Lisa (Elisabeth) N. Roberts


Copyright © Lisa (Elisabeth) N. Roberts 2016

A Dissertation Submitted to the Faculty of the


In Partial Fulfillment of the Requirements

For the Degree of



In the Graduate College






As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Lisa (Elisabeth) N. Roberts, titled “I have a connection!”: The Situated Sense-making of an

Elementary Student about the Role of Water in Modeled vs. Experienced Ecosystems, and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of


______________________________________________________Date: (November 18, 2015)

Walter Doyle

______________________________________________________Date: (November 18, 2015)

Kristin Gunckel

______________________________________________________Date: (November 18, 2015)

Marcy Wood

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

________________________________________________ Date: (November 18, 2015)

Dissertation Director: Bruce Johnson



This dissertation has been submitted in partial fulfillment of the requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that an accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Lisa (Elisabeth) N. Roberts



It has been a long, strange trip, yet one made infinitely more meaningful, fun, and tolerable through the countless small and large acts of help and encouragement from more people than I have room or memory to cite.

This journey started many years ago in the English classroom of joan mountford at Alvirne

High School in Hudson, New Hampshire. Joan was my first muse and task master in the art of inquiry and writing; any good work in this thesis is her legacy; the bad is completely my responsibility.

The second kick in the pants to start this meandering path came from Susan Doubler, who asked “why not a Ph.D.?” in the middle of a late autumn walk in the New England woods. Thank you, Sue, for believing it possible. M. Jean Young, Katherine Paget and Susan Loucks Horsley added momentum to that initial kick. Thank you Jean for your constant noodging and for teaching me the art of methodological bullshit detecting when reading and reviewing studies. Thank you Kathy, for your work with Sue in guiding me through the transition from evaluator to researcher, and for navigating the twists and turns of many projects that Shall Not Be Named. Thank you Susan, wherever you now are, for your vision of rigorous and high-quality science learning for elementary students.

My advisor, Bruce Johnson, aka Miracle Max, has guided and stuck with me on this journey through multiple twists and turns in this process, encouraging me with patience, insistence, rigor, and the dreaded raised eyebrow. Each member of my committee, Walter Doyle, Kristin Gunckel, and Marcy Wood, pushed my thinking far beyond any ledge I was comfortable walking, and in so doing taught me the art of research. Other faculty at the University of Arizona who guided me through the thickets include Patricia Anders, Alberto Arenas, Iliana Reyes, and the deeply missed

Richard Ruiz. A special thank you to Kathy Short for introducing me to the practices and perils of teacher research, the particular rigor of qualitative research, and the ability to figure out “what goes with what.”

I am grateful to all fellow teachers and colleagues I had the honor of apprenticing with at

TERC, in the NSF Systemic Change Initiatives, and my current projects. You each were inspiring examples of the highest values of teaching, and agents of change whose work is still active in this world despite the surface turbulence of educational policy and politics. A particular shout-out to the

5 teachers, students, volunteers, and extended community of Manzo Elementary School, who are literally laying the groundwork for children to develop ecological literacy through building a resilient and healthy community using school gardening and permaculture practices.

This work would not have been possible without the support of Dr. Lisa Elfring and the incredible group of teachers I had the honor of working with through the Systems in Middle School science project, one of whom became the teacher in this study. Thank you for trusting me with your students, Ellie. Thank you, students from Room 5. Your ideas were so amazing I could have written

28 dissertations.

Saving the most personal for last, I am deeply grateful to all of my family, friends and fellow graduate students who stood by, picked me up, fed me, gave me love and support, taught me meditation, and (mostly) avoided asking if I was done yet. Special thanks to Drr2, Drr3, and Drr4 for your models of perseverance, grace under pressure, and strange sense of humor at just the right moments. Thank you, Janet MacGregor, for helping me re-light the flame. My deep appreciation to the staff and counter mates at Frank’s diner, who eschewed elegant dining for unconditional love and cheap breakfasts.

And most of all, to the Ph.D Cupcakes: Erin Dokter, Sanlyn Buxner, Jamie Carson, Audra

Baleisis, and especially Jessie Antonellis-John, who had the enormous task of editing this beast while keeping me laughing and (mostly) sane through the end.


This dissertation is dedicated to the memory of three women who helped me learn to love this world and how it works:

My mother, Elaine Poehler Roberts

My sister, Cynthia Roberts Sullivan

My sister-from-another-mister, Patricia Roberts

May this work keep your light and the green fire alive




ABSTRACT ....................................................................................................................................................... 15

CHAPTER 1: PROBLEM AND FRAMEWORK .................................................................................... 17

Introduction ...................................................................................................................................................... 17

Rationale ............................................................................................................................................................. 18

The Need for Ecological Literacy ........................................................................................................... 18

Challenges to Understanding Ecosystems ............................................................................................. 19

Ecosystems in Science Education ........................................................................................................... 19

The Problem ...................................................................................................................................................... 20

Research on Ecological Understanding ................................................................................................. 21

Developing Ecological Knowledge through Modeling Practices ...................................................... 23

The power of microcosms .......................................................................................................................... 23

Extending physical models with inscriptional practices .............................................................................. 27

Challenges of translating design-based research studies for everyday classrooms ........................................... 28

Conceptual Framework ................................................................................................................................... 30

Situativity as Interaction of Learners with Complex Instructional Systems ..................................... 32

Situations and Problems Emerge in Activity ......................................................................................... 33

Agency and Activity Shape the Emerging Situation ............................................................................. 34

Affordances and Constraints Shape Interaction and Agency ............................................................. 35

Affordances and Constraints of Physical and Cultural Tools Mediate Sense Making .................... 36

Models as Tools for Sense Making ......................................................................................................... 38

Grounding Situativity in Classrooms with Academic Task Theory ................................................... 40

Conceptual Framework Summary ........................................................................................................... 41

Research Questions .......................................................................................................................................... 42

Overview of the Dissertation ......................................................................................................................... 43

CHAPTER 2: LITERATURE REVIEW .................................................................................................... 45


Ecology and Systems Thinking in Western Science .................................................................................... 46

Ecology ........................................................................................................................................................ 46

Systems ........................................................................................................................................................ 48

Definitions and delimitations: Ecosystems, Environment, and Nature ............................................ 50

Review of Empirical Literature ...................................................................................................................... 51

Children’s Conceptions and “Misconceptions” .................................................................................... 51

Materials cycling ...................................................................................................................................... 52

Energy in ecosystems ................................................................................................................................ 56

Limitations of misconceptions studies ........................................................................................................ 59

Development of Biological and Causal Reasoning ............................................................................... 62

Naïve biology ........................................................................................................................................... 64

Causal reasoning ..................................................................................................................................... 66

Limitations of Developmental Studies ...................................................................................................... 68

Phenomenographic Experiencing of Ecosystems ................................................................................ 69

Ecological insights .................................................................................................................................... 70

The insight of transformation ................................................................................................................... 71

The Water Cycle as Perceivable Model for Non-Perceptual Materials Cycles ................................. 72

Idiosyncratic Trajectories of Ecological Insight ................................................................................... 73

Experiencing Ecosystems from “The Bottom Up” ............................................................................. 73

Limitations of phenomenological studies .................................................................................................... 75

Teaching and Learning about Ecosystems: Structures, Functions, Behaviors, and Models ................ 76

The Practice of Modeling ......................................................................................................................... 77

Inscriptional models ................................................................................................................................. 79

Models as a form of explanation .............................................................................................................. 82

Scaffolding modeling practices ................................................................................................................... 82

Conclusion ......................................................................................................................................................... 85

The Power of Microcosms ....................................................................................................................... 85


Extending Physical Models with Inscriptional Practices ..................................................................... 88

Challenges of Translating Design-based Research for Everyday Classrooms ................................. 89

CHAPTER 3: METHODS ............................................................................................................................. 92

Study Design ..................................................................................................................................................... 92

Unit of Analysis .......................................................................................................................................... 93

Context ........................................................................................................................................................ 95

Site .......................................................................................................................................................... 95

Negotiating site access .............................................................................................................................. 96

Teacher .................................................................................................................................................... 97

Focal student ........................................................................................................................................... 97

Researcher Role ........................................................................................................................................ 100

The BioBottle Project Design ................................................................................................................ 101

Data Collection and Management ................................................................................................................ 102

Management, logistics, and timelines .................................................................................................... 102

The plan vs. the reality ........................................................................................................................... 102

Classroom Observation, Videotaping, and Field notes ..................................................................... 103

Informal Conversations with the Teacher ........................................................................................... 104

Other Data Sources ................................................................................................................................. 104

Data Management .................................................................................................................................... 105

Digital record transcriptions ................................................................................................................... 105

Interview ................................................................................................................................................ 107

Academic task artifacts ......................................................................................................................... 107

Analysis ............................................................................................................................................................ 108

Analysis Level 1: Phenomenographic Semiotic Analysis (Emic Perspective) ................................ 108

Example of the semiotic analysis process ................................................................................................. 109

Analysis level 2: Jorge’s Expressions in Relation to the Content (Etic Perspective) .................... 113

Example of the coding process ................................................................................................................ 114


Analysis Level 3: Role of Water in Ecosystems .................................................................................. 114

Methodological Trustworthiness, Credibility, and Limitations ............................................................... 116

Conclusion ....................................................................................................................................................... 116

CHAPTER 4: FINDINGS ........................................................................................................................... 117

Overview .......................................................................................................................................................... 117

Academic Tasks .............................................................................................................................................. 117

Task 1: Pre-Assessment .......................................................................................................................... 119

Task and context .................................................................................................................................. 119

Resources ............................................................................................................................................... 120

Interactions of child and task ................................................................................................................. 120

Role of water .......................................................................................................................................... 123

Practices of modeling .............................................................................................................................. 123

Task 2: Decomposition Bottles ............................................................................................................. 124

Task and context .................................................................................................................................. 124

Interaction of child and task ................................................................................................................... 128

Role of water .......................................................................................................................................... 130

Modeling practices .................................................................................................................................. 130

Task 3: How does sun energy affect water in a bottle model? ......................................................... 131

Task and context .................................................................................................................................. 131

Interactions of child and task ................................................................................................................. 132

Role of water .......................................................................................................................................... 132

Practices of modeling .............................................................................................................................. 135

Task 4: How Does Water Interact with Different Soils? .................................................................. 136

Task and context .................................................................................................................................. 136

Interactions of child and task ................................................................................................................. 137

Role of water .......................................................................................................................................... 141

Practices of modeling .............................................................................................................................. 142


Task 5: The Connection Circle .............................................................................................................. 142

Task and context .................................................................................................................................. 142

Interaction of child and task ................................................................................................................... 144

Role of water .......................................................................................................................................... 148

Practices of modeling .............................................................................................................................. 148

Task 6: Designing and Observing a Sealed Ecosystem Model: The BioBottle .............................. 149

Task and context .................................................................................................................................. 149

Interactions of child and task ................................................................................................................. 150

Role of water .......................................................................................................................................... 153

Practices of modeling .............................................................................................................................. 154

Task 7: Observing the BioBottles ......................................................................................................... 156

Task and context .................................................................................................................................. 156

Interactions of child and task ................................................................................................................. 156

Role of water .......................................................................................................................................... 158

Practices of modeling .............................................................................................................................. 160

Interlude 1: Water Cycle Lesson ........................................................................................................... 160

Task 8: Scientific Posters ........................................................................................................................ 164

Task and context .................................................................................................................................. 164

Interactions of child and task ................................................................................................................. 164

Role of water .......................................................................................................................................... 168

Practices of modeling .............................................................................................................................. 169

Interlude 2: Setting Free the Worms .................................................................................................... 170

Task 9: Written Post-Assessment .......................................................................................................... 171

Task and context .................................................................................................................................. 171

Interactions of child and task ................................................................................................................. 171

Role of water .......................................................................................................................................... 175

Practices of modeling .............................................................................................................................. 176

Summary Across Tasks ........................................................................................................................... 177


Water in Experience, Water in a Model System ........................................................................................ 178

Water in Experience ................................................................................................................................ 178

“The roots suck up the water” ................................................................................................................ 178

"This is a moister area plant" ................................................................................................................ 179

“My cousin said, ‘looook, it's a rock.’ I said ‘no, it's my mango.’” .......................................................... 183

Water in a Model System ........................................................................................................................ 186

"I'm guessing like two drops of water…” ............................................................................................... 186

Conclusion ....................................................................................................................................................... 192

CHAPTER 5: DISCUSSION ...................................................................................................................... 195

Research Findings ........................................................................................................................................... 195

Interactions of Learner, Academic Task, and Modeling Practices ................................................... 196

Role of the Microcosm ........................................................................................................................... 199

Enactment of Modeling Practices ......................................................................................................... 202

The Role of Water in Ecosystems: Ideas In and Out of Focus ....................................................... 205

Broader Implications ...................................................................................................................................... 207

Teaching, Curriculum, and Teacher Education .................................................................................. 207

Assessment and Research ....................................................................................................................... 210

Recommendations for Future Research ..................................................................................................... 212

Reflections on the Research and Limitations of the Study ...................................................................... 212

Conclusion ....................................................................................................................................................... 213

REFERENCES .............................................................................................................................................. 215



Figure 1: Timeline with major tasks of the Bottle Biology unit by week ............................................... 103

Figure 2: Jorge’s pre-assessment work product ........................................................................................... 111

Figure 3: Academic tasks across the study ................................................................................................... 118

Figure 4: Jorge’s references to ecosystem structure-function across academic tasks ............................ 119

Figure 6: Compost bottle models at 24 hours ............................................................................................. 125

Figure7: Bottle 3 showing mold and fungal growths ................................................................................. 126

Figure 8: Jorge’s notebook entry in response to “What is a model?” ...................................................... 127

Figure 9: Jorge and his table partner explore the contents of the decomposition model .................... 130

Figure 10. Jorge’s assessment work product: “What is the importance of water in an ecosystem?” .. 134

Figure 11: Jorge explains how soil with bigger pieces lets the water drain through ............................ 139

Figure 12: Jorge’s notes on results of the water-soil interaction ............................................................... 140

Figure 13: The teacher points at the whole-class-constructed Connection Circle ................................. 144

Figure 14: Jorge’s finished Connection Circle work product .................................................................... 145

Figure 16: Jorge and Rosa’s written plan for the design of their BioBottle model ................................ 151

Figure 17: Jorge and Rosa’s illustrated plan for the design of their BioBottle model ........................... 151

Figure 18: Example of a data sheet from Jorge’s observation of the BioBottle Model on Day 4 ...... 157

Figure 19: Jorge’s first BioBottle observation drawing .............................................................................. 158

Figure 20: Summary of Jorge and Rosa’s BioBottle observations ............................................................ 159

Figure 21: Teacher Illustration of the Water Cycle during her lecture .................................................... 161

Figure 22: Jorge and Rosa’s Bottle Biology Poster ..................................................................................... 166

Figure 23: Jorge and Rosa’s methods section of their poster .................................................................... 168

Figure 24: Jorge post-assessment mind-map ............................................................................................... 172

Figure 25: Jorge’s post-assessment representation of water in the BioBottle ........................................ 173


Figure 26: Jorge’s Post-Assessment diagram of the life cycle of a plant ................................................. 174

Figure 27: Jorge’s Post-Assessment Representation of the Water Cycle ................................................ 175

Figure 28: Jorge and Rosa’s design of the closed system BioBottle ......................................................... 187

Figure 29: Jorge’s explanation of how water is conserved in the closed BioBottle ............................... 191



Current policy and research have led the field of science education towards a model of “science as practice.” In the past decade, several research programs on model-based reasoning practices in education have articulated key dimensions of practice, including constructing and defending models, comparing models to empirical data, using representations to identify patterns in data and use those as inscriptions to buttress arguments.

This study presents a detailed case of how the use of a physical microcosm and children’s selfdirected representations of an ecosystem constrained and afforded student sense-making in an urban elementary classroom. The case analyzed the experiences of a 10-year old fifth grade student, Jorge, and the variation in his expressed understanding of ecosystems as he interacted with academic tasks, along with models and representations, to design, observe and explain an ecological microcosm.

The study used a conceptual framework that brings together theories of situated cognitionm phenomenology, and academic task to explain how and why Jorge’s perception and communication of dimensions of ecosystem structure, function, and behavior appear to “come in and out of focus,” influenced by the affordances of the tools and resources available, the academic task as given by the teacher, and Jorge’s own experiences and knowledge of phenomena related to ecosystems.

Findings from this study suggest that elementary students’ ability or inability to address particular ecological concepts in a given task relate less to gaps in their understanding and more to the structure of academic tasks and learning contexts. The process of a student interacting with curriculum follows a dynamic trajectory and leads to emergent outcomes. As a result of the complex interactions of task, tools, and his own interests and agency, Jorge’s attunement to the role of water in ecosystems comes in and out of focus throughout the unit. The instructional constraint of needing to integrate the FOSS Water Cycle curriculum into the Bottle Biology Project became an

affordance for Jorge to ask questions, observe, and theorize about the role of water and the water cycle in an ecosystem. The practice of modeling a closed ecosystem made salient to Jorge the boundaries of a system and the conservation of water within that system. The closed ecosystem model also presented constraints to students’ sense making about the role of interactions when students lack domain knowledge in ecology. Relying on students’ own talk, photographs and representations as explanations of phenomena in the Bio Bottle, without establishing norms of representational conventions and communication, resulted in missed opportunities for Jorge to reinforce his sense making during the activity and to develop conventions of scientific representation.

Findings from this study can be used to inform the design and implementation of learning environments and curricular activities for elementary and middle school students that address all three dimensions of the Next Generation Science Standards: a) developing conceptual understanding of key concepts in the domain of ecology, b) the cross-cutting concept of systems, and c) multiple practices that ecologists use in developing and evaluating models that explain ecosystem structures, functions, and change over time.




Ecology, like the whole of science, has been a house with many doors, some leading to one view of nature, some to another (Worsted, 1977, p. 348).

[K]nowledge, instead of being pursued in depth and integrated in breadth, is pursued in depth in isolation. Rather than getting a continuous and coherent picture, we are getting fragments—remarkably detailed but isolated patterns. We are drilling holes in the wall of mystery that we call nature and reality on many locations....But it is only now that we are beginning to realize the need for connecting the probes with one another and gaining some coherent insight into what is there" (Laszlo, 1996, pp.



This qualitative case study investigated how the ecological sense-making of an 10-year-old fifth grader, Jorge, was influenced by the curricular tasks and learning contexts as he engaged in the practices of designing, observing, and explaining the changes to a model ecosystem, the BioBottle.

The situation and setting of this study provided a unique opportunity to explore student’s sense making of ecology through the modeling experience after engaging in a year of school science curriculum, an informal outdoor learning program, and activities related to the teacher's emerging expertise with integrating systems-thinking tools and habits of mind (Richmond, 1993) throughout her teaching. My work with this teacher through a systems thinking professional development grant, as well as research connected to the informal education programs, led me to the questions investigated in this study.

This chapter identifies how the study addresses a critical problem in both science education and society: the need for all people to develop what Orr (1992) calls “ecological literacy.” Following

18 this, I present the study’s conceptual framework, which is a synthesis of situated learning theory, academic task, and scientific practices, particularly modeling and representations. The chapter ends with the research questions.


The Need for Ecological Literacy

Ecology is the study of the interrelationship of biotic and abiotic systems and how those interrelationships sustain the order and organization of earth’s biosphere and provide resources necessary for human life. Science and environmental educators face the challenge of developing ecological literacy in a world where most children think that milk comes from the grocery store

(Louv, 2005) and fewer than 20 percent of Americans have attained a level of scientific literacy to understand a science article in a major newspaper (Bauer, Allum & Miller, 2007).

Ecological literacy—people’s ability to understand and manage their role in complex ecosystems—is central to human quality of life and our future as a species in a limited, fragile earth system (Bowers, 1996; Capra, 1996; Orr, 1992). As Orr defined it, ecological literacy is the

“demanding capacity to observe nature with insight, a merger of landscape and mindscape” (1992, p.

86). The Ecological Society of America provides a more scientifically focused definition: “[a]n ecologically literate person exhibits awareness about local habitats, can link local issues to global concerns, and has an understanding of spatially independent concepts and issues” (Jordan, Singer,

Vaugn & Berkowitz, 2009). Knowledge about how ecosystems work is a necessary but not sufficient foundation for the insight of ecological literacy (Orr, 1992; Van Matre, 1990). We need to understand that an ecosystem, like all systems, is a whole; a change to any function or interaction perturbs the entire system (Carlsson, 1999; Odum, 1968). We also need to understand that humans are an integral part of multitudes of systems, both affecting and affected by the accumulation of our actions and relationships with other organisms and with the material resources of Earth. Humans

19 and all living beings inhabit and depend on multiple ecosystem services, both local and global, to support their lives. Human demand for resources have resulted in change, damage, and destruction to ecosystems in ways that may be too subtle or too distant for most of us to perceive at the “life world” level of scale (Thomashow, 2002).

Challenges to Understanding Ecosystems

Ecological systems and their processes are multidimensional, dynamic, complex, and emergent. Ecosystems result from the interactions of abiotic Earth systems, including the lithosphere, atmosphere, hydrosphere, and climate, with living organisms that are self-organized

(Priogine & Stengers, 1984; von Bertalanffy, 1968) to maximize the available energy and materials available in a specific location. Challenges inherent in making sense of ecological systems—their structures, functions, interactions (causal mechanisms), and change over time—are related to this dynamic complexity (Capra, 1996; D’Avanso, 2008). A key challenge for understanding and instruction is how and when to think beyond our human scale of daily life, or meso-scale, to consider the macro-level structures and functions of ecosystems such as energy flow, materials cycles, populations, and climate (Carlsson, 1999). A reciprocal challenge is to understand how ecological systems operate at the micro-level of scale—e.g., not only the biological functions, interactions and transformations such as photosynthesis and respiration, but also the biogeochemical interactions of the biotic and abiotic systems such as the water cycle, carbon cycle, nutrient cycle, and energy flow. A third challenge for humans is to make sense of ways that multiple sub-systems and causal structures, including feedback, contribute to the emergent behavior of ecosystems such as thermodynamic equilibrium, ecological succession, and flux in biodiversity (Booth Sweeney, 2004;

Grotzer, 1993).

Ecosystems in Science Education

The recently released Framework for K-12 Science Education (NRC, 2012) and accompanying


Next Generation Science Standards (NGSS Lead States, 2013) present a new policy vision for science education that integrates three dimensions: Practices, Cross-Cutting Concepts, and Disciplinary Core

Ideas. The authors of this framework propose a solution to the "mile wide, inch-deep" science standards from the 1990s, which have arguably contributed to the piecemeal and disconnected opportunities for students to develop ecological understanding. The framework positions

Ecosystems, Interactions, Energy and Dynamics as one of four Core Ideas in the life sciences, and specifies seven Cross-Cutting concepts (NRC, 2012, pp. 84-100). These cross cutting concepts—1)

Patterns, 2) Cause and effect; 3) Scale; 4) Systems and systems models; 5) Energy and matter: Flows, cycles and conservation; 6) Structure and Function; 7) Stability and Change—are all integral to understanding ecosystems. While all eight Essential Practices identified in the Framework (p. 41-82) are necessary for the construction of scientific knowledge, three—developing and using models, engaging in analyzing and interpreting data, and constructing explanations—represent central practices in the field of systems ecology and in the project of this study (Canham, Cole & Lauenroth,

2003; Capra, 1996; Orr, 1992, Pickett, Kolasa & Jones, 2007; Worster, 1977).

The Problem

Public education can and must play a role in developing students’ ecological literacy, starting in the elementary years. To achieve the outcomes envisioned by the new policy requires detailed understanding of children's sense-making, in both disciplinary content and cross-cutting domains.

While a century ago most of the world’s people understood ecosystems at the life world level through living in rural areas and the practices of farming, food gathering, and homesteading, significant changes in our society over the past 50 years have caused most children in Western culture to grow up disconnected from the natural world (Louv, 2005). In the past ten years, a “green revolution” movement across our culture and within schools has begun to address the need for literally re-grounding children in ecological experiences. In parallel with school science, the “green

21 revolution” has catalyzed a renewed interest in creating hybrid formal/informal educational experiences such as earth education (Van Matre, 1979,1990; Van Matre & Johnson, 1988), school lunch gardens (Center for Ecoliteracy, 2006), and urban gardening (Fusco, 2001). School curricula and informal programs, while not a panacea, nevertheless provide an important site for practicebased ecological learning.

While these programs are promising, there is much to be understood about how children participate in these systems, and as a result, whether and how such experiences serve to increase children’s ecological literacy. It remains up to children (with the help of more knowledgeable others) to weave a coherent personal understanding of ecology and experiences with multiple ecosystems spread across different times, spaces, and contexts.

Research on Ecological Understanding

The research literature in science education has until recently painted a discouraging picture of children’s and adults’ lack of ecological knowledge and systemic reasoning. For example, expert models such as food chains and food webs are ubiquitous in K-12 curriculum texts and textbooks

(e.g., BSCS, 1999), yet research studies over decades and across cultures suggest that most children

(and adults) interpret these not as representing the flow of energy across trophic levels but canonical representations of “what eats what” (Adeniyi, 1985: Barman, Griffiths & Okebukola, 1995; Barman

& Mayer, 1994; Boyes & Stannistreet, 1991; Gallegos, Jerzano & Flores, 1994; Gotwals & Songer,

2010; Griffiths & Grant, 1985; Hogan, 2000; Leach, Driver, Scott & Wood Robinson, 1996; Lin &

Hu, 2003; Munson, 1994; Palmer, 1997; Roberts & Johnson, 2005, 2006; Webb & Boltt, 1990, 1991).

Further, these studies suggest few people of any age can accurately predict or explain what might happen to a food web if one population is disturbed. A likely explanation for the well-documented difficulties students show when attempting to explain food webs or the result of perturbations is that they have memorized the components or visible structures without understanding the

22 underlying organizing principle of transformation of energy and materials (Carlsson, 1999). Students also lack domain-specific knowledge of the types or needs of organisms in an ecosystem (Magntorn

& Helldén 2005, 2007), which may lead them to reason by assuming all organisms can choose what they eat (as humans do), or in over-simplified food web representations, that the animal (or population) who loses a food source will simply die.

A parallel body of research demonstrates difficulties students and young adults demonstrate when trying to explaining the cycling and transformation of materials in ecosystems and earth systems, such as decomposition (Andersson, 1991; Eskilsson & Helldén, 2003; Helldén, 1995;

Hogan & Fisherkeller, 1996; Lofgren & Helldén, 2008; Smith & Anderson, 1986), exchange of gases and cycling of materials through an ecosystem (Wilson, Anderson, Heidemann, Merrill, Merritt,

Richmond, Sibley, & Parker, 2006) biogeochemical cycles such as carbon exchange (Mohan, Chen &

Anderson, 2009) and the water cycle (Covitt, Gunckel & Anderson, 2009; Gunckel, Covitt, Salinas &

Anderson, 2012). A common naïve form of reasoning shown across these studies is that participants describe the visible components of the system, which again may be limited by their knowledge of the system.

People’s inability to apply ecological or systems principles may be based on lack of domain knowledge, such as the adaptability of feeding patterns of species at different trophic levels (e.g.,

Edgerton, 2007; Leach et al, 1996b), overly-linear and limited causal or systems reasoning (Grotzer,

1993, 2002, 2003; Grotzer & Bell Basca, 2003; Grotzer & Solis, 2015; Hogan, 2000; Perkins &

Grotzer, 2005), or lack of experience using models of dynamic systems (Hogan, 2000; Hogan &

Weathers, 2003). Given the complexity of both ecology and systems as domains, the challenges to learners are likely a combination of all three (D’Avonso, 2008; Jordan, Singer, Vaughan & Berkowitz



Developing Ecological Knowledge through Modeling Practices

A critique of the earlier “misconceptions” research is that most assess students’ reasoning about ecosystem form and function using reductionist, verbal-response items focused on individual organisms, static concepts rather than processes, and do not determine students’ domain-based knowledge of the particular ecosystems used in items (Carlsson, 1999; Leach et al., 2006b;

Magntorn, 2005; Tunnicliffe & Ueckert, 2007). A more recent body of research suggests that given enough time to develop expertise in a content area, and instructional scaffolds that support extended inquiry and sense making, children in elementary grades are able to engage in modeling practices to make sense of biological and ecological processes (Ero-Tolliver, Lucas & Schauble, 2013; Lehrer &

Schauble, 2004; Lehrer, Carpenter, Schauble & Putz, 2000; Lehrer, Schauble & Lucas, 2008; Manz,

2012). In the following section I discuss three dimensions examined in more recent studies, elaborating on how the present study builds on and extends existing research: 1) the power of microcosms as model systems; 2) extending the limits of physical models with representations and inscriptional practices, and 3) translating large-scale proof-of-concept design studies to implementation in “everyday” classroom contexts.

The power of microcosms

Teaching ecosystems through the use of physical models (microcosms) is one approach that has shown promise from the college level to first grade. Tamponnet and Savage (1994) argued that developing and studying closed ecosystems (containing associations of organisms that form materials cycles, but open to energy exchange from the outside) can serve as tools to determine the combinations necessary to attain the long-term succession and balance needed to support long-term space exploration, or to learn further about the functioning of terrestrial systems (p. 167). The

Biosphere2 Center at the University of Arizona was originally designed as such a system, which led to important new scientific understandings of how large-scale microcosms function (e.g., Tubiello,


Lin, Druitt, Marino, 1999) as well as how such model systems alter the function and biodiversity of wild biomes (e.g., Leigh, Burgess, Marino & Wei, 1999).

Two studies illustrate the potential of using microcosms in research and instruction. Britta

Carlsson investigated Swedish student teachers’ qualitative variation of ideas about the structures and functions of an ecosystem through two interview-facilitated tasks of designing ecosystem models (1999; 2002a). Her approach was to question the teachers in ways that reveal their discernment of the whole, and the interrelationship of the components by designing two microcosms: one in a glass cylinder, and one a thought experiment about what would be needed to supply a spaceship to support human exploration over 6000 years. She suggested that the variation in categories of descriptions she identified might offer teachers tools for understanding students’ pre-instructional thinking, which could lead to pedagogical interventions. Carlsson found that of her

10 participants, five (50%) viewed photosynthesis (energy fixation) as a “black box” process in which plants take in some materials, while producing others. Most of the teachers (70%) expressed ideas that materials are transformed as they recycle. The key conceptual shift for understanding how ecosystems develop and function, Carlsson argued, requires an understanding of transformation as key to explaining both energy flow and materials cycling (2002a,b). Carlsson’s use of a task of designing a microcosm, the materially accessible Eco bowl, served to elicit participants’ ideas and questions, to which she could respond with further prompts or clarifications. Carlsson’s study suggested the Eco Bowl design task could be a fruitful strategy for research with students, encouraging them to express their mental models and test ideas through the design of a physical model.

Helldén (1995) followed 26 Swedish students’ ideas about the transformation of matter in ecosystems from age 8 to 15 using a combination of interviews and a closed microcosm in which plants were grown and leaf litter changes studied. Helldén emphasized that water plays a key role in

25 biological and ecological processes, both in the fixation of carbon through photosynthesis, and in the processes of respiration and decomposition, which release water, oxygen, and other key molecules for reuse. The water cycle and phase changes of water play a key role at the level of the ecosystem. While molecules in gaseous form are not perceivable by an “ordinary” observer, the processes of decomposition and rot are discernable. To challenge students’ naïve explanations that dead materials simply “disappear,” or that plants increase mass (grow) due to continuous uptake of water from the soil, Helldén cultivated plants in sealed transparent boxes to focus students on the conditions for growth and to provide visible limits for decomposition and materials cycles. Students noted the transformation of water in the sealed system, which Helldén claimed became a prototype for thinking about “cycles” that they incorporated to explain both decomposition and gas exchange by age 15.

The methods and findings from these studies suggest that engaging students in modeling microcosms challenges students to consider what are the necessary structures and proportions of biotic and abiotic materials needed to create a functioning ecosystem. Designing microcosms with feedback from peers challenged participants to make their thinking visible to themselves and others—to express their tacit and personal mental models so that others could evaluate and give feedback on their ideas. Carlsson’s (1999) approach, allowing her participants to decide what should go into the Eco Bowl to constitute a functional ecosystem, became the basis of this study’s design task. Helldén’s (1995) success in using his closed microcosm to focus students on water’s phase changes and the visible processes of materials and decomposition cycles also suggests that such models could be used to challenge students’ naïve explanations of materials “disappearing” and to focus their perception on the theoretical need to identify boundaries in the analysis of systems by instantiating these as physical boundaries in the model system.

A limitation in both these studies is that neither Carlsson nor Helldén made explicit the ways

26 that their microcosms functioned as simplified models of the Earth’s materially-closed but energetically-open system. To emphasize this epistemological connection, in the current study, the teacher and I made explicit the function of the BioBottle as a model of Earth’s system, using the intermediate model of Biosphere 2 as an exemplar and a story frame.

Lehrer's and Schauble’s extensive body of design-based research studies suggests modeling practice at the elementary level is most successful when instruction begins with physical models that maintain resemblance to the phenomenon under investigation (Lehrer & Schauble, 2005). Their approach builds on children’s limited understanding of models as physical copies of the world (e.g.,

Grosslight, Unger, Jay & Smith, 1991). Shifting the instructional emphasis from structures

(components and relationships) to functions (causal relations and the resulting changes over time) has been shown to elicit more detailed explanations in related studies of children designing a functioning elbow (Penner, Giles, Lehrer and Schauble, 1997), exploring the biomechanical functions of levers (Penner, Lehrer, Schauble, & Putz, 1998), reasoning about gears (Lehrer &

Schauble, 1998), explaining natural variation in populations and exploring growth patterns in organisms (Lehrer & Schauble, 2004), designing artificial lungs (Hmelo, Holton & Kolodner, 2000), and aquarium systems (Hmelo-Silver, Marathe, & Liu, 2007; Lehrer, Schauble, & Lucas, 2008). The model-design task used in the current study extends the focus of modeling studies into the domain of ecology by challenging students to design a materially-closed model that functioned like a garden ecosystem. Two other modeling studies related to ecosystems focused on discrete processes such as decomposition with first-grade students (Ero-Tolliver, Lucas & Schauble, 2013) and reproductive success with third graders (Manz, 2012). These are important precursor studies for developing students’ attunement to key ecological ideas such as materials cycles and building toward more complex ecological understandings such as evolution (Lehrer & Schauble, 2012), but leave open the question of how and at what ages students are capable of considering an ecosystem as a functional

27 and complex whole.

The limitations of students’ domain-based knowledge, which were shown to have an effect in some studies (e.g., Hmelo-Silver, Marathe, & Liu; Hmelo-Silver & Pfeffer, 2003) were minimized in this study by introducing the project at the end of the school year after students had completed two related science kits, Ecosystems (BSCS, 1999) and Earth Materials (FOSS, 2005), and were concurrently engaged in a unit on Water (FOSS, 2003). Given that the teacher had approximately seven years’ experience in systems dynamics thinking (which includes both inscriptional and computer-based modeling practices), and that the student in the study had worked with this teacher for three years, this unique case provided an opportunity to contribute an additional perspective in the research on modeling practices and ecological thinking of elementary students.

Extending physical models with inscriptional practices

Deciding which aspects of the natural world to include or highlight in their model, and which to ignore, is a key metacognitive process in modeling (Schwarz & White, 2005) and inscriptions (diSessa, 2004) that develops through production, evaluation, and revision (Lehrer &

Schauble, 2002; Lehrer, Schauble, Carpenter & Penner, 2000). For example, Lehrer, Schauble,

Carpenter and Penner (2000) described engaging third-graders in an iterative series of investigations in which students grew populations of Wisconsin Fast Plants™ and were challenged by the researchers to use graphical representations of variations in plant height to decide what was a

“typical” height. Researchers also had students graph change in height through the plants’ life cycle, which with careful questioning by researchers and teachers led students to identify the characteristic

“S” shape growth curve. This in turn led to questions and investigations related to why plants and humans demonstrated both a growth spurt and a leveling off of growth at maturity.

While Lehrer and Schauble advocate for students to develop and evaluate their own forms of model-based representations, the case described in Lehrer, Schauble, Carpenter and Penner


(2000) suggests that the inscriptional and modeling decisions were primarily introduced and orchestrated by the researchers with specific mathematical goals in mind. They acknowledged that the work of constructing inscriptions to reveal underlying patterns and relationships in nature is itself a complex practice that needs careful scaffolding by teachers who clearly know which patterns and processes they expect students to make sense of. One aim of this current study was to explore what kinds of inscriptions students in this urban classroom, working with a teacher who herself was developing systems dynamics thinking, generated to express and extend their ideas about ecosystem’s structures and functions. The current study analyzed the inscriptional practices of one student across various tasks in which he expressed his ideas about ecosystem structures and functions using a variety of representations (drawings, diagrams, concept maps, and related visual models). Those related to the design and observation of the physical microcosm offer evidence for which aspects of the system are perceived and considered salient to the student, albeit in a complex interaction of his ecological knowledge, emergent sense-making, and representational capabilities.

Challenges of translating design-based research studies for everyday classrooms

As Lehrer and Schauble (2006) acknowledged, there are many unique features inherent in conducting design research that are not typically found in most educational settings. Such studies require interdisciplinary teams with a high degree of communication and collaboration. Replication of an intervention over multiple grade levels and classrooms takes years, money, graduate students, and institutional support. For example, the plant diversity and growth modeling studies of Lehrer &

Schauble (2003, 2004) entailed working with an entire class of 23 students, with data collected by up to three researchers on whole-class and small-group interactions, as well as follow-up individual interviews audiotaped and transcribed. They noted that appropriate contexts for such design-based studies included a need for a large number of measures of multiple attributes, time and opportunity for exploring leads or revisiting problematic areas, were detailed enough to track change in learning

29 over time, and entailed inquiry practices that could be conducted within the limits of school time and materials (2004, p. 670). While the school itself could be considered “typical” in terms of student diversity, teacher experience, and financing, the school and several teachers also had an extended history of collaboration with university researchers, and leadership commitment to the professional development and autonomy of teachers as well as to promoting innovation. Further, the researchers also designed an extended series of professional development experiences for teachers over the course of several summers and during the school year (Lehrer & Schauble, 2000).

With the introduction of “No Child Left Behind” and the increased emphasis on high-stakes testing of mathematical computation and linguistic fluency, such extended design-studies became more difficult to implement (Lehrer & Schauble, 2006). At present, most elementary teachers and students, particularly in underserved urban classrooms, do not have access to the conditions that supported Lehrer and Schauble’s earlier design studies, but that should not be accepted as an excuse not to search for more scaled-down or different approaches that reflect the underlying principles of successful design-based modeling research to date. To that end, this study attempted to translate the principles of their large-scale design research by exploring how the use of ecological microcosms and other forms of modeling in a more “typical” classroom might contribute to students’ ecological literacy.

A related characteristic of most of the studies cited above relates to the unit of analysis—the bounds of the studies examine impact of instruction at the level of the classroom. While individual group vignettes highlight different representational choices (Lehrer & Schauble, 2004) or analyze group abilities to manage the complexities of inquiry related to an aquatic ecosystem (Lehrer,

Schauble & Lucas, 2008), data on the impact of instruction or the nature of student modeling practice were analyzed at the classroom level. This level of analysis makes a compelling case for the educational value and efficacy of approaching science instruction through the practices of modeling.


Yet at the level of day-to-day teaching and assessment practice, the focus of performance outcomes remains on the individual student, and the challenge to teachers is how to identify and build upon individual students’ ideas and practices—to make informed instructional decisions that take into account individual learners’ potentials (as well as managing the ideas of thirty-odd students in the typical urban elementary classroom). In short, design-studies suggest that model-based instruction can be a powerful force for learning, but cannot address questions of how such model-based practices are taken up, interpreted, and used by diverse students over the course of an investigation.

By delving into the complexities of the day-to-day interaction of modeling practices and sense making of one student across the various contexts of the project, this study enriches the existing literature at a more detailed level of scale. A focus on one student’s navigation of the complexity of the content domain and the multiple structures and functions embodied in this study highlighted the possible points of connection or disjunction in making sense of the multiple complexities of an ecosystem, and of the modeling practices that bring that student’s ideas about ecosystems into being.

Conceptual Framework

The conceptual framework for this study is guided by situated cognition theory, using

Doyle’s (1977; 1983) framework of academic task to ground the theory of situativity in classroom practices. Analyzing the phenomenological perspective, what the child brought to the academic task, enabled me to illuminate what ecological understandings were uniquely Jorge’s (Carlsson, 1999), and identify the ecological and systems concepts, structures, functions, and change over time that he perceived and considered most salient in the modeling task—his “knowledge in action” (Driver &

Erickson, 1983). Model-based-reasoning guided the design of the task of investigating, designing, and observing the BioBottle and its components. The interaction of Jorge’s unique perspectives on ecological concepts with the affordances and constraints of the physical and representational models created by the teacher and students, in the context of the academic tasks, result in an explanation for

31 the variation in Jorge’s sense making about ecological concepts across the timeline of the project.

A situated perspective of learning is grounded in the philosophy of American Pragmatism as framed by John Dewey (Bredo, 1994; Brown, Collins & Duguid, 1989). One of Dewey’s (1902,

1938/1963) core arguments is that people negotiate meaning through experience, reflection, and communication. As Bredo elaborates, Dewey sought to resolve the dualism between mind and world, and to “place theory within practice rather than outside of it,” framing activity as a

“transaction between person and environment that changed both” (p. 24). In his essay, The Child and

the Curriculum (1902), Dewey sought to synthesize the dual perspectives of child centered education vs. discipline-centered education. Attunement to the phenomenon of the transitory world of the child, he argued, provides the “germinating seed, or opening bud, of some fruit to be borne” (p.6).

Effective curriculum/instruction “psychologizes” or translates disciplinary content through the child’s experiences that reflect the “origin and significance” of the subject matter (p. 7). The child is strategically invited into an interaction with disciplinary content. Dewey concluded that in this interaction “all activity takes place in a medium, a situation, and with reference to its conditions.

…All depends upon the activity which the mind itself undergoes in responding to what is presented from without” (1902, p. 10).

Theories of situated cognition can be traced back to multiple roots in cognitive psychology, ecological psychology, sociocultural psychology, social anthropology, and the sociological studies of scientists and their work (Brown, Collins & Duguid, 1989; Greeno, 1997, 1998; Lave & Wenger,

1991). Situated cognition theories represent a significant epistemological shift away from cognitive theories of information processing and “mind as computer” and towards a view of learner-inactivity, engaged in particular social and material practices, using cultural tools that provide affordances and constraints for action and sense-making (Greeno, 1991, 1997, 1998, 2009). The social and material practices, physical and cultural tools, and academic tasks (Doyle, 1983) together

32 form the “situation” or context with which learners interact as they make sense of disciplinary content (Greeno, 1998). The following sections highlight key tenets of the conceptual framework used in this study.

Situativity as Interaction of Learners with Complex Instructional Systems

In research on situated learning, the focus shifts from inferring what is in the mind of the individual, to describing how the individual interacts with the complex dimensions of their learning environment, particularly the physical environment (e.g. Gibson, 1977), physical tools such as laboratory equipment (c.f., Latour, 1987; Pickering, 1993), and cultural tools such as language and other symbol systems (Vygotsky, 1978). Within classrooms, both curriculum and pedagogy are shaped through the presentation and negotiation of academic tasks, or cognitive and activity goals for student work and how that work is to be achieved (Doyle, 1992). Greeno and his colleagues

(1998) framed these dimensions in terms of “situativity theory,” arguing that the situative perspective in research captures the learning of either individuals or groups as “trajectories of dynamic systems” (p. 5), focusing on how learners make sense of content through participation, and how such systems function (p.7). As argued by early proponents of a situated approach to instruction:

Many methods of didactic education assume a separation between knowing and doing, treating knowledge as an integral, self-sufficient substance, theoretically independent of the situations in which it is learned and used.…The activity in which knowledge is developed and deployed…is not separable from or ancillary to learning and cognition. Nor is it neutral. Rather, it is an integral part of what is learned.

Situations might be said to co-produce knowledge through activity” (Brown, Collins

& Duguid, p. 32).

Bredo (1994) as well as Greeno and colleagues (1997, 1998) emphasized that theories of

33 situated cognition do not refute or ignore important findings from the earlier bodies of cognitive research that focus on information processing and neurological functions. In this study, I assume that there are certain universal features of human cognition such as limits of short term/working memory, processes of memory storage and retrieval, processing of visual images, symbols, space and time. The key is that situated cognition recognizes that such processes are always influenced by the specifics of a context, that interpretation and meaning will vary with and be contingent on context.

As Bredo observed, “one never steps into the same river twice” (1994, p. 32).

Situations and Problems Emerge in Activity

The situative perspective expands on individual cognition to take into account how learners or agents actively work with unfolding or “emergent” problem spaces (Greeno, 1998), emphasizing that and solution strategies are emergent and rarely controllable; the situation or context shifts as learners or actors make decisions about how to meet their goals. A classic study is that of Lave,

Murtaugh and de la Rocha (1984), who studied how people in supermarkets solve problems of selection and cost not by using formal mathematical algorithms taught in school, but by interacting with their environment and taking advantage of information in that environment to make decisions.

Greeno (1998) proposed that situativity theory offers a new perspective on work and learning by emphasizing interactions among participants and with physical and cultural tools as they construct meaning. This sociocultural perspective originated in the work of Vygotsky (1978) on the interrelationship of thought and language. Lave and Wenger (1991) drew on Vygotsky and Marxist theory to describe cultural-historical structures they term “communities of practice,” in which novices learn from more knowledgeable others through legitimate peripheral participation in a practice (e.g., tailoring, learning to be an insurance adjuster). In these communities meaning emerges through joint interaction and continual work toward shared reference. For example, Goodwin and

Goodwin (1996), in their study of airplane ground controllers, showed that the ability to “see” a

34 plane, is embedded in larger organizational tasks (getting planes safely to and from gates to runways), local tasks (a ground handler loading baggage on to the correct plane), and mediated by supporting tools (written manifests, aircraft ID numbers). Participants with different perspectives

(e.g., maintenance crew) may use some of the same tools and larger organizational structures to accomplish a different task. Participants’ talk constructs the context, or activities that participants are accomplishing through mutual negotiation and accountability. Thus accomplishing the work of

“seeing” planes is anything but straightforward and predictable.

Agency and Activity Shape the Emerging Situation

Situativity theory positions learners as active agents who bring their individual identity, beliefs about themselves as learners, and perspectives of phenomena, to the goals and functions of the group (Greeno, 1998). A phenomenological lens emphasizes humans’ perceptions of and relationships with the natural world in general and ecosystems in particular (Carlsson, 1999). This lens provides an important counterpoint to criticisms that canonical Western science fails to respect multiple cultural worldviews, or to consider personal perceptions, motivation and agency in relation or opposition to Western scientific knowledge (Aikenhead, 1996; Calabrese-Barton, Koch, Contento

& Hagiwara, 2005; Helldén, 2004; Patrick & Tunnicliffe, 2011). Few studies in environmental learning or science education investigate children's understanding of ecological systems from the perspective of the children themselves. While science and environmental learning researchers often invoke the learner as the source for future change, they rarely consider the voices and perspectives of these learners in their formulation of problems or solutions. If children are to bear the burden of changing human relationships to the environment, their voices need to be represented in research.

The phenomenological lens contributes affective and agentive dimensions to learning about and living within ecosystems.


Affordances and Constraints Shape Interaction and Agency

Affordances and constraints, according to Greeno (1994), relate the abilities of an agent in

interaction with the situation or context. Gibson, an ecological psychologist, specified an affordance as all actions that an actor’s can perceive for interaction with the physical environment (1977). While an affordance may be as simple as a handle on door, the affordance is not the physical handle but the opportunity for an actor to twist it in order to open a door. As Greeno elaborated, “The term

affordance refers to whatever it is about the environment that contributes to the kind of interaction that occurs” (1994, p. 338). The abilities of the actor are in reflexive relationship with the affordances that an environment or situation offers. The designer Donald Norman (1999) extended

Gibson’s notion of affordance into the realm of human-machine interaction and emphasized that an actor’s expectations of common usage or past experience play into the notion of affordance: those who have never had to twist a handle to open a door will not likely perceive the affordance the handle offers. In science education, the affordances and constraints of materials, artifacts, tools and models may serve to “open up possibilities” for new ways to solve a design task (Roth, 1996); conversely, physical models may “push back” against student expectations and challenge students to reconsider their assumptions about components, relationships, and complex interactions (Lehrer,

Schauble & Lucas, 2008). The materiality of scientific work and the scientist’s grappling with the affordances and constraints of the laboratory of the field is captured in Pickering’s (1993) metaphor of the “mangle of practice.”

Building on Gibson’s ideas in framing situativity theory, Greeno (1994) argued that the notion of constraint is more subtle and complex than the everyday notions of a barrier or resistance.

Constraints do not act as a barrier; in a sense they are also affordances, presenting a limit on how it is possible for a learner to interact with her environment such as through the provision (or lack of) laboratory materials, time, space in the classroom, or design limits in engineering (Roth, 1996). A

36 person skilled in driving a car is likely attuned to constraints that relate the turning of the steering wheel to the forward direction of the car. In a learning situation, the performance of a child who is instructed to construct a concept map is supported by affordances of the paper and pencil, the conventions of design, and her abilities. Constraints on her performance include her interpretation of what a concept map is meant to do, her knowledge of the conventions used to construct a concept map, and her prior experiences of constructing concept maps. In the context of this study, a learner who has prior experiences with building and observing complex biological models has developed attunements to the constraints of the scale of a model on its function in ways that a novice model-builder may not.

As Greeno (1994) emphasized, the question of how an affordance is perceived by a learner is a “straightforward research problem” (p. 340), whether one is investigating a learner’s interactions with real objects, symbol systems, or models. Over time, repeated experiences of discovering affordances and constraints through situations enable a learner to develop “elements…of a practice”

(Greeno, 1994, p. 339). A learner’s attunement (selective perception) to the constraints and affordances in an activity system “support but do not determine” their participation in activity

(Greeno, 1998, p. 9).

Affordances and Constraints of Physical and Cultural Tools Mediate Sense Making

Both situativity theory and sociocultural theories of learning focus on the role of physical and cultural tools in mediating learners’ actions and sense making; this assumption underlies work model-based inquiry (Greeno & Hall, 1987; Lehrer & Schauble, 2000; Vygotsky, 1978; Windschitl,

Thompson, & Braaten 2008). Vygotsky (1978) distinguished between physical tools as those which support activity, such as laboratory equipment and physical models such as a terrarium or aquarium from cultural tools such as language and symbol systems. Both physical and cultural tools embody the practices, values and norms as well as the history of a community of practice (Lave & Wenger,


1991). Kirsh (2009) argued that humans live most of our lives in socially designed environments rich with tools, such as computers, that are deliberately designed to make us smarter and more efficient in achieving goals. Research following the Vygotskian tradition tends to privilege the linguistic forms of mediation, as evidenced in the “discursive turn” in social science studies since the mid-1970’s

(Wertsch, 1985; 1991). These theories and studies have led to a rich understanding of how social participation structures (Erickson, 1982) in school can invite diverse children into domain-based science discourse (e.g., Lemke, 1990; O’Connor & Michaels, 1996; Wertsch, 1991).

In contrast the emphasis on written and spoken language, researchers in multimodal literacies have argued that privileging the linguistic over other forms of signs such as symbols, art and visual representations can exclude children who see and experience the world actively and visually in their sense making practices (Eisner, 1994; Gardner, 1983; Greene, 1993, Wertsch, 1991).

Eisner (1994) proposed that learners’ ability to express an idea is limited to the types of representations they have experience and skill with, or access to in their sociocultural world. Every form of representation limits what one can express about different aspects of the world. Likewise, children’s experience with different forms of representation will shape their attention and perception of particular features of the world.

The privileging of the linguistic over other forms of mediation is particularly problematic for the vision of science education that builds on the multimodal practices identified in actual scientific communities via social studies of science (Latour, 1987; Lemke, 1998; Jewitt, 2008; Kress, Jewitt,

Ogborn & Tsatsarelis, 2001, Tang & Moje, 2010; Roth & McGinn, 1998). As Lemke (1998) argued,

“[T]he language of science is a unique hybrid: It is natural language as linguists define it, extended by the meaning repertoire of mathematics…, contextualized by visual representations of many sorts, and embedded in a language (or, more properly, a semiotic) of meaningful, specialized actions afforded by the technological

38 environments in which science is done” (p. 33).

Scientific arguments and explanations are inherently multimodal (Lemke, 1998), and make thinking visible to self and others for reflection, communication, and negotiation (Salomon &

Perkins, 1998; Greeno and Hall, 1997; Lehrer and Schauble, 2002, 2005, 2010). Each semiotic mode can “modulate and multiply” possible meanings for both writer and reader (Lemke, 2000, p. 92).

Further, these modalities have been invented and refined within communities of scientific practice that are shaped by the paradigms of a discipline (Kuhn, 1962), which encompass particular worldviews, assumptions of the object studied, and the tools, technologies, and strategies of investigation, communication, and argument (Latour, 1990).

Models as Tools for Sense Making

A key practice emphasized in the new science standards, model-based reasoning, is the main epistemological tool used by modern ecologists to study systems at the micro (microscopic), meso

(human) and macro (bioregional or earth systems) scales (Canham, Cole & Laurenroth, 2003).

Arguments for engaging students in the practices of modeling trace back to the rise of constructivist theory and concern for engaging students in education that reflects the epistemology of science as theoretical, tentative, testable, and predictive (Gilbert, 1991; Hestenes, 1992; Lehrer, Horvath &

Schauble, 1994; Clement, 2000). The Next Generation Science Standards (NGSS Lead States, 2013) emphasize the importance of engaging students in the epistemological practices of constructing science knowledge. In this study, the overarching framework of situated learning and engaging students in the practices of science, particularly modeling, guided the design of the project and the analysis.

As most of the key processes and dynamics of ecosystems are not directly observable at the human (mesoscopic) scale, the discipline of ecology relies strongly on models and modeling to construct explanations and predictions, which are validated, refined, or challenged using empirical

39 data from field studies, experimentation, and construction and observation of microcosms (Canham,

Cole & Lauenroth, 2003; Cruzan, 1988; Jordan et al., 2009; Tamponnet & Savage, 1994). Canham et al. noted that a frequent definition for the purpose of modeling in ecology was “to clarify conceptualization of system structure” (p. 3). Thus the educational design for ecological literacy should address domain-specific knowledge in ecology, domain-general understanding of systems, and the practice of using models to articulate the structure and function of both.

Models of ecosystems are ubiquitous in elementary science classrooms—terraria, aquaria, vermiculture, school gardens. A child first learns to literally see the biotic and abiotic components of ecosystems either by observing the distribution of organisms across a particular geographical and climatic regions, or through physical models. However, simply observing physical models limits students to descriptive, rather than explanatory, accounts of ecosystem structure and functions. To induct students into the scientific view of ecosystems requires curriculum representations and learning experiences that enable students to “see” beyond the physical and concrete features of place or model and perceive the underlying structures (defined as components and interactions) and functions (organization, flow of matter and energy) of the ecosystem. For children to understand ecological systems, they must be inducted into the discourse community and genres

⁠ of communicating about ecology (Bazerman, 1988), which include not only physical, but symbolic and mathematical models. Kirsch (2009) argued that situated cognition “owes us a theory” that will explain how people use physical objects to help them reason and solve problems. He proposed a key question for the design of learning environments:

“C. S. Peirce (1931- 1958) first mentioned this idea - that people use external objects to think with - in the late nineteenth century, when he said that chemists think as much with their test tubes as with pen and paper. What characteristics must a thing to think with have if it is to be effective, easy to use, and handily learned?” (p. x).


Grounding Situativity in Classrooms with Academic Task Theory

Doyle’s framework of academic task (1977, 1979, 1981, 1983, 1986), extends the more theoretical dimensions of “situativity” and delineates the specifics of how learning is situated in classrooms. While sociolinguistic studies of classroom processes and student sense making have contributed significantly to understanding of how meaning is constructed through social processes and discourse, these studies tend to background or omit a key dimension: academic content (Ford &

Forman, 2006). Doyle’s work on academic task emphasizes the role of domain-based knowledge, the type of products and cognitive processes students are expected to use to produce products, and an

“accountability” system (expectations for performance, feedback, and grades) to the situative framework.

A classroom “task” refers to three dimensions of student work in relation to the curriculum: the products or “answers” students are expected to produce, the “operations” or processes that students should use to produce the product, and the resources available to students—not only physical resources but conceptual models such as an example or model of expected level of work

(Doyle, 1983, p. 161). The design of an academic task and its expected outcome frame for students the specifics of the content they should attend to, and the specific cognitive operations they are expected to use, e.g., memorization for a spelling test vs. reading for comprehension. Thus, the notion of academic task specifies the “basic treatment unit in classrooms” (1983, p. 162).

Doyle’s research and analysis of classroom effects and academic tasks illustrate ways in which classrooms are systems: “complex settings with multiple dimensions, many of which operate simultaneously” (1981, p.4), and are characterized by immediacy and unpredictability (Doyle 1977).

Doyle argued that what students learn about a content subject is embedded in the task situation, that in fact, “students experience subject matter as classroom tasks” (1979, p. 142). Therefore, a task in which students memorize and reproduce information results in different knowledge outcomes than

41 a task that requires the construction of inferences. Outcomes can also be mediated by how students interpret or misinterpret the academic task, or find ways to do the work with minimal cognitive engagement, by relying on more vocal or persistent peers, or by overhearing the teacher coach other students. Students’ familiarity with task structures may allow them to be more efficient in attending to “what counts” in a task and therefore affect their engagement and performance (Doyle, 1979).

Doyle’s theory of academic task emphasizes the interaction of individuals and groups in interaction with the content and practices of a conceptual domain, in this case, ecosystem structures and functions. Academic tasks entail a continual negotiation between students and teacher about the nature and outcome of the classroom work. Each student interprets the academic tasks from her or his prior knowledge, experience, perceptions, imagination, motivations, and intentionality. The interpretation of an academic task by the student connects back to the phenomenological dimension: how learners make sense of ecosystems with their unique consciousness.

Doyle’s academic task framework narrows and specifies the boundary of a classroom as a particular type of situated learning environment in which individuals and groups in interaction with academic content. In order to infer meaning from student work produced in classroom activities, research must account for how the academic task was structured by the curriculum and enacted

(Doyle, 1992) by the teacher and students in dynamic interaction. All interpretation of data from student work, both phenomenological and contextual, must consider interpret evidence through the lens of academic task.

Conceptual Framework Summary

The conceptual framework as outlined above leads to the focus of the study on the interaction of the learner, Jorge, with the physical and representational models that he and his peers construct and share through the series of academic tasks designed to elicit his sense-making about the structures and functions of ecosystems. The theory of situativity and its tenets emphasizes that

42 all learners are engaged in a continual process of coming to know, what Sfard (1998) termed the

“participation metaphor” of learning. The metaphor of knowing as participation shifts perspectives away from the characteristics or “intelligence” of an individual student and toward a view of continual change and adaptation:

The participation metaphor does not allow for talk about permanence of either human possessions or human traits. Being “in action” means being in constant flux.

The awareness of the change that never stops means refraining from permanent labeling. Actions can be clever or unsuccessful, but these adjectives do not apply to the actors. For the learner, all options are always open, even if he or she carries a history of failure. This…bring[s] a message of an everlasting hope: Today you act one way, tomorrow you may act differently (Sfard, 1998, p. 8).

Viewed in this way, a research framework grounded in situativity opens space for “the having of wonderful ideas,” Eleanor Duckworth’s (1987) lyrical description of the emergence of meaning in the context of messy, complex situations. This study affords the opportunity to celebrate complexity of how systems, both individual and instructional, operate dynamically and in many cases, unexpectedly. My goal in this study is to learn from that complexity. As Meadows (2008) reminded us:

Let’s face it, the universe is messy. It is nonlinear, turbulent, and dynamic. It spends its time in transient behavior on its way to somewhere else, not in mathematically neat equilibria. It self-organizes and evolves. It creates diversity and uniformity.

That’s what makes the world interesting, that’s what makes it beautiful, and that’s what makes it work (p. 181).

Research Questions

This study entails a detailed case of how the use of a physical microcosm and children’s self-

43 directed representations of an ecosystem constrained and afforded student sense-making in an urban elementary classroom. The case analyzed the experiences of a 10-year old fifth grade student, Jorge, and the variation in his expressed understanding of ecosystems as he interacted with academic tasks, along with models and representations, to design, observe and explain an ecological microcosm. The overall questions addressed by the study were:

Within the context of a culminating class project to design a functioning closed ecosystem, how did the interactions between a learner, the academic tasks, and modeling practices influence what he perceived and communicated about the structures and functions of an ecosystem?

A second, question that further bounds this case emerged from the contingencies of designing the project and findings from the initial analysis:

In particular, how did this student perceive and communicate his sense of the form and function of water in the microcosm and the interaction of water with other living and non-living components of an ecosystem?

Overview of the Dissertation

In this chapter I have provided an introduction to the problem and rationale of the study, and have established the study’s conceptual framework. Chapter 2 provides an expanded discussion of the findings and limits of research related to the study. In Chapter 3, I describe my research process, using my conceptual framework to justify the design and methods used for the study. I present a history of my prior work with the teacher and class, then describe the setting for the study, including the school, classroom, teacher, students, and my position in the classroom as both researcher and scientist. I present the timeline of the study and the flow of lessons across the 12 weeks. I describe data collected, and analysis procedures, with examples to address trustworthiness.

Finally, I analyze how the changes from my original plan to what played out limited some of my initial goals for the study, and discuss limitations in the methodology.

Chapter 4 presents findings from my analysis of Jorge’s work with the BioBottle model. In

Chapter 5 I bring together the findings to address my research questions, relate my findings to the literature, and discuss limitations of the study. I then propose implications for teachers and researchers, and explore possible research directions.




The literature addressing students’ understandings of ecosystems and ecological concepts is vast and scattered across multiple and sometimes incommensurate traditions of research. Multiple lines of research in education have grappled with how children understand ecological concepts such as photosynthesis, decomposition, materials cycles, and energy flow, as well as how to design learning experiences or learning environments that can support children’s sense-making about ecosystems. Ideally, studies from different research perspectives may help us move closer to a shared understanding of how to support children’s sense making about ecosystem structures, functions, and change over time using either models or fieldwork (Cobb, 1996; Cobb & Bowers, 1999).

First, I present an overview of the conceptual domains of ecology and systems. I then review the theoretical and empirical literature related to this study. Following a roughly historical timeline, I first review studies in the conceptual change paradigm that focus on students’ “misconceptions,” or alternative conceptions related to ecosystem topics. I contrast this work with studies from two paradigms—developmental and phenomenographic studies—that shift emphasis from what scientific understandings children lack and towards the naïve but sensible and often coherent ways that children reason about biology and causality, and their experiences of ecosystems beyond the cognitive. Lastly, I examine studies that approach the teaching and learning of ecosystems from a holistic and dynamic perspective, many of which entail modeling practices as a core instructional or research strategy. In the summary section I reiterate how this research influenced the questions and methods of the present study.


Ecology and Systems Thinking in Western Science

Ecology, like the whole of science, has been a house with many doors, some leading to one view of nature, some to another (Worster, 1977, p. 348).


The concept of an ecosystem and field of ecology is a relatively new development in the history of science (Odom, 1968; Worster, 1977). The essence of ecology as a science lies in the assumption that there is order in the natural world that can be predicted and explained (Worster).

The term "ecology" was first coined by Ernst Haeckel in 1866 as "Oecologie,” from the Greek root

oikos, a reference to the family household and the logic, or economy, of its operations, both cooperative and competitive. The term captured not only fields of inquiry across geology, botany, species diversity, and Darwin's emerging theories of evolution, but also by 1915, in the journal

Ecology, a philosophical lens emphasizing integration of biological organisms and the specifics of their environment as a dynamic whole (Worster).

Early 20th century ecology studies focused on observation and description of organisms, populations, communities across the earth. In the middle of the 20th century, ecological investigations shifted toward developing models and theories to quantify and predict the amount of energy generated, utilized, and lost due to the disorganization of heat energy resulting from each energy transformation in an organism or across the system (Odom, 1968). Likewise, the cycling of materials in various systems began to be quantified and modeled mathematically to predict limits to growth as populations expand and compete for limited resources of air, soil, water, and nutrients.

Today the discipline of ecology has evolved into a constellation of fields with a strong focus on population ecology and evolutionary biology (D’Avonso, 2008). Across these domains, ecologists employ a dialectic of two epistemological practices: fieldwork and modeling.

The practice of ecology has also broadened significantly to study and ameliorate

47 environmental problems, such as pollution and chemical persistence, related to human actions.

Catalyzed by Rachel Carson's (1962) groundbreaking study of bioaccumulation of organic pesticides, the field has increased focus on the growing human population's role in disrupting the evolution and equilibrium of earths ecosystems, and also the biosphere (Starr & Taggert, 1984; Worster, 1977).

Exchanges of energy, materials and information at the level of ecosystems, and alterations of those by humans has led to one of the most pressing socio-scientific challenges to the survival of the

Earth's biosphere: climate change. Understanding the structure, function and behavior of ecosystems underlies understanding of these larger issues.

Steve Van Matre, the founder of earth education programs, argued that ecological understandings are key to changing human knowledge and behavior. He outlined a succinct explanation of ecosystems and the underlying concepts of energy flow, materials cycles, interactions, and change over time:

To understand life on the third planet from the sun, you have to understand the flow of light energy bathing the planet each day, how it is captured and utilized, and how it powers the great cycles of building materials of all living things—the air, the water, and the soil. Together, that energy and those materials combine in varying amounts in different places and times across the surfaces of the earth. From the beginning, those variations in the quality and quantity of energy and materials available have given rise to various communities of life whose inhabitants are continually interrelating with one another as they go about obtaining their own energy and material needs. And within those percolating pools of life there is the constant ebb and flow of change. Things change one another, they change their surroundings, and their surroundings, in turn, change them. In short, all living things draw upon sunlight energy for their existence, and each represents a temporary ordered

48 arrangement of matter interacting with its neighbors. Each builds up, then breaks down as the materials of its own body inexorably crumble over time (Van Matre,

1990, p. 105).

Water links the abiotic and biotic structures of an ecosystem on multiple scales, yet it is not typically taught in conjunction with ecosystems at the elementary school level. On a macroscopic scale, water is a key abiotic resource in Earths' biosphere and ecosystems; the overall amount and quality of the water available to an ecosystem, along with the amount of energy from the sun, determines the climate of an ecosystem, and in interaction with processes of evolution and succession determines the species, populations, communities, and distribution of organisms in an ecosystem. On the microscopic scale, water plays a central role in the biological processes of synthesis, respiration, and metabolism; all organisms need water to survive.


The language Van Matre used constructs a perspective on ecosystems not as an inert fact or thing, but as a process continually unfolding. The language is congruent with the perspective of general systems theory, which views phenomena as emerging from a structure of interrelationships among components, that results in emergent patterns of behavior over time (Capra, 1996; Laszlo,

1996; Meadows, 2008; von Bertalanffy, 1968). General systems theory emerged within the field of biology (von Bertalanffy, 1968) with the recognition that all life has evolved into complex, nested structures which are organized and function to make maximum use of available energy and materials in its environment (Capra, 1996). The ontological shift from reductionism to holism starting in the late 18 th

century gave rise to the maxim that “the whole is greater than the sum of its parts,” and that patterns rather than parts are key to explaining the structures and functions of living systems.

As Kauffman (1980) explained, “a system is a collection of parts which interact with each other to function as a whole” (p. 1). In living systems such as an ecosystem, structure and function

49 are explained as the interrelationships among the “parts” or subsystems. Or as Kauffman vividly put it, if you divide a cow in half, you do not get a smaller cow—you get a mess. You may create hamburger, but the system functioning as cow is lost. The Soviet psychologist Lev Vygotsky applied a systems definition to distinguish between the informal concepts acquired through experience, which he called “heaps,” with the “scientific” concepts one develops by mastering language and symbol systems in a coherent network of knowledge (1978).

A critical structure in all systems is that of mutual causality, or the feedback loop (Kauffman,

1980; von Bertalanffy, 1968). Systems have evolved complex networks of balancing feedback loops to maintain stability in the context of change in the external environment or disruptions to parts of the system. For example, in an ecosystem, the organization of relationships in a community of plants and animals, the food web, experiences constant flux in predator and prey populations, as well as populations of producers (plants) and consumers (organisms that obtain energy and materials from plants or other organisms). Yet the overall pattern of the biomass or energy “pyramid” (an expert model accepted by systems ecologists) remains constant.

The structure/function perspective of systems theory also applies to the interaction of the biotic (living) and abiotic (non-living) components of an ecosystem. All life depends on the energy of the sun, which plants and blue-green algae use to transform water and atmospheric carbon (in the form of carbon dioxide) into the sugar molecules that become incorporated into the plant’s structure, and are used to carry out metabolic functions. Non-autotrophic organisms depend on plants to obtain the energy and materials they need to maintain their structures and functions.

Likewise, the function of decomposers in an ecosystem transforms these once living material components back into soil and basic molecules so that materials can continuously recycle. These systems principles are vital for all citizens to understand, not only to maintain the balance of

“natural” ecosystems but to reform our human practices of using Earth’s resources and producing

50 non-decomposable waste at a rate that is destroying the stability and resiliency of ecosystems globally.

The reductionist science of the Enlightenment gave Western civilizations a powerful tool for analyzing phenomena; that perspective is complemented by systems theory. As Laszlo (1996) argues, in the complexity of today’s interlinked global systems, we need to be able to “see” through the artificial disciplinary boundaries that Western scientific culture has constructed. We need to be able to think beyond categories of “thing” and “part” to recognize and describe processes (Chi, Slotta and deLeeuw, 1994)—most importantly, the interactions of parts within systems that generate dynamic, sometimes chaotic or nonlinear change over time (Capra, 1996, Resnick, 1995, Senge,

1990). Thus our understandings of ecosystems must be both particular and general, a paradox for both living and for learning.

Definitions and delimitations: Ecosystems, Environment, and Nature

Before delving into the literature it is important to note that the terms “ecosystem,”

“ecology,” “environment” and “nature” are often used interchangeably by researchers, in the fields of science education and environmental learning. In this study, ecology and ecosystems refer to the interdisciplinary field that studies the interactions among organisms, and interactions of organisms with their environment (Earth systems). This was the conceptual goal of the Investigating Ecosystems curriculum used in the classroom of this study (BSCS, 1999). The field of environmental science extends the discipline of ecology with a focus on designed human-environmental interactions such as conservation management, waste and remediation, and toxicology (Journal of Environmental

Sciences, 2015). The term nature refers to any natural phenomena of the physical earth or environment, including the solar system and the universe beyond human life.


Review of Empirical Literature

Children’s Conceptions and “Misconceptions”

In the early 1980s science education researchers first began to investigate children’s ideas about ecosystems or ecological processes, at different ages, with and without instructional interventions. The underlying assumption in these studies posited learning as a process happening in the mind of an individual, with the goal of instruction to challenge “naïve” ideas and promote conceptual change toward more “expert” or scientific, understandings of ecological phenomena.

The goal of these studies was to identify children’s conceptions and misconceptions or “alternative conceptions” (Driver, 1983; 1989) in order to develop new curriculum or instructional strategies.

John Leach and his colleagues (Leach, Driver, Scott, & Wood-Robinson, 1995, 1996a,

1996b) conducted extensive studies of the cognitive/naturalistic studies on children’s thinking related to ecological concepts and summarized other’s work for the Leeds National Curriculum

Science Support Project from 1988-1992. They assume a constructivist theory of learning—that children’s ideas about the world develop from their experiences, and often hold onto these

“everyday” beliefs even after formal instruction in science.

This group conducted a large-scale study of students’ understandings of six interrelated concepts: photosynthesis, respiration, decay, habitat, transfer of matter and energy between organisms, and transfer of energy with the environment. These ideas are subsumed under three “big ideas” of ecology: the cycling of matter, the flow of energy, and interdependency among organisms.

Five diagnostic “probes” were designed to elicit students’ conceptual understanding of the above ideas in two or more contexts. Responses were gathered from children near an industrial city in

Great Britain through a combination of written tasks and interviews, and analyzed qualitatively through iterative coding and tests of inter-rater reliability. The study design clustered participants in four discrete age ranges— 5–7 , 7–11, 11–14, and 14–16 —who were interviewed at roughly the

52 same point in time. This cross-sectional design allowed the researchers to suggest how children think at different ages. However it did not allow them to make claims about how students’ ideas develop over time or in interaction with specific instruction (Leach et al., 1995, p. 730).

The findings from this study are summarized below, along with citations of other studies that found similar results. For the Leach results I discuss findings related to ages 7–11, which match the age of the student in this current study. Data were reported in two articles related to the cycling of matter (Leach et al. 1996a)—including needs of plants, photosynthesis, the sources of matter for plant and animal growth, and the process and role of decay), and the interdependency of organisms

(Leach et al. 1996b)—including communities, relative population size, relationships of organisms in food webs, and forms of interdependence. The following sections are organized around the three

“big ideas” of materials cycling, energy flow, and interdependence.

Materials cycling

The cycling of matter entails both biotic and abiotic processes and components of ecosystems, such as the transformation of inorganic carbon in the form of CO


to the organic carbohydrates which comprise the majority of a plant’s physical mass and also provide energy and nutrients for animals in food webs.

Role of plants and photosynthesis. One common alternative conceptions revealed by multiple studies regards plants getting their “food” and material for growth from the soil rather than by manufacturing carbohydrates through photosynthesis (Barker & Carr, 1989a, 1989b; Bell, 1985;

Stavey, Eisen, & Yaakobi, 1987; Wandersee, 1983; Wood-Robinson, 1991). In the Leach et al. study

(1996a), children were asked what a plant “needed” to stay alive and healthy. For children ages 7–11 the most common responses were water (~75%), soil (~55%), and sunlight (~40%), followed by a small percentage (less than 10%) who cited oxygen, carbon dioxide, or air. When asked about sources of food for plants, almost no students age 7–11 stated that plants make their own food, but

53 they continued to cite water, sun, soil, carbon dioxide as possible sources of “food” for producers.

Most children tended to use the terms food, nutrient, and energy interchangeably in their interviews or explanations. Likewise, when asked about sources of matter for animal growth, students in this study focused on the needs of animals rather than the actual processes involved in digesting, transforming, or excreting waste (Barker & Carr, 1989a; Smith and Anderson, 1986), with a common notion that food gets “used up,” possibly to provide energy for living and growing. Barker and Carr (1989a) noted that children’s ideas about food are context dependent—different for people than for plants.

Decomposition. The process of decay is typically seen by upper elementary students as a natural feature of change over time without any causal mechanism (Leach et al. 1996a, Hogan &

Fisherkeller, 1996; Smith & Anderson, 1986). Hogan and Fisherkeller (1996) argued that nutrient cycling, like photosynthesis, is a “linchpin concept” in ecological thinking as it addresses how organisms and the environment continually exchange materials such as nutrients and gasses, and how decomposers release nutrients locked in dead organisms so that they can be reused. Leach et al. found about 40% of upper elementary students in their study provided causal explanations of decay based on food being eaten by visible or invisible organisms. About 25% of students age 7–11 also referred to physical processes such as air, sun, and heat. After age 6 many students cited other experiences with decay, either food in the refrigerator or dead animals; almost none mentioned composting. Very few students at the age of 16 provided an explanation that included the principle of conservation of matter (cf. Helldén, 1995). Leach et al. (1996a) note that many students made no reference to processes of decay in their life experience, either the phenomenon or the time scale.

Explanations rarely referred to the role of detritivores, particularly microorganisms, other than some children’s vague reference to “germs” (Leach et al. 1996a), which is a common idea across young children (Byrne, 2011).

The role of water and the water cycle in ecosystems. Research on children’s understandings of the

54 water cycle tend to be positioned in the field of earth or physical sciences, with few connections made between forms and transformations of water as a substance and the importance of water to all living organisms. For example, an early constructivist study by Bar (1989) described the water cycle as a series of “steps:”

[W]ater evaporates from the sea and other water sources, when conditions of a barometric low prevail this vapor rises, cools and condenses into little crystals of ice and small droplets of water; these droplets increase in number, become heavy and fall down as rain; rain water flows back to the sea, sinks into the ground, becomes ground flow, or re-evaporates (Bar, 1989, p. 481)

Bar found that students younger than 11 years tended to focus on visible water, and explained processes of water transformation as water “disappearing.”

Ten years later in a study with children age 9-11 in a rural part of the UK, Dove, Everett and

Preece (1999) explored children’s drawings of to gauge their understanding of the concept of a

“river basin” and “where the river began and ended” (p. 489). Like Bar (1989), Dove et al. identified levels of complexity of understanding as well as conventions students used to represent rivers. Most children admitted in interviews that their drawings of rivers were “made up” as opposed to being based on experience. About twelve percent of student drawings included people, but as this was not a focus of the research, the interaction of people with rivers is not explored.

With the rise of systems ecology and a new focus on developing curriculum to support students’ development of environmental literacy, there is a new movement to identify learning progressions (NRC, 2012, NGSS Lead States 2013), which are intended to trace students’ sense making from their initial novice understandings of systems to constraint-based or model-based reasoning. For example, Covitt, Gunckel and Anderson (2009) reported initial results from a series of studies that examined how students’ understanding of water in environmental systems may

55 progress from kindergarten to Grade 12. Their definition of environmental science literacy is based on a framework developed by the US. Long Term Ecological Research Network, Research

Initiatives Subcommittee (2007). In this systems-based framework (p. 39), environmental systems that include fresh water systems include living systems as well as geological systems; however the biological processes that move water or alter water composition only include transpiration. On the human and social economics side of the model, water is provided as an ecosystem service to humans for “personal use and consumer products.” (p. 39). As the focus of this work is on environmental literacy, not ecological literacy, the foregrounding of interaction between humans and their environment is appropriate to the field of environmental science, but also informs this study.

Covitt et al.’s (2009) first research question addressed student understanding of natural systems: “What do students know and how do they reason about the structure of environmental systems through which water flows?” (p. 40). The second question focused on the processes that move water and other materials through the system. Results from an initial rubric-based assessment of 120 randomly-chosen student responses from elementary, middle and high school confirmed findings from earlier research. Students focused on the visible movements of water in the environment, but failed to trace water from visible to invisible parts of the system. Likewise, students did not represent microscopic elements of the system, nor macroscopic dimensions such as watersheds. Covitt et al. also found a lack of evidence that students were using the principle of conservation of matter, even at the high school level. From a more positive perspective, the results also showed most students starting at the upper elementary level recognize that water is a component in the atmosphere, and that stores of water exist underground. They recognized that humans rely on natural systems to get their fresh water, and that human actions such as pollution or water distillation can affect water quality.

Explanations for these findings reflect explanations discussed earlier: the school curriculum

56 is designed so that students study different water systems and processes in different science disciplines. Unless they take an environmental science class, students may not encounter humandesigned water systems at all during schooling. Covitt et al. (2009) recommended that students have more informal and formal experiences with dimensions of natural and human designed water systems as well as work with models such as solar stills. These recommendations are important, but they do not address bridging the gap between the geophysical science teachings about water and the role of water in the life and health of all living organisms and ecosystems, not just humans and human systems.

Energy in ecosystems

Multiple studies of students’ misconceptions or partial understandings of energy have been conducted in the physical sciences (Pfundt & Duit, 1994; for specific studies see Solomon, 1982,

1983a, 1983b; Kesidou & Duit, 1993). One challenge is that the term energy has many uses in everyday life, typically associated with ideas about stamina, strength, power (Jennison & Reiss, 1991).

Scientists use the same word but in a discourse in which the term has very specific meaning (Driver,

Asoko, Leach, Mortimer & Scott, 1994). A related challenge is for learners to understand that principles related to energy in the physical world also apply to biological systems (Barak,

Gorodetsky, & Chipman, 1997). Jennison and Reiss suggested that it may be more useful in biology to talk about energy using verbs rather than nouns: transferring, warming, and “becoming less useful.”

H.T. Odom, the pioneer of mapping energy in an ecosystem, used the metaphor of “flow” to describe the transfer of energy from the sun to plants and to animals at different trophic levels in a food web (1968). In his words—

The behavior of energy in ecosystems can be conveniently shorthanded as “energy flow” because energy transformations are directional in contrast to the cyclic

57 behavior of materials. The potential and kinetic components of energy flow through an ecological system are lumped under the designations, production (P) and respiration (R) (p. 11).

Food webs as models of energy flow in ecosystems. In bioenergetics, or the study of energy in biological systems, entropy (disorganization of energy organization due to its transfer to heat in the process of respiration) limits the total biomass of populations at trophic levels in the food web. The structure of food webs in actual ecosystems is more complex than those models commonly taught at the elementary grades and depends on the ratio of P and R within organisms as well as the flow through organisms at different trophic levels (Odum, 1968).

Models of food webs, as well as the pyramid of trophic population numbers or biomass are a staple of ecology education, starting in the early grades, most often used as an illustration of the concept of interdependence among organisms (Leach, Driver, Scott & Wood-Robinson, 1996b).

Regardless, a multitude of studies in various research traditions have found that elementary, middle, and high school children’s explanations of the structures of food chains and food webs indicate limited and stereotypical understanding such as only being able to trace a food chain “up” from producers but not down from consumers, or failure to consider that a perturbation in one population in a food web might result in changes to populations throughout the web (Barman,

Griffiths & Okebukola, 1995; Eilam, 2002 & 2012; Gallegos, Jerzano, & Flores, 1994; Gotwals &

Songer, 2010; Grotzer & Bell Basca, 2003; Leach, Driver, Scott & Wood-Robinson, 1996b; Lennon

& DeBoer, 2008; Reiner & Eilam, 2001; Roberts & Johnson, 2005; Webb & Bolt, 1990). Even undergraduate students responding to a survey struggled with basic concepts of energy flow in biological systems (Boyes and Stanisstreet, 1990, 1991). In a phenomenological study with upper elementary children about food chains and lunch conducted by Roberts and Johnson (2005) results showed that students easily recited the basic model “up” a human food chain in an illustration of

58 their local desert ecosystem, but when asked to trace back the components of their lunch “down” the food chain, few could do so accurately.

Multiple explanations have been posed to explain students’ lack of understanding of energy and food chains, along with multiple recommendations for instructional interventions to address misconceptions. Leach et al. (1996b) note that upper elementary children constructed communities of plants and animals based on their familiarity with particular ecosystems, especially woodlands, and described interdependence in terms of obtaining food or shelter. They also found that about 50 percent of children at all ages in the study identified producers as the most numerous organism in an ecosystem, children in the range of ages 7–11 relied either on teleological reasoning (more producers are there for the benefit of providing food for other species), or descriptions of natural systems.

When the same children were asked a different question about a particular food chain in the “Eat” probe, their explanations were often different. One feature Leach et al. noted was that most children talked about the animals in a food web as individuals, not populations (often reinforced by the illustrations of a food web used in research probes and textbooks). Students rarely considered competition among organisms for limited resources, or the interdependency of predator and prey populations.

The explanations given in many studies relate back to children’s unfamiliarity with the phenomenon of energy: that population sizes at tropic levels decrease logarithmically as energy is used in the processes of photosynthesis, digestion, and made unavailable (less organized) as it is transformed into heat as a byproduct of biological functions. Chi (1992) and Chi, Slotta and deLeeuw (1994) proposed that students make an ontological error of categorizing energy as a type of

“thing” rather than “process,” or use materials cycles as an analogy for thinking about energy in ecosystems. Given this, making sense of food chains is likely made more difficult for children who are concurrently learning about materials cycles. A related explanation offered is children’s limited

59 understanding of what is meant scientifically by the term food, with a lack of microscopic processes such as chemical transformation into the physical materials of an organism (Rowlands, 2004).

Additional challenges exist for students to think macroscopically about an ecosystem as undergoing complex, dynamic, and simultaneous processes, and that many factors affect size of a population other than predator-prey relations. For example, Green (1997) and White (1997) concluded from a series of experimental studies that adults typically have difficulty reasoning about the feedback relationships between predator and prey relationships; thinking in both children and adults tends to be all-or-nothing: predators devour prey until there are no more prey to eat, and then the predators die.

Limitations of misconceptions studies

Leach et al. (2006b) made an important point that food webs are idealized, “expert” models constructed from many specific field studies of different ecosystems—a point also made by H.T.

Odum early on in the field of systems ecology (Odum, 1968). Thus the relationships between size of populations in ecosystems depend not only on the evolution of food webs stabilized over time, but on factors affecting births and deaths, and the multiple choices of food for omnivores (Polis &

Strong, 1996). An important finding in Leach et al.’s studies (1996a, 1996b) was that students tended to rely on their experiences or knowledge with specific plants or animals, suggesting that their reasoning was linked to specific contexts, or lack of contexts in terms of decay, communities, and food webs, described above. Although most of the older students had encountered scientific explanations of photosynthesis, respiration, carbon fixation, and the role of energy in the transformation of molecules, few used these explanations. Leach et al. (1996a) surmise this was due to these microscopic processes having less “intuitive appeal” and not being reinforced in students’ life worlds or culture (p. 32). Cultural references such as “plant food” may further reinforce everyday notions that conflict with scientific explanations (c.f. Hogan & Fisherkeller, 1996, Wood-Robinson,


1991). Many children after age eight mentioned gas cycling between plants and humans, typically in the form that plants give off oxygen so human beings can breathe, and that plants clear carbon dioxide that humans exhale. Hardly any students recognized that plants also undergo a process of respiration (Leach et al. 1996a). Leach et al. note that even older students failed to make a connection between the physical environment (moisture, temperature) and its impact on cycling of matter or decay.

Attempting to explain theirs and others’ similar findings, Leach et al. (1996a) proposed three possible causes. One was that science teaching in the United Kingdom’s secondary schools typically does not address the process of decay or connect this process to cycling of matter (Leach et al.

1995). The second was that many of ecological and biological processes are taught in isolation (Bell,

1985, Leach et al. 1996a) The third was that students simply do not consider a need to scientifically explain micro processes because these are familiar and not salient to students’ everyday lives (Leach et al. 1996a, Barker & Carr, 1989a. The researchers conjectured that a teaching approach that emphasizes “relationships between organisms and life processes may lead to more integrated learning” (p. 33), although they still assumed a need to teach the processes in isolation as well.

There may be other explanations for students’ apparent misunderstandings or alternative conceptions found in the studies above. The first is that the school curriculum, textbooks and the first round of science standards, particularly the NRC National Science Education Standards (1995) are organized around disciplinary perspectives and focus on the details within a discipline to the detriment of connections across scientific fields (Booth-Sweeney & Sterman, 2007; Carlsson, 1999;

Leach et al. 1996a, 1996b; Hogan & Fisherkeller, 1996; Hogan & Weathers, 2003). This means physical science such as energy transformation is taught separately from biological science, which is taught separately from human health (if health is taught at all). These first generation standards also emphasize the canon of scientific knowledge as a thing to be attained and assessed accordingly. The


Next Generation Science Standards (NGSS Lead States, 2013) emphasize “big ideas” in science such as energy flow in organisms and ecosystems, as well as cross-cutting themes such as systems. However, teachers and students are still left to grapple with how to make the connections until a next generation of curriculum focused on holistic views is developed.

A second possible explanation, directly connected to situated cognition theory, is that for the majority of people, neither students nor their families at the time of these studies had daily interaction with ecosystems processes such as growth or decay in contexts such as gardening, composting, or interaction with an ecological system such as a local woods. Many of the prompts in these studies questioned children about unfamiliar or stereotypical ecosystems about which they may have had little knowledge or experience, such as forest ecosystems used with children in urban environments. For example, in Byrne’s (2011) study of children’s understandings about microorganisms, he found that most students between age 7 and 11 considered microorganisms as sources of disease rather than beneficial to ecosystems, evidenced by describing compost heaps as a place where they might get infections. Or, as Adenyi (1985) pointed out, it could be that students did not make the connection between science taught in school, the research questions, and their cultural practices of farming.

Leach (1996), and Carlsson (1999) made an important point that the research methods used in the above studies also rely on a positivist and reductionist view of science, using tasks that focus on specific topics related to ecosystems but not connecting those topics in a holistic manner.

Surrounded by a culture that emphasizes linear cause and effect and de-emphasizes systems thinking, children and adults make reasonable, if scientifically incorrect, sense of phenomena related to complex dynamic systems. For example, Booth Sweeney and Sterman (2007) found that both students and teachers from two middle schools demonstrated limited systems thinking abilities, such as one-way causal thinking vs. feedback loops, lack of attention to time delays. Carlsson argued that

62 to understand how students conceive of complex, dynamic systems, not only do school practices need to emphasize systems relationships, but researchers also must adopt holistic research strategies and focus on the interrelated structures and functions of ecosystems to reveal nascent holistic thinking in research participants.

Lastly, and leading to the importance of developmental research, Metz (1995, 2000) advances the argument that much of the research findings on children’s “limited” ability to reason has less to do with their cognitive maturation, and more to do with their lack of content knowledge.

As Metz eloquently argues:

A vicious cycle has emerged here. Children’s performance in the laboratory is frequently handicapped by weak knowledge of the domain within which they are tested. This weak knowledge has resulted in poor reasoning and thus an underestimation of their reasoning capacities. This underestimation of their reasoning capacities, interpreted as a ceiling on age-appropriate curricula, has resulted in unnecessarily watered-down curricula. The watered-down curriculum has led to less opportunity to learn and thus weaker domain-specific knowledge, again undermining children’s scientific reasoning (Metz, 2000, p. 374).

Development of Biological and Causal Reasoning

In the studies of students conceptions and misconceptions of ecological concepts discussed above, children’s ideas about ecosystems are compared to expert scientific understanding, with a goal of developing (and justifying) instructional interventions that will promote conceptual change toward more scientific views (Posner, Strike, Hewson & Gertzog, 1982). More recent educational policy emphasizes instead that children’s everyday understandings and experiences are a potential source of strength and a resource to build on while inducting children into more scientific practices and views of phenomena (NRC, 2012). This section reviews the developmental literature on how

63 children reason about the biological functions of organisms from their own embodied experiences and perceptions, which is both an affordance and a constraint on their sense making about ecosystems. This literature also introduces a focus on children’s ways of perceiving and making sense of processes that support living systems, which affect how they interact with real or model ecosystems. Studies in the developmental and phenomenological tradition address how children (and adults) develop sense-making capacity and how they see the world from the position of the research subject. Developmental and phenomenological studies start with an assumption that “naïve” or everyday explanations of biology or ecology are strengths to build upon as opposed to misconceptions that need to be eradicated.

In expanding the boundaries of previous stage-based and constructivist explanations of learning based on Piagetian theory and research methods, Donaldson (1987) questioned whether

Piaget’s findings on the limits of students’ ability to reason abstractly until adolescence might have a different interpretation. She told a joke from a comic about a child’s misunderstanding of the idea of

“present” on the first day of school. She then observed:

The obvious first way to look at this episode is to say that the child did not understand the adult. Yet it is clear on a very little reflection that the adult also failed, at a deeper level, in understanding the child—in placing himself imaginatively at the child’s point of view.

Donaldson suggested it is the adult who forgets the gap in understanding between herself and the child. Who in this situation should be labeled “egocentric?” Slight changes in wording or perceptual cues change children’s understanding of what question is being asked, what task they are responding to. Donaldson also argued that understanding is not an all-or-nothing state; the meanings of words and utterances develop as children gain experiences of producing and comprehending them.


In a review of Piaget’s work and theory, Metz (1995) provides further evidence that science curriculum and instruction based on misinterpretations of Piaget’s work underestimated children’s ability to engage in scientific inquiry and to reason about abstract scientific concepts such as interdependency in ecosystems. Metz provided considerable evidence that children’s expression of scientific ideas and practices is more constrained by their lack of domain-based content knowledge rather than limits of their cognitive structures.

Naïve biology

Studies of young children’s naïve biology illustrate ways that children draw on their everyday experiences and bodily knowledge to reason about organisms and their needs. Developmental studies in children’s biological reasoning suggest that even very young children reason systematically and plausibly about many biological processes. These studies demonstrate that children come to school with a richness of ideas, that learning is a process, and that it develops through both physical maturation and experiences in the world (instructional or informal).

The body of research by Inagaki (1990a) and Inagaki and Hatano (2002) suggests that children’s everyday reasoning about biological phenomena demonstrated plausible and “generative” analogical reasoning based on their experiential knowledge about living organisms, especially themselves. For example, in their 1987 study, Inagaki and Hatano asked questions of five-year-olds that did not rely on life experiences to answer, such as keeping a baby rabbit small forever. The majority of students gave predictions and some gave explanations, reasoning from knowledge of themselves needing to eat and inevitably growing. Children in another study reasoned more effectively about frog behavior based on their experience raising goldfish (Inagaki, 1990b).

Children’s analogical reasoning is not only generative but also constrained by similarity to humans and their knowledge about the attributes of a living organism. Not all analogies children constructed were scientific; many showed anthropomorphic reasoning (e.g., if a human who is taking care of a

65 tulip dies, the tulip will feel sad). On the other hand, by age 6 most children could distinguish between living and non-living things, and recognized both plants and animals as living based on biological criteria, such as the need for food or water, and the phenomenon of growth and change

(Hatano & Inagaki, 1994).

Inagaki (1990a) proposes three factors based on a situated account of cognition to explain these findings. In the real world, guessing by using prior knowledge is more likely to be rewarded. In school or laboratory test settings, students are more aware of the expectation that there is one

“right” answer expected, and may be less likely to venture an explanation or plausible analogy.

Secondly, people judge inferences based on whether they would make sense in a realistic situation.

Third, in real life, people are motivated to propose and try out plausible solutions, not by an external reward but to meet their own needs in a situation. Inagaki (1990a) emphasized that reasoning anthropomorphically works best in scientific fields closer to human experience, such as biology or psychology, but less so in fields such as physics (e.g., diSessa, 1993).

This “naive biology” of children (Hatano & Inagaki, 1994) is not yet a complete theory, but one possible explanation for children’s responses. Children’s responses based on teleological or purposeful explanations in biology (e.g., the plant transforms carbon dioxide into oxygen so that humans can breathe) emphasize the “why” of things, while their vitalistic causal explanations (e.g., blood delivers energy to keep our bodies alive) are an attempt to explain the “how” or process of biological functions when they have incomplete knowledge of the inner organs and organ systems of living things. According to these researchers, the change towards conceptual and causal biological reasoning is notable by the age of 8.

These findings support findings about naïve or young learners in more recent work on learning progressions (e.g. Covitt, Gunckel & Anderson, 2009; Gunckel, Covitt, Salinas & Anderson,

2012), particularly that younger children tend to reason from what they observe and experience at

66 the meso-scale of their daily life—their bodies, short timelines, visible phenomena.

Causal reasoning

Developmental studies on children’s understanding of ecosystems are few. A notable exception is Tina Grotzer’s and colleagues’ work applying theories on the development of children’s causal reasoning to their understanding of the interrelated structures of ecosystems (Grotzer, 1993;

Grotzer & Bell Basca, 2003; Perkins & Grotzer, 2005). In her dissertation research study (1993),

Grotzer investigated how 7- to 9-year-olds understand linear vs. complex cause and effects in natural systems under two teaching interventions (linear causal relations vs. complex causal relations), compared to a control group. Grotzer's (1993) findings suggested that given experience with complex causal relationships in a natural system, children appeared more competent at tracing multiple causal links than would be predicted from earlier studies.

Key to children's developing of understanding was the use of explicit models and causal

“structures” to demonstrate multiple relationships and complex causality, an intervention Grotzer and colleagues call RECAST, or “activities to Reveal the underlying Causal STructure.” Children in this “explicit models” study condition showed the strongest gains from pre- to post-test, and also demonstrated some transfer to new systems. In a modified reenactment of this study with children in a local school district, Grotzer and Bell Basca (2003) found that students in the group which experienced the curriculum along with both causal experiences and discussion about those causal patterns showed a significant difference from the control group (t (26) = 2.75, p = 0.01). However, they did not compare the discussion group to the control group, so their results could be explained as the effect of two interventions over no intervention.

Grotzer (1993) proposed three analogs or external models of complex processes of causality: radiating, branching, and cyclic (p. 19), taken from patterns found in the natural world. She argued that these are basic “building blocks” that can be combined to construct more complex causal

67 models. A radiating causality model describes a pattern of “many effects [that] emanate from one cause” (p. 19). The pattern may be two or three dimensional. Radiating causality may be seen as perturbations in natural or social systems. A “branching causal model” (p. 21) describes how causes may result in effects on either the whole system or a part. In her food webs research tasks, Grotzer reasoned that both the radiating and branching complex models could explain different effects of causes at different levels of the food chain. For example, both a radial and branching model explains why interrupting the flow of energy at the level of sun or plants would have a catastrophic effect on all remaining components, whereas the disruption of one population of consumers would have less of an effect overall. The model that Grotzer calls “cyclic” appears to be that described in the systems dynamics field as a feedback loop (Kauffman, 1980; Meadows, 2008). Grotzer (1993) explores only reinforcing feedback (an increase in A causes an increase in B, causes an increase in A) and fails to acknowledge the complementary “balancing” feedback loops where an increase in A causes a

decrease in B, which reinforces stability in a system (Kauffman, 1980; Meadows, 2008).

To explain her findings, Grotzer (1993) proposed that linear and proximate causal patterns may be more accessible to our senses, leading us to overgeneralize all causality as linear and proximate, reinforced by selective observation (observing what we expect to observe). Grotzer

(1993) also reviewed early developmental research studies that suggested reasoning about complex relationships does not develop without multiple experiences and support in multiple content domains, an assertion supported today by multiple lines of research in cognitive science and education (c.f., Bransford, Brown, & Cocking, 2000). Grotzer and colleagues’ work supports more recent arguments that explicit teaching of underlying structures or patterns in the natural world as models is one of several strategies that can support children learning across scientific domains, discussed in more detail below (e.g., Lehrer & Schauble, 2000; Schwarz & White, 2005).


Limitations of Developmental Studies

Developmental studies, particularly those in children’s naïve biology, are typically concerned with the natural, unschooled unfolding of children’s ideas through physical maturation. As Hatano and Inagaki (1994) cautioned, children’s naïve biology has its limits, such as lack of factual knowledge about different organisms, lack of systematically organized biological categories, and lack of knowledge of physiological mechanisms or processes (causality), and biological concepts such as evolution. There may also be cultural differences in children’s biological reasoning based on language (Stavy & Wax, 1989), or cultural beliefs, such as the Buddhist teaching that all objects in the natural world have a mind or a purpose (Hatano, Siegler, Richards, Inagaki, Stavy & Wax, 1993).

As in the constructivist literature, a gap in the literature on children’s development of biological reasoning is the lack of studies conducted from a systems perspective. While topics such as “human body system” may be found in school curriculum, upper elementary children’s understanding of the structures and functions of their body are vague at best (Reiss, Tunnicliffe,

Andersen, Bartoszeck, et al., 2002; Rowlands, 2004). Additionally, the findings that children come to school with naive biological understanding and use their knowledge of their own bodies to construct analogical explanations for biological systems may present a challenge when they are introduced to ecology and must begin to consider the role of abiotic structures and functions in ecosystems. On the other hand, experiences structured to enable elementary children to consider how their own bodies use and transform energy in processes of digestion and respiration may help them make sense of the role of energy in living systems. Methods and arguments made by Grotzer in her original 1993 and subsequent studies were of the first wave of studies that sought to understand how children’s natural cognitive maturation interacts with instructional experiences. With the introduction of the methodology and philosophy of design studies (Brown, 1992; Collins, 1992), educational researchers have argued that learning is a complex interaction of natural development

69 and instructional opportunity, with the recognition that young children are quite capable of sophisticated reasoning about ideas and events much earlier than predicted by Piaget’s original studies (cf., Metz, 1995). Lehrer, Schauble and colleagues take the approach that instructional design must identify children’s naïve ideas not to change them but to build upon them through extended instruction towards the “big ideas” of a discipline (e.g. Lehrer & Schauble, 2000; 2006). Children's prior knowledge and experience in a domain, rather than their age, determine how they can engage in complex disciplinary ideas.

Phenomenographic Experiencing of Ecosystems

Phenomenography is an approach to qualitative research in the interpretivist tradition

(Marton, 1986) which seeks to identify the range of variation of meaning of an experience across individuals. A phenomenographic approach to research focuses on an individual’s interpretation of meaning from experience, and is therefore an important perspective to consider when exploring how children and adults experience ecosystems from the perspective of relationship and engagement with the natural world and their unique interpretation of embodied experience. The goal is to reveal patterns of interpretation and meaning which can describe the range and variation of human experiences and relationships to the natural world (Carlsson, 1999, 2002a, 2002b; Helldén, 1995,

2004). In the framework of situativity, phenomenographic studies provide insight into the worldview of the learner, which can inform the design of learning interventions. Ways of conceiving or perceiving are not constants within an individual but change continually as a result of our acting in particular contexts. By mapping and understanding the underlying patterns of variation in people’s experience, studies in the phenomenological tradition provide educators with a potential landscape of learners’ ideas with which to work. The studies in this section complement developmental and model-based research by providing a synoptic view of ways that teachers and children interpret the structures and functions of ecosystems, what Carlsson (1999) termed “ecological understanding.”


This understanding goes beyond scientific knowledge to include the insights, beliefs, and relationships with nature that people bring to the formal study of ecosystems.

Ecological insights

Carlsson (1999; 2002a, 200b) investigated the phenomenon of ecological understanding and patterns of understanding across 10 pre-service teacher educators in Sweden to identify variations in how people experience ecosystem structures and functions, based on the elements they discern as different and dimensions and attributes that are perceived as simultaneously present, whether consciously recognized. The question Carlsson (1999) addressed was: what particular aspects do participants discern when they are considering the function of an ecosystem? An insight is defined by Carlsson (2002a) as a “way of thinking about” key ecological functions. To develop insight, a person must discern dimensions of a phenomenon from its context while continuing to relate these to the whole. There must be variation in phenomena to enable discernment, with multiple variations or dimensions simultaneously perceived, if only tacitly. In other words, Carlsson took a holistic approach to studying understandings about ecosystems. The research of Carlsson revealed three insights related to ecological understanding that are necessary for a holistic understanding of ecosystems: the photosynthesis insight, the recycling insight, and the energy insight. All of these insights point to the idea of transformation as key to ecological understanding (Carlsson, 1999, 2002b).

She connected these three in a dialectical relationship with each of the others. As she put it, "the functional aspects between the living and the non-living elements tie the processes into a system of wholeness" (2002a, p. 683).

To elicit participants’ views of an ecosystem, Carlsson (1999) used two task-related interview contexts with ten pre-service science teachers in Sweden. In the first task, teachers were presented with an empty bowl and asked what they would need to build or create an ecosystem in the bowl to enable the system to survive for two years. Materials were available but hidden until a participant

71 mentioned them, at which time their explanation about the function of the material in the system was probed. Carlsson also intervened by challenging or questioning ideas; in one case she introduced a lid and “closed” the bowl permanently to bring focus to the survival of the living organisms. In the second task, participants were given a thought experiment: to design a spaceship and specify supplies needed to ensure survival of 100 humans for 6,000 years. This second task brought in the relationship of humanity with ecosystems. The analysis compared individual and group responses until four qualitatively different ways of thinking were identified.

In terms of photosynthesis, Carlsson (2002a) describes categories also found in previous studies: 1) the plant takes in some components and produces others without explanation of internal functions (expressed by half of her ten teachers); 2) plants are viewed as an essential component of all ecosystems due to their ability to transform sunlight energy, water, and carbon dioxide into carbohydrates. The remaining categories add complexity to the second: 3) plants also respire and are therefore self-sufficient organisms, and 4) photosynthesis creates order and resources within the plant. Carlsson found little variation across the context of the two tasks; she cautions that the two tasks were quite similar in their underlying structure.

In analyzing participants’ ideas about materials cycling and energy flow, Carlsson (2002b) again finds similar results, with most teachers expressing a level 1 “input-output” model in which ecosystem resources such as water, nutrients, or energy are “used up.” Individuals expressed similar levels of understanding for each insight across all tasks.

The insight of transformation

Carlsson (2002b) argued from her findings that achieving a systems-level view and understanding transformations is the key "shift" learners must make to become ecologically literate.

Transformation appears to be the conceptual “gatekeeper” to more complex forms of understanding ecosystems. Transformation of energy and materials explains both structure and

72 function—the interrelationships between organisms and their abiotic environment. Carlsson theorizes that the shift to transformational understanding is a shift of paradigm, that once achieved is now a powerful new way of “seeing” and explaining. On the other hand, she conjectured that “old ways of thinking” may not disappear but remain useful for different contexts, a subject for further research.

The Water Cycle as Perceivable Model for Non-Perceptual Materials Cycles

A colleague of Carlsson’s, Helldén, a practicing biology teacher, noticed students' difficulties in explaining ecological processes. He designed a study of 25 pupils from age 8 to 15, using a modified clinical interview and models of plants cultivated in closed (sealed) transparent boxes

(Helldén, 1995). Hellden constructed the sealed plant boxes to challenge students' ideas about the conditions needed for growth and how matter cannot simply “go away.” Helldén found students’ thinking quite similar to Carlsson’s teachers: most held a limited input-output model of plant biology and a model of natural resources being “used up.” From his findings Helldén identified three processes that he claims students need to understand: 1) All organisms maintain equilibrium or homeostasis through the exchange of energy and matter; 2) Several ecological processes require water; the water cycle and phase changes of water, and 3) the products of decomposition are carbon dioxide and water. These are all processes of gas exchange, which cannot be observed directly, so it is difficult for students to accept their materiality (Andersson, 1991). Thus key transformations of matter, including photosynthesis, become difficult to understand until children learn that all gasses have mass, and are integral components of ecological cycles of growth and decay. Hellden (1995) concluded that the closed box models seemed “useful” for challenging children’s ideas, and that introducing the water cycle at an early age could serve as a useful model for later “cyclic’ thinking.

He cited Boshhuizen and Brinkman (1991), who argue that the water cycle is easier to understand because water always remains as the same molecule, whereas in photosynthesis, digestion, and

73 respiration, molecules are broken down and recombined to form different kinds.

Idiosyncratic Trajectories of Ecological Insight

In a longitudinal study of these same students, Helldén (2003, 2004) found that many students held onto initial ideas. He identified individual themes in the ways that each student explained the function of the ecosystem. Further, students often referred back to concrete experiences they had in childhood, suggesting a contextual and episodic form of reasoning. While many students expanded the complexity of their explanations or developed more scientifically

“correct” explanations as they aged, they also expressed earlier ideas as part of their “repertoire” of explanations. The development of individual students’ “trajectories” of ecological understanding appear “idiosyncratic” (2004, p. 73) despite continuing in the same two schools and experiencing the same curriculum. Thus each person’s individual interests, perceptions, discernments, and life experiences shaped their interaction with the domain of ecological understanding.

Experiencing Ecosystems from “The Bottom Up”

In further studies, Magntorn and Helldén (2005, 2007) moved their teaching and research away from pure phenomenology and into instructional design in the field with pre-service teachers, high school students, and elementary students. I discuss their study with elementary students here because it is a natural extension of their original work and most pertinent to this study. Using a

“bottom up” approach with elementary students (age 10-11) suggested by Slingsby and Barker

(1998), Magntorn and Helldén (2007) began with a focus on a single organism (a fresh water shrimp) and focused on its population’s autecology (relationships with the other living and nonliving components of the ecosystem), taxonomy in terms of adaptations, synecology (relationships of the organism with other organisms in the community, and ending up at a systemic view of the relation of biotic and abiotic components in an ecosystem, especially matter cycling and energy flow. Data were collected via a series of interviews spaced throughout a field-based instructional sequence.


During the study students also constructed and observed sealed aquatic ecosystems based on their developing knowledge of the ecosystem.

Of the 23 students in the study, their reasoning shifted from pre-structural/uni-structural

(vague or one-way connections such as “fish need water”) to 15 “reading” the river on a relational level (linking organisms to ecosystem function) and six reading the river on an extended, abstract level (explicitly discussing the role of materials cycling and energy flow, populations and life cycles, and variation in biodiversity). However, roughly one-third of the students continued to use unistructural explanations during the post-experience interviews, which suggests there are more factors in play such as student interest and motivation, memory and recall, or artifacts of the research design.

The researchers noted that students continued to maintain qualitatively different ways of

“reading” the river ecosystem based on their individual interests and ways of seeing. In the study they contrast two students, one who continued to focus on morphological features of individual organisms while the other focused more on ecological relationships and made more synecological and systems-based links, despite both having approximately the same domain-based knowledge and similar experiences. What also stand out is that species knowledge, experience with decomposition processes, and models of relations based on trophic levels rather than focusing on specific food webs appear to be fruitful or “linchpin” concepts for students at both the elementary and secondary levels.

Additionally, the authors remark that many students continually referred back to their experiences in the field (experienced context) while constructing their explanations. This supports a situated explanation of cognition and the argument to involve students more in fieldwork (e.g.,

Bowen & Roth, 2007; Dillon, Rickinson, Tearney, Morris, Choi, Sanders & Benefield, 2006;

Sukhontapatipak & Srikosamatara, 2012). Further, the approach of starting “from the bottom up”


(Magntorn & Helldén, 2007) with the autecology of one organism and gradually expanding outward to its multiple interactions with the system enabled most students to progress from a focus on parts or simple links (e.g., fish need water) to at least a level of multistructural relationships with the life cycle of several organisms linked, and two thirds of students toward a relational or systemic view of an ecosystem that addresses the two major organizing principles of energy flow and materials cycles.

To summarize, both Carlsson’s and Helldén’s studies identify common patterns of experiencing ecosystems and also identified key insights that lead to more complete and scientific understanding of how ecosystems function. The insights of transformation (Carlsson, 2002b) and the role of gas exchange and transformation (Helldén, 1995) both strongly support the argument that micro-level processes and interactions between the abiotic and biotic components of ecosystems are simultaneously essential for making sense of ecosystem function and difficult to grasp. It is interesting that neither researcher included the process of respiration as a key insight for completing the cycle of carbon exchange.

Limitations of phenomenological studies

While phenomenological studies contribute to understanding the range and variation of people’s ecological reasoning, the purpose of phenomenology is not to delve into why such variation exists. In Carlsson’s studies with teachers, it is clear that some teachers had very little prior knowledge of or experience with ecosystem form or function. Several had limited science course work. In Magntorn and Helldén’s (2007) study with upper elementary students, just over one-third

(8 of 23) did not achieve a level of relational understanding (recognizing the interrelationships between factors or organisms in the system)—although the researchers acknowledge that neither the instruction nor the interview prompted students to do so explicitly.

The use of “closed systems” models in these studies do appear to provide an affordance for participants to reconsider their naïve models of materials cycles that matter can simply “disappear”

76 or be “used up.” In Carlsson’s original study (1999), the participants who were presented with the closed model did reconsider their ideas to an extent in terms of plants as self sufficient organisms.

However, many students from all studies still failed to identify connections between the exchange of gasses between animals and plants, and the micro processes of photosynthesis and respiration that drive the behavior of the system over time. The present study addresses some of the limitations of phenomenographic research with its example of how a student brings his phenomenological perspective to a range of academic tasks, with the goal of developing insight into how and why he uses and possibly amends his phenomenological perspective as a result of interacting with the curriculum.

Teaching and Learning about Ecosystems:

Structures, Functions, Behaviors, and Models

The studies discussed in this final section approach the teaching and learning of ecosystems holistically, that is, as integral systems and nested subsystems. Studies in this paradigm assume that the cross-cutting concept of understanding systems is an essential dimension of ecological understanding. In everyday language, thinking about natural systems requires asking “what goes with what?” (Meadows, 2008, Orr, 1992), which is recognition of the components of ecosystems in the world or in the model. More deeply, thinking about systems requires focusing not just the components (the form of the system), but their interrelationships, functions, and dynamics as a whole (von Bertalanffy, 1968; H.T. Odum, 1983). This more recent body of research suggests that given enough time to develop expertise in a content area, and instructional scaffolds that support extended inquiry and sense making, children in elementary grades are able to engage in modeling practices to make sense of biological and ecological processes (Ero-Tolliver, Lucas & Schauble,

2013; Lehrer & Schauble, 2004; Lehrer, Carpenter, Schauble & Putz, 2000; Lehrer, Lucas, &

Schauble, 2008; Manz, 2012). This literature extends research and instruction beyond using physical

77 models of ecosystems as observational tools towards engaging students in what Hestenes (1992) termed “the modeling game”—the interactive development of domain general and domain specific approaches to understand phenomena through “model based reasoning” (NRC 2012, NGSS Lead

States 2013).

The Practice of Modeling

With the shift to a “practice” view of science education, recent research and science policy based on this research emphasizes the importance of constructing knowledge in science actively through a dialectic between data and theory, via the process of model-based reasoning (NRC, 2012,

NGSS Lead States, 2013). Lehrer & Schauble (2006b) noted that until recently, the idea of “model” in school science typically represented a thing or a product, as opposed to a result of the process of modeling. In contrast, Lehrer and Schauble (2005, 2006a, 2006b) argued that modeling is a form of disciplinary argument that requires extended practice on the part of students and teachers to develop coherent conceptual and epistemic structures. Modeling as a process entails:

…[T]he construction and test of representations that serve as analogues to systems in the real world. These representations can be of many forms, including physical models, computer programs, mathematical equations, or propositions. Objects and relations in the model are interpreted as representing theoretically important objects and relations in the represented world. (Lehrer & Schauble, 2006b, p. 177).

From this perspective, modeling is a form of disciplinary argument that requires extended practice on the part of students and teachers to develop coherent conceptual and epistemic structures.

A key pursuit in Lehrer and Schauble’s research program is to identify which types of models and science themes provide “easy entry” for young children and can be deepened and extended as challenges for older students (2006a, 2006b). They proposed thinking of instructional design as an

78 interaction among variations in learners and richly structured tasks:

Designing for education must encourage emergence and variability or else risk pruning the potential for development to sanctioned pathways. Faced with such complexity, educators can choose the path for students and use teaching assistance primarily to minimize straying from the predetermined route. Or instead, one can foster and encourage variability in student thinking and then capitalize on the local opportunities that emerge from it. In that case, the design problem is to craft situations and tasks that are most likely to produce forms of variability that are rich with instructional potential…If one reconceives of variability not as error or noise but as grist for development (Siegler, 1996), then documenting and accounting for contingency become an essential part of the research enterprise (p. 191).

In Lehrer & Schauble’s research program, they attempt to address three goals of science education grounded in model-based reasoning:

1) understanding of particular scientific ideas through their models,

2) build a repertoire of models that are usable across different situations, and

3) understanding how processes of modeling contribute to construction of scientific knowledge (Lehrer & Schauble 2006b, p. 177)

For example, an a review of their work with students on modeling tasks, Lehrer and

Schauble (2006a) argued that engaging in modeling practices in school science settings entails two strategies of representational practice. They advocated providing students with the opportunity to develop their own idiosyncratic forms of representational practice within a classroom culture that continually shares, negotiates, and critiques the meaning of those representations. They also advocated introducing students to mathematical modeling practices so that students shift their attunement from specific instances to general patterns, and from a focus on the concrete attributes

79 of individual organisms to the patterns, variations and distributions of populations of organisms. For example, in a 2004 study, Lehrer and Schauble built on their earlier work with a group of fifth grade students on the statistical concept of distribution (Petrosino, Lehrer, & Schauble, 2003) to explore the concept of natural distribution of growth over time in a population of Wisconsin Fast Plants™.

This design study extended over 28 sessions and approximately 45 hours, about one-third more than the time allotted to the present study. The academic task involved students designing displays of the heights of a collection of plants at a single day of growth (Day 19). These students settled on plant height as the most direct measure of growth, although there were debates about whether the height of a plant included the roots, or how to count multiple branches. The teacher-researchers scaffolded student discussion about the usefulness of different representations by having students work in small groups to create inscriptions, then trade with another group and provide feedback. Over time this led to discussions of how a particular data display highlights some features of data and backgrounds others. Through this extended investigation, students developed multiple ways to represent the data and also developed what diSessa, Hammer, Sherin & Kolpakowski (1991) refer to as “meta-representational competence.” Students were continually encouraged to reference their representations or inscriptions back to the population of actual plants. This study extends the argument made by Lehrer, Schauble, Carpenter and Penner (2000) that “tools, talk, notation, and modes and means of argumentation…are effective means for fixing and composing the process of design in replicable plans and outcomes” (p. 326); the development of inscriptions are interrelated with conceptual outcomes.

Inscriptional models

From the beginning of their work, Lehrer, Schauble and colleagues positioned mathematics

“as a tool that both describes the world and serves as a resource for meaning-making (Lehrer,

Schauble, Strom & Pligge 2001). The authors argue that mathematical understanding that underlies

80 descriptions of distribution, variation, generalization, and other structures of data are necessary to ground students’ understanding of “big ideas” in science such as growth or diversity. The central process of inventing, revising, and contesting models relies on a wide range of discursive and mathematical practices—using multiple forms of representation to explore aspects of the world and move toward “mathematizing” natural phenomena by quantifying, visualizing geometrically, and using graphs. (see also Lehrer & Schauble, 2005; Latour, 1990).

The emphasis on mathematics is grounded in research that demonstrates younger students are able to think mathematically by engaging in the practices of examining geometry (Lehrer &

Pritchard, 2002), measurement (Lehrer & Schauble, 2000), and data (Lehrer & Schauble, 2000c,

2005). Lehrer & Schauble argued that students need access to these mathematical resources in the context of “big ideas” in science (e.g., form and function) in order to construct explanations based in data and models (2006b). Underlying their approach is the assumption that with children, the most effective approach is to begin with models that resemble observable phenomena, for the reason argued above, that children need scaffolding to successfully connect and remember the connections between the model and the phenomenon the model represents (Brown, 1990; Lehrer &

Schauble, 2000c).

With careful instruction, first grade students were able to shift their modeling of an elbow from its form (“looks like” an elbow) toward its function (“works like”) (Penner, Giles, Lehrer &

Schauble, 1997). By third grade, the same group of students was able to use mathematics to describe the function of the elbow tendon as a lever able to move a load (Penner, Lehrer & Schauble, 1998).

The researchers followed the children’s focus on the importance of plant height under different conditions to establish standards of measurement and representation (Lehrer, Carpenter, Schauble &

Putz, 2000). Examination of the change of height over time led children to ask if one plant grew faster than others, and how much faster. By comparing narcissi and amaryllis, the children also

81 recognized that different plant species showed different rates of growth but eventually growth

“levels off.” In a third-grade study, children investigated changes in a population of Wisconsin Fast

Plants ™, using pressed plants in series to record changes over time, correlate plant height to canopy “width” and three-dimensional cylinders to investigate plant volume (Lehrer, Schauble,

Carpenter & Penner, 2000). The pressed plants drew students’ attention to possible relationships between the growth of roots and shoots, which led them to notice how the growth of roots first predominates but is then followed and exceeded by the growth of the plant, yet both parts of the plant system showed the same general pattern (an “S” shaped logistic curve).

The increasing sophistication in students mathematical and visual modeling practices led students to important biological ideas about form and function, growth, and patterns of change over time, which they could explain and defend using data from their measurements (Lehrer & Schauble,

2005, 2006b). By fifth grade, the students were familiar enough with variation and distributions in growth patterns of a population of plants to relate variables such as amount of fertilizer and light with plant height and number of seeds and seed pods (Lehrer & Schauble, 2004).

Lehrer and Schauble (2000c, 2003, 2005) described “characteristic shifts in children’s understanding of modeling” across the elementary years. They found—

1) Children began by emphasizing literal depictions, but could shift to use and understand more “symbolic and mathematically powerful” forms of representation as they gained experience;

2) Children employed diverse forms of representation and use of mathematical resources; these inscriptions worked in dialectic with children’s ideas to bootstrap more sophisticated thinking;

3) Children came to understand biological growth as dynamic rather than linear;

4) Repeatedly using, creating, and revising model-based representations helped students co-

82 develop understanding of the uses and purposes of scientific or mathematical inscriptions as well as the content-grounded arguments they support.

Models as a form of explanation

A key need is to help teachers and students shift activity in science classrooms away from surface observations and descriptions towards constructing scientific explanations (Driver, Asoko,

Leach, Mortimer & Scott, 1994; McNeill, Lizotte, Krajcik, & Marx 2006). In the philosophy of science, a scientific explanation goes beyond description to theorize about unseen entities, laws, and principles that provide causal explanation for phenomena—explanations answer “why” (Braaten &

Windshitl, 2011). These authors argued that a more robust conception of explanation is needed in the science education community so that teachers and children understand what “counts” as an explanation, what distinguishes explanation from description, and how might students’ explanations be evaluated (p. 651). Model-based reasoning serves as one legitimate form of knowledge construction, explanation, and evaluation in ecology and in learning about complex and dynamic ecosystems (Grotzer & Basca, 2003; Lehrer & Schauble, 2012; Perkins & Grotzer, 2005). Yet this practice is new and challenging for both teachers and students (Windschitl, Thompson & Braaten,

2008). It is one thing to advocate for a model-based reasoning approach as a design study with stable classrooms, ongoing teacher support, and a phalanx of researchers to collect and analyze data. It is quite another for a single elementary teacher to change her worldview and practice in teaching science while also challenging the beliefs and habits of her students (Windschitl & Thompson, 2006;

Windshitl, Thompson & Braaten, 2008).

Scaffolding modeling practices

Findings from these studies are not surprising. Children were faced with the tasks of learning about two domains simultaneously: a scientific phenomenon and the practices of pictorial, diagrammatic, and mathematical representation of that phenomenon. A case from Lehrer and


Schauble (2002) echoes findings from a study done by myself and several colleagues on children’s interpretation of line graphs of solar house models heating and cooling (Doubler, Settlage &

Roberts, 1992). Lehrer and Schauble discovered while working with a group of 1st graders on the growth of plants that the children were able, with scaffolding, to construct a bar graph series of plant height changing over time. However, the children at first insisted on each strip of the bar graph being colored green like a plant stem, and topped each bar with a conceptual drawing of a flower. In our study, an assessment of children’s “telling the story” of a line graph of temperature change in a solar house model showed that children ranged in their ability or willingness interpret the line as a representational object vs. a model of a phenomenon. About one quarter focused their

“reading” solely on the shape of the line (“it goes up slowly, then fast, but then goes down again”) , half described the change in the line in terms of numbers or temperature (“it starts at 15, goes slowly up to 17, then slowly back down to 16”), and about one quarter were able to describe the change in the line as a story of how the air temperature changed in the model house quantified by data from the graph).

These studies suggest students need scaffolding and opportunities to construct conceptual links between their experiences and observations of phenomena and the more powerful but less contextualized inscriptions used in mathematical and scientific modeling. Lehrer & Schauble (2006b) cite Hestenes (1992) and Olson (1994) to argue that representational tools not only communicate but actual shape thinking. This provides further rationale for ensuring children develop familiarity with multiple forms of representations and criteria for when/how to use them.

Another important understanding children need to develop is that models constructed in science necessarily simplify the world and literally focus the reader’s eye on what the scientist believes to “count” as explanation and evidence. This can be seen in the history and sociology of science as well: Newton’s (1704) diagram of a beam of light being seen through a prism includes the

84 objects of prism and eye, as well as conceptual lines and angles demonstrating the path and refraction of light. Latour (1990) describe how the “facts” of science are constructed through

“cascades” of inscriptions, leading to “the transformation of rats and chemicals into paper” (p.22).

Thus Latour and other sociologists of science frame the notion of science as a worldview: “how a culture sees the world and makes it visible” (Latour, 1990, p. 30). To acquire the discourse practices of science children must learn to “see” beyond the visible observations to the underlying patterns and principles (Eberbach & Crowley, 2009). They must also learn how inscriptions and models work to support and “conscript” others into accepting a particular explanation or argument about a phenomenon (Latour, 1990). Latour argued that the inscription itself cannot “carry the burden of explaining the power of science” by itself…The phenomenon we are tackling is not inscription per se, but the cascades of ever simplified inscriptions that allow harder facts to be produced at greater cost” (p. 41, original italics). In actual scientific training, this often takes place in the field alongside experienced scientists, as Goodwin (1994) demonstrated through his study of how soil scientists learn to “match” the physical soil to a soil color classification on a Munsell chart. The ecologist does not “see” a food web but constructs a model of relationships among organisms in an ecosystem to explain how populations and communities of organisms self-organize to maximize the use of available energy. Simply presenting a visual diagram of a food web to an elementary student does not help the student to “see” what the ecologist sees (Eberbach & Crowley, 2009). Thus children interpret food web diagrams in terms of their previous knowledge and experiences, as “the bunny eats the grass, then the snake eats the bunny.”

In spite of their robust findings on how children learn to engage in model-based reasoning,

Lehrer and Schauble (2006b) outlined the many challenges inherent in conducting design research of this scope. Such studies require interdisciplinary teams with a high degree of communication and collaboration, particularly between researchers, teachers, and school administrators. Replication of

85 an intervention over multiple grade levels and classrooms takes years, money, graduate students, and requires constant flexibility to fit the realities of a particular group of students, teachers, and classroom contexts. Even in 2006, they describe the changing political climate of schooling, instability of leadership, turnover of staff, and student mobility. Their assessment of school climate is blunt:

The legitimate agendas of schools often inadvertently put them at cross-purposes to the goals of the research…within the past several years we have found the politics of education to be especially disruptive to any agenda that includes systematic capacity building (2006b, p. 189).


In this section I bring together key ideas from the empirical research, elaborating on how the present study builds on and extends existing research across three themes: 1) the power of microcosms as model systems; 2) extending the limits of physical models with representations and inscriptional practices, and 3) translating large-scale proof-of-concept design studies to implementation in “everyday” classroom contexts.

The Power of Microcosms

Teaching ecosystems through the use of physical models (microcosms) is one approach that has shown promise from the college level to first grade. Tamponnet and Savage (1994) argued that developing and studying closed ecosystems (containing associations of organisms that form materials cycles, but open to energy exchange from the outside) can serve as tools to determine the combinations necessary to attain the long-term succession and balance needed to support long-term space exploration, or to learn further about the functioning of terrestrial systems (p. 167). The

Biosphere2 Center at the University of Arizona was originally designed as such a system, which led to important new scientific understandings of how large-scale microcosms function (Tubiello,


Druitt, Marino, 1999) as well as how such model systems alter the function and biodiversity of wild biomes (e.g., Leigh, Burgess, Marino & Wei, 1999).

Two studies illustrate the potential of using microcosms in research and instruction. Britta

Carlsson investigated Swedish student teachers’ qualitative variation of ideas about the structures and functions of an ecosystem through two interview-facilitated tasks of designing ecosystem models (1999; 2002a). Carlsson’s use of a task of designing a microcosm, especially the materially accessible Eco bowl, served to elicit participants’ ideas and questions, to which she could respond with further prompts or clarifications. Carlsson’s study suggested the Eco Bowl design task could be a fruitful strategy for research with students, encouraging them to express their mental models and test ideas through the design of a physical model.

Helldén’s (1995) study followed 26 Swedish students’ ideas about the transformation of matter in ecosystems from age 8 to 15 using a combination of interviews and a closed microcosm in which plants were grown and leaf litter changes studied. To challenge students’ naïve explanations that dead materials simply “disappear,” or that plants increase mass (grow) due to continuous uptake of water from the soil, Helldén cultivated plants in sealed transparent boxes to focus students on the conditions for growth and to provide visible limits for decomposition and materials cycles. Students noted the transformation of water in the sealed system, which Helldén claimed became a prototype for thinking about “cycles” that they incorporated to explain both decomposition and gas exchange by age 15.

The methods and findings from these studies suggest that engaging students in modeling microcosms challenges students to consider what are the necessary structures and proportions of biotic and abiotic materials needed to create a functioning ecosystem. Designing microcosms with feedback from peers challenges students to make their thinking visible to themselves and others—to express their tacit and personal mental models so that others can evaluate and give feedback on their

87 ideas. Carlsson’s (1999) approach, allowing her participants to decide what should go into the Eco

Bowl to constitute a functional ecosystem, became the basis of this study’s design task. Helldén’s

(1995) success in using his closed microcosm to focus students on water’s phase changes and the visible processes of materials and decomposition cycles also suggested that such models could be used to challenge students’ naïve explanations of materials “disappearing” and to focus their perception on the theoretical need to identify boundaries in the analysis of systems by instantiating these as physical boundaries in the model system. A limitation in both these studies is that neither

Carlsson nor Helldén made explicit the ways that their microcosms functioned as simplified models of the Earth’s materially-closed but energetically-open system. To emphasize this epistemological connection, in the current study, the teacher and I made explicit the function of the BioBottle as a model of Earth’s system, using the intermediate model of Biosphere 2 as an exemplar and a story frame.

Lehrer's and Schauble’s extensive body of design-based research studies suggests modeling practice at the elementary level is most successful when instruction begins with physical models that maintain resemblance to the phenomenon under investigation (Lehrer & Schauble, 2005). This approach builds on children’s limited understanding of models as physical copies of the world (e.g.,

Grosslight, Unger, Jay & Smith, 1991). Shifting the instructional emphasis from structures

(components and relationships) to functions (causal relations and the resulting changes over time) has been shown to elicit more detailed explanations in related studies of children designing a functioning elbow (Penner, Giles, Lehrer and Schauble, 1997), exploring the biomechanical functions of levers (Penner, Lehrer, Schauble, & Putz, 1998), reasoning about gears (Lehrer &

Schauble, 1998), explaining natural variation in populations and exploring growth patterns in organisms (Lehrer & Schauble, 2004), designing artificial lungs (Hmelo, Holton & Kolodner, 2000), and aquarium systems (Hmelo-Silver, Marathe, & Liu, 2007; Lehrer, Lucas & Schauble, 2008). The

88 model-design task used in the current study extends the focus of modeling studies into the domain of ecology by challenging students to design a materially-closed model that functioned like a garden ecosystem. Two other modeling studies related to ecosystems focused on discrete processes such as decomposition with first-grade students (Ero-Tolliver, Lucas & Schauble, 2013) and reproductive success with third graders (Manz, 2012). These are important precursor studies for developing students’ attunement to key ecological ideas such as materials cycles and building toward more complex ecological understandings such as evolution (Lehrer & Schauble, 2012), but leave open the question of how and at what ages students are capable of considering an ecosystem as a functional and complex whole.

The limitations of students’ domain-based knowledge, which were shown to have an effect in some studies (e.g., Hmelo-Silver & Pfeffer, 2003, Hmelo-Silver, Marathe, & Liu) were minimized in this study by introducing the project at the end of the school year after students had completed two related science kits, Ecosystems (BSCS, 1999) and Earth Materials (FOSS, 2005), and were concurrently engaged in a unit on Water (FOSS, 2003). Given that the teacher had approximately seven years’ experience in systems dynamics thinking (which includes both inscriptional and computer-based modeling practices), and that the student in the study had worked with this teacher for three years, this unique case provided an opportunity to contribute an additional perspective in the research on modeling practices and ecological thinking of elementary students.

Extending Physical Models with Inscriptional Practices

Deciding which aspects of the natural world to include or highlight in their model, and which to ignore, is a key metacognitive process in modeling (Schwarz & White, 2005) and inscriptions (diSessa, 2004) that develops through production, evaluation, and revision (Lehrer &

Schauble, 2002; Lehrer, Schauble, Carpenter & Penner, 2000). For example, Lehrer, Schauble,

Carpenter and Penner (2000) describe engaging third-graders in an iterative series of investigations

89 in which students grew populations of Wisconsin Fast Plants™ and were challenged by the researchers to use graphical representations of variations in plant height to decide what was a

“typical” height. Researchers also had students graph change in height through the plants’ life cycle, which with careful questioning by researchers and teachers led students to identify the characteristic

“S” shape growth curve. This in turn led to questions and investigations related to why plants and humans demonstrated both a growth spurt and a leveling off of growth at maturity.

While Lehrer and Schauble advocate for students to develop and evaluate their own forms of model-based representations, the case described in Lehrer, Schauble, Carpenter and Penner

(2000) suggests that the inscriptional and modeling decisions were primarily introduced and orchestrated by the researchers with specific mathematical goals in mind. They acknowledge that the work of constructing inscriptions to reveal underlying patterns and relationships in nature is itself a complex practice that needs careful scaffolding by teachers who clearly know which patterns and processes they expect students to make sense of. One aim of this current study was to explore what kinds of inscriptions students in this urban classroom, working with a teacher who herself was developing systems dynamics thinking, generated to express and extend their ideas about ecosystem’s structures and functions. The current study analyzes the inscriptional practices of one student across various tasks in which he expressed his ideas about ecosystem structures and functions using a variety of representations (drawings, diagrams, concept maps, and related visual models). Those related to the design and observation of the physical microcosm offer evidence for which aspects of the system are perceived and considered salient to the student, albeit in a complex interaction of his ecological knowledge, emergent sense-making, and representational capabilities.

Challenges of Translating Design-based Research for Everyday Classrooms

As Lehrer and Schauble (2006) acknowledged, there are many unique features inherent in conducting design research that are not typically found in most educational settings. Such studies

90 require interdisciplinary teams with a high degree of communication and collaboration. Replication of an intervention over multiple grade levels and classrooms takes years, money, graduate students, and institutional support. For example, the plant diversity and growth modeling studies of Lehrer &

Schauble (2003, 2004) entailed working with an entire class of 23 students, with data collected by up to three researchers on whole-class and small-group interactions, as well as follow-up individual interviews audiotaped and transcribed. They note that appropriate contexts for such design-based studies include a need for a large number of measures of multiple attributes, time and opportunity for exploring leads or revisiting problematic areas, are detailed enough to track change in learning over time, and entail inquiry that can be conducted within the limits of school time and materials

(2004, p. 670). While the school itself could be considered “typical” in terms of student diversity, teacher experience, and financing, the school and several teachers also had an extended history of collaboration with university researchers, and leadership commitment to the professional development and autonomy of teachers as well as to promoting innovation. Further, the researchers also designed an extended series of professional development experiences for teachers over the course of several summers and during the school year (Lehrer & Schauble, 2000).

With the introduction of “No Child Left Behind” and the increased emphasis on high-stakes testing of mathematical computation and linguistic fluency, such extended design-studies became more difficult to implement (Lehrer & Schauble, 2006). At present, most elementary teachers and students, particularly in underserved urban classrooms, do not have access to the conditions that supported Lehrer and Schauble’s earlier design studies, but that should not be accepted as an excuse not to search for more scaled-down or different approaches that reflect the underlying principles of successful design-based modeling research to date. To that end, this study attempted to translate the principles of their large-scale design research by exploring how the use of ecological microcosms and other forms of modeling in a more “typical” classroom might contribute to students’ ecological

91 literacy.

A related characteristic of most of the studies cited above relates to the unit of analysis—the bounds of the studies examine impact of instruction at the level of the classroom. While individual group vignettes highlight different representational choices (Lehrer & Schauble, 2004) or analyze group abilities to manage the complexities of inquiry related to an aquatic ecosystem (Lehrer,

Schauble & Lucas 2008), data on the impact of instruction or the nature of student modeling practice are analyzed at the classroom level. This level of analysis makes a compelling case for the educational value and efficacy of approaching science instruction through the practices of modeling.

Yet at the level of day-to-day teaching and assessment practice, the focus of performance outcomes remains on the individual student, and the challenge to teachers is how to identify and build upon individual students’ ideas and practices—to make informed instructional decisions that take into account individual learners’ potentials. Design-studies suggest that model-based instruction can be a powerful force for learning, but cannot address questions of how such model-based practices are taken up, interpreted, and used by diverse students. By delving into the complexities of the day-today interaction of modeling practices and sense making of one student across various contexts, this study enriches the existing literature at a more detailed level of scale. A focus on one student’s navigation of the complexity of the content domain and the multiple structures and functions embodied in the BioBottle model highlights the possible points of connection or disjunction in making sense of the multiple complexities of an ecosystem, and of the modeling practices that bring that student’s ideas about ecosystems into being.



Study Design

Through this interpretive case study (Merriam, 1998), I investigated how an elementary-age child, Jorge interacted with and made sense of ecosystem form and function through the practice of designing, building, and observing a “BioBottle” model of a garden. The study took place in the ongoing classroom work in “Room 5,” whose participants were children in a mixed grades 3-5 (ages

9-12) classroom who had completed two science units on ecology and earth materials, and simultaneously were doing work in a FOSS science unit on the water cycle. The modeling task, which I designed in cooperation with the lead teacher, served as a “thought revealing” research activity (Kelly & Lesh 2000) as well as an academic task (Doyle, 1983) in which students had opportunity to make sense of the components, structures, functions, and behavior of model ecosystems over 12 weeks.

As argued in my theoretical framework, children's understandings related to science content knowledge can be seen as a developing and complex system that children construct and extend in practice with more knowledgeable others in the context of meaningful activity. Given that the phenomenon of understanding is multidimensional, then evidence of understanding must be multidimensional as well. I propose that children's understandings can best be “seen” through learners’ actions, or performances (Wiggins & McTighe, 1998), and their reflection on those performances with the help of “more knowledgeable others” (Vygotsky, 1987). Children’s learning emerges through the complex interactions of task, talk, and tools as “all functions are incorporated into a new structure, form a new synthesis, become parts of a complex whole; the laws governing this whole also determine the destiny of each individual part” (Vygotsky, 1986, p. 108). The phenomenon of understanding unfolds at multiple levels of scale, as individual reasoning, social

93 discourse, and interactions of learners with the physical world of objects and natural environments

(Greeno, 2009). Children's writing, drawing, and purposeful action across scales and across situations served as the data to be collected and analyzed.

As defined by Stake (1995), this study investigated a best case bounded system of a class of students who had the opportunity to develop understanding of ecological concepts through both an informal, field-based environmental learning program (Earthkeepers), and formal classroom curriculum (BSCS Ecosystems, FOSS Earth Materials, and FOSS Water with district adaptations).

Students were also participants in a community of practice as systems thinkers and ecological thinkers, with a teacher who viewed herself as both, demonstrated good conceptual understanding, and held a personal and public commitment to making conscious ecological decisions in her daily life. This study was designed to extend research on children's ability to use physical models to reason about structures and functions of ecosystems. The research design and analysis follows a systemic view, investigating the interaction of multiple variables in a dynamic environment (Salomon &

Perkins 1998; Greeno, 2009), “zooming in” and “zooming out” to describe the practices of an individual student and some of his collaborative work with another student to document and explain the changes they observed over time in their BioBottle.

Following the conceptual frame, the study design employed the research tradition of fieldbased, naturalistic, and interpretive research (Bogden & Biklen, 1992; Creswell, 2003). Research strategies of pre- and ongoing-individual assessments, participant observation, and field notes in the form of audio and video recordings enabled the collection of a wide range of data at the scale of whole class, small group, pairs, and individuals.

Unit of Analysis

A key benefit of drawing on situated cognition as a lens is that it allows researchers the choice of when and how to study learning processes at the level of the individual, small group,

94 whole class, or community of practice. All are systems in which learning emerges through interaction, that function at different levels of scale. A useful analogy proposed by Salomon and

Perkins (1998) is to consider conceptions of individual vs. social learning as

…[T]wo levels of analysis, each of which sometimes neglects the other. One can make an analogy with two perspectives on the spread of a flu: cell biology and epidemiology…Clearly the two complement each other: Subverted cellular mechanisms figure in the invasion of individual cells by viruses, but the viruses have to arrive at individual cells to infect them. Although each process can be understood in its own right, understanding the interplay yields a richer and conceptually more satisfying picture” (1998, p. 2).

This use of situativity theory aligns with Greeno and colleagues’ (1998) proposal that situativity theory contributes to thinking about design beyond the level of instructional methods toward a larger concept of designing a learning environment (Greeno, 2009). Roth’s (2001) metaphor of

“zooming,” borrowed from filmmaking and photography, frames for the researcher the question of where to focus and how far to zoom (p. 31). The patterns and relationships will be different based on the chosen scale of time or space.

Following Wilson and Myers (2000), I use the theory of situativity as a framework for

description rather than prescription. In their terms, the questions are: “Is the learning environment successful in accomplishing its learning goals? How do the various participants, tools, and objects interact together? What meanings are constructed? How do the interactions and meanings help or hinder desired learning?” (pp. 20-21, ms). This conceptual frame respects that learning situations and communities of practice emerge from the diversity of participants as well as the contingency of interactions:

[A]uthentic communities of practice are not so much designed, but rather emerge

95 within existing environments and constraints. They fill ecological niches where certain opportunities open up, based upon the environment, people, tools, organizational structure and power dynamics, etc. How can we talk about designing authentic learning environments when so much of what goes into a learning environment is pre-determined by constraints, or emerges based on the participants themselves?” (Wilson & Myers, 2000, pp. 21-22, ms).

Thus new models of design for science learning environments must take into account this tension and acknowledge that “what works” (or doesn’t) in one situation of educational practice may not in others. For this study, the unit of analysis was the interaction of Jorge, a fifth-grade student, with multiple academic tasks. Typically, a lesson and a task were the same unit until the class began designing their BioBottle models. The trajectory of the BioBottle process of design-observe-explain was examined as an extended task. Likewise, student assessments were also considered a task, because each resulted in a written (or representational) work product.



The study involved a mixed grade 3-5 class at an urban elementary magnet school in a large city in the Southwestern United States. This school and this class presented a unique opportunity to investigate how student understanding of ecological concepts emerges in a setting which envisions children as part of a community of learners. The school curriculum program emphasizes literacy and inquiry learning across all disciplines, and integrates the arts through the district’s “Opening Minds

Through the Arts (OMA)” curriculum in collaboration with community artists. Located in the heart of the valley and city, the school brings together students of diverse ages and abilities in a “total communication” environment in which everyone uses both sign language and English (School brochure, 2010).


The teacher, Ellie (a pseudonym), and the principal committed to working with a local environmental learning center to build better connections between the Earthkeepers (Johnson & van

Matre, 1989) program experience, classroom learning, and teacher professional development. About two-thirds of Ellie’s current class attended Earthkeepers at a local environmental learning center in the spring of 2009 (when they were third or fourth graders), the year before the present study was conducted; this year’s class (2009-10) with 10 new students also attended an overnight experience at the same center in late September 2009.

Negotiating site access

At a preliminary meeting in May 2009, the two grade 3-5 teachers and principal met with the director of the field center and myself to imagine strategies for integrating Earthkeepers and the four environmental concepts into the science and mathematics curriculum. I asked at the end of this meeting whether the school and the teachers would be open to my doing a dissertation study with them and their students. All replied they would. Ellie and I met approximately every other week from August 2009 to January 2010 to brainstorm the project further while I developed a research plan. During the project, Ellie and I met weekly to discuss progress and revise plans during the study, except for a weekend in the middle when she was out of town.

This phase followed general principles for qualitative design outlined by Bogden & Biklen

(1992). I had already gained access to the teacher and classroom through our previous work together, and gotten to know the students informally through volunteering in class and at

Earthkeepers (Fall, 2009). The teacher, co-teacher, interpreters, and principal reviewed the Human

Subjects applications before submission and provided a letter of site approval. Once Human

Subjects permission was gained, I spent a week consenting students and “hanging around” (Becker,

1998) to learn the setting and develop further rapport.



Ellie attended two years of a professional development program, Systems Thinking in Middle

School Science, designed and co-taught by a senior university microbiology faculty member, a Systems

Dynamics professional, and myself in 2004 and 2005.

Ellie employs systems thinking practices and tools throughout her curriculum, especially in science and social studies. We have continued to work together as colleagues in a study group on systems thinking, and more recently, through our work in earth education and learning about natural systems. We share two core beliefs: that all children are capable of developing sophisticated understanding of natural systems, and that the diversity of her students’ interests and abilities brings funds of knowledge that enlighten everyone in the learning community.

At the school, teachers of grades 3-5 “loop” so that the students stay with the same teacher for three years. Ellie’s class of 28 students during the study (spring 2010) included eight children in

Grade 3, ten in Grade 4, and ten in Grade 5; there were 15 girls and 13 boys. All students and their parents chose to be included in the study; both parents and children signed Human Subject Consent forms. One parent declined to have her child be interviewed but consented to other parts of the study.

Focal student

The original study plan called for a cross-case comparison of four focal students. For the first two weeks (6 sessions) I observed, videotaped and made field notes across the whole class of 28 students. In order to closely examine student thinking about ecosystems and the four concepts, I then selected a sample of four students from the entire class to study in depth. Two criteria guided my selection, which was reviewed and discussed twice with input from the teacher. The first criterion was to select only children who were able to hear and speak spoken English. This class included hearing-impaired and non-hearing children, and I did not know American Sign Language


(ASL) fluently. It is a communicative norm within the classroom that everyone should sign as much as they can, and a classroom norm that everyone helps everyone else communicate and understand.

Both teachers were fluent in sign language, and there were two sign-interpreters. Thus I was able to follow the flow of whole class discussions without a need for fluency in ASL.

Two of the four students selected, Jorge and Rosa, used hearing aids and worked with a speech therapist on speaking and listening but were able to carry out conversations with other students, the teacher, and myself without sign interpretation. For the times in class discussion where they relied on sign instead of speaking, the teacher or an aide translated their sign into speech immediately. Students’ use of sign was recorded in the video, but not transcribed unless I decided the sign or gesture was essential to include in order to interpret their meaning related to the study.

The second criterion for selecting focal students was to maximize the diversity of students included in the study. While students were administratively categorized as grades three, four or five, the class included students with a wide range of abilities. During the first three weeks of the study students sat at their “science tables” in groups previously arranged by the teacher. After setting up the video recorder to capture the whole class from the back of the room, I moved from table to table with the teachers to observe and interact with students. In several activities I carried the second, hand-held video recorder with me and recorded these interactions. At the end of the fifth session, about three weeks into the study, I reviewed these videos and student work artifacts with the purpose of selecting students of different ages, grade levels, and use of different representational strategies (drawing, writing, diagramming, etc.). I then met with the teacher to confirm the ages/grade levels and elicit her perspective on my initial selection of four focal students and four back-up students. We discussed re-arranging the science tables to group the eight possible students in two tables of four, with two sets of partners at a table. The teacher suggested partner pairs based on her knowledge of who worked well together, and whom she had previously teamed based on

99 different strengths of each student. The two groups consisted of five girls, three boys, ranging from ages 9 to 11, and grade levels 3, 4, and 5. At the start of the following week, students were assigned

“new” science tables for the remainder of the year.

The first set of four students I selected quickly demonstrated they were uncomfortable with being videotaped and audio taped as a small group. After reviewing their work across two sessions and consulting with the teacher, I decided to switch focus to four students at an adjacent table, one of whom was Jorge. They were already selected as the backup focal group and were also being videotaped and audio taped separately. The teacher and I discussed the first group of students’ behaviors and concerns with them privately. All agreed they wanted to continue being in the study. I asked if it was OK with them to continue videotaping and audio taping their work as a group “to help me think more about your thinking.” They all agreed, and appeared more relaxed. The new group of focal students, Table 4 (Jorge, Rosa, Annie, and Sonya) paid little attention to the audiovisual recording of their work.

While focusing on these four students I continued to film the whole class during discussions.

A second video and audio set-up focused on Table 4. Unfortunately, the noise level in the room combined with Jorge and Rosa’s soft speaking voices and movement of the children limited the transcription I could accomplish for their work as a team and a table. After all data were collected and I made a first pass on the data for Table 4, I recognized that Sonya was absent from a number of lessons. Her absence and the amount of data led me to a decision to focus on Jorge and Rosa.

Further into the analysis it became obvious that there were enough data on Jorge’s thinking to limit the case study to his work. I include Rosa’s work on their BioBottle design and science poster. Field notes and the video data from Table 4 indicate that Rosa often deferred to Jorge or agreed with his ideas. Thus Jorge became the focal student for this study.


Researcher Role

This study focused on how students think about ecosystems in the context of a modelingactivity. My framework specifies that learning takes place in and through interaction with others, as well as with physical objects and information generated by authoritative texts. In Lesh & Doerr’s

(2003) and Lehrer and Schauble’s (2006) models and modeling framework, teachers and researchers interact with children when they see opportunities to facilitate or focus children’s thinking by asking questions and suggesting possible strategies or tools. It is up to the students to decide whether and how to respond to such questions or suggestions. In my teaching and research practice with students and teachers, I have found a useful stance to take is one where I offer to help them talk about or think through an idea they are having, but not to answer questions directly because they are capable of developing an answer or getting more information themselves. My role as researcher, as Eleanor

Duckworth aptly summarizes, is “to probe children’s thinking, to appreciate how they are making sense of a situation, and to understand their understanding” (1996, p. 84).

My ethical stance on research is to ensure that any research is done with teachers and students, not to them, and that both students and teachers’ voices need to be respected and heard. At all times the research should contribute to students’ learning rather than take away from the already precious time of instruction. Lastly, I believe the best research, like the best science, is done with a spirit of adventure and inquiry. I did not know the “answers” to this study, and continually checked in with myself during analysis to ensure that neither I nor the teacher felt we were “proving” that systems thinking is “what works” for all students. In fact, during the course of the study and initial analysis I found myself increasingly skeptical about whether the direct instruction of systems “tools” or “thinking” was a cognitive scaffold or source of overload for students. My reflections eventually became impetus for further explorations of the data, about the relationship of children’s ideas and the representations they chose to frame the expression of their ideas.


The BioBottle Project Design

After initially thinking that I would lead the study lessons, both the teacher and I reconsidered. We agreed that I would be responsible for planning the tasks for the study, with input from Ellie in order to integrate the Water (FOSS) science unit she was teaching that semester. Ellie would take responsibility for “delivering the content,” as well as for classroom management and discipline. Ellie suggested she should coach me on her expectations for management and discipline so that we used the same language with students. We agreed that my role in relation to the students was that of a teacher and researcher who is working with them and their classroom community

(including the other adults) to better understand their thinking. This was the frame I used in introducing myself and the study to children when asking their consent and throughout the course of our work together (Field Notes, January 10, 2010).

A models and modeling task (Lehrer & Schauble, 2006) provided the context to elicit children’s thinking related to the research question. The BioBottle Project and decomposition bottles activities were adapted from the Bottle Biology curriculum guide (Ingram, 1993). The project as it unfolded over twelve weeks engaged students in iterations of considering the overarching problem of how to design a sealed model ecosystem (the “BioBottle”) that could support a Wisconsin Fast

Plant™ through its life cycle alternating with tasks that engaged them in considering ecosystem structures and functions to fill important ecological ideas so they would have enough information to make informed decisions about design. The task engaged students in “thought-revealing activity” in which students conduct a scientific inquiry and construct scientific and mathematical inscriptions to negotiate and reflect on their thinking (Lesh, Cramer, Doerr, Post, & Zawojski, 2002; Latour, 1990;

Roth & McGinn, 1998). Building and observing models of ecosystems provided students with opportunity to explain and negotiate their ideas about what to put in the model and how that component might interact with other components to enable a species known as a Wisconsin Fast


Plant™ (a variety of mustard) to complete its roughly 28-day life cycle from seedling to seed production. Students applied the understandings they had developed from both school and informal learning about ecosystems over an extended period of time. The task was “counted” as a regular part of students’ academic schoolwork, with performance expectations and grades assigned by the teacher for dimensions of participation.

Data Collection and Management

Data for this study include field notes recording informal conversations, videotaped wholeclass session , my reflective journal, photographs of the classroom, and a complete record of Jorge’s artifacts, including notebook entries, data sheets, and formative assessment tasks. Artifacts collected from students included—

1) Individual student work products such as science notebooks, informally-generated inscriptions, and five individual assessment tasks distributed across the 12 weeks of the study;

2) Video and audio recordings of selected whole class and small group activities;

3) Science posters generated by collaborative pairs or teams;

4) A post-project, semi-structured interview with selected focus students (video recorded);

5) The model ecosystems themselves (observed in real time and photographed at intervals).

Management, logistics, and timelines

This section provides an overview of the general sequence of the research plan, documenting phases of gaining access, mapping current knowledge and questions, and conducting the modeling task. For study planning and management, I have relied on Bogden and Biklen (1996) and Miles and Huberman (1994).

The plan vs. the reality

As with any study conducted in real-time classrooms, plans for the unit quickly shifted to

103 accommodate contingencies such as changes to the teacher’s schedule and plans, flow of lessons, and the myriad interruptions of special projects, assemblies, an unexpected cross-classroom social studies unit, illnesses, absences, and chance. Figure 1 presents the actual implementation of the study, with dates, sessions (hours), and data artifacts collected.

Figure 1: Timeline with major tasks of the Bottle Biology unit by week

Classroom Observation, Videotaping, and Field notes

The teacher of this class often used video recordings in her own action research and encouraged her students to use a digital camera or the video recorder any time they felt that

“something important” should be captured. Thus, adults observing and video recording activity was a norm in this classroom, with expectations that children would stay focused on the work at hand. I

104 originally planned to use two cameras, one for the whole class and one for the focus group of students at one table. I had originally designed time in between sessions to write field notes, do a rough transcription of video, and make adjustments to the next weeks’ plan. In reality, most of the video uploading and review was done on the weekend following the week, and my field notes were less of a system than a jumble of notebook pages, scraps of paper, and reflections from memory. I managed to observe and video record every class session related to the study except the second hour of the final assessment, which I was not aware of until afterwards. In addition to the main video camera that focused on the whole class or the group of four focus students, I used a second video camera carried with me to record my interactions with individuals or table groups. During the Bottle

Biology observations, I placed this second camera on a tripod to film the focal students at Table 5.

The data from these videos and accompanying audiotape turned out to be very difficult to use for the study. Students frequently moved around beyond camera range, and ambient noise in the classroom interfered with the quiet talk of the focal students. The whole-class camera captured more useful data.

Informal Conversations with the Teacher

The teacher and I met approximately once a week to plan how the project would integrate with her science curriculum unit on the water cycle. Each week during the study the teacher, coteacher, and I scheduled a debrief session to check in on logistics and any adjustments to the plan. I kept notes during the meetings, which were incorporated into my field notes and future lesson plans.

These meetings became less frequent and less productive toward the end of the year as the teacher became overwhelmed with the number of other activities she needed to coordinate and complete.

Other Data Sources

• Field notes

• Personal reflective journal entries


• Pertinent email and logistical communications

• Lesson plans, physical artifacts and other materials used during the modeling task

• Conceptual storylines and description of enactment for each of the three science curriculum units and the Earthkeepers experience (as background information on children's opportunities to learn the concepts prior to the study).

Data Management

For data storage, management, and analysis I used the Microsoft Office suite of productivity software as well as HyperResearch software version 2.8.3 for categorical coding and cross-referencing.

Each session video recording(s) and audio recording(s) were uploaded and checked for quality at the end of each week. All student work was photographed weekly and then summatively at the end of the semester. I took digital photos of all artifacts at multiple points in the study. This includes a complete record of every page in a child’s notebook, their individual or shared data sheets, individual assessments, and collaborative posters as well as a record of any changes to their thinking over time.

Original identities of students and teachers remain in the video and audio recordings, which are not part of the data corpus. Transcribed portions of audio or video were used as data in the analysis; these transcripts have original identities replaced by pseudonyms. The original video and audio files are stored on two hard drives that are kept in separate locations, locked, and inaccessible to others or any machines connected to the Internet.

Digital photos of student work are also kept on these drives. For Jorge’s work products, I made a copy of each of his artifact photographs and edited out any identifying information with an image processing software program. All data with identifying information are stored and backed up securely following the plan outlined in the UA Human Subjects approved plan (March 15, 2010).

Digital record transcriptions

A rough database of each session at the level of task, events, participants, and topics was

106 created the summer after the study was completed (see Table 3.2, above). This database also included an index of all photographs or hard copies of artifacts that all 28 students produced.

Photographs of work produced by the two focal students were edited to de-identify each artifact and copied onto a password-protected hard drive for analysis.

Transcriptions of video and audio recordings of whole class discussion were made over the summer following the study. A first rough “gloss” transcription was typed almost in real time as I watched the videos, much as if I had been observing and taking notes during the actual class. From these I reconstructed Table 3.2 above, which is the record of driving question(s) and purpose(s) for each session. I also constructed a database of all work artifacts that students produced, quality of the data recordings, and notes on events or contingencies that seemed important to further interpretation cross referenced to the class sessions. For example, one contingency affected results from the session investigating the interaction of the bottles with sunlight energy. It was the only day of the entire semester with heavy clouds and rain. We switched at the last minute to working indoors with lamps, but these did not have enough power to effect change in the bottle system. I made notations of these types of contingencies in the database and took them into account during analysis when I felt they may have affected students’ engagement or sense making.

This first level of transcription in almost-real-time was a decision I made to force myself to view the actual flow of events and outcomes of the unit with a level of detachment. After transcribing each lesson, I wrote privately about what I call the “woulda-coulda-shoulda” hypercritical thoughts I typically experience when reviewing my own teaching or interaction with students.

This process somewhat enabled me to conceptually and emotionally separate my own investment in relationships with the teacher and students and attachments to the study as a teacher and researcher with the “is that is.” In this process I worked to make a conscious shift away from judgment about

“what worked/didn’t work” about the curriculum design or lesson enactment, and towards a focus

107 on the students’ interactions with and responses to events as they flowed. I then set this level of transcript and my journaling “to the side” while I conducted the analysis of student work. Of course, the thoughts and perceptions, including my own biases, remained active during analysis. However, I did not refer to these notes directly, nor did I share them with the teacher. I revisited my reflections on what worked/didn’t work when writing the implications and conclusions of the study.


Near the end of the unit, I interviewed Jorge on videotape with an audiotape backup for the purpose of probing his ideas about changes in the BioBottle beyond what he and his partner represented in their science poster. During analysis I made a decision to transcribe the interview and work from the transcription, using the video as a backup for interpreting meaning through tone of voice, action, gesture, and gaze of the student when needed. The interview, being primarily languagebased, was not analyzed in the first semiotic analysis but was included in the second, etic analysis.

Academic task artifacts

I also transcribed all of Jorge’s written work artifacts such as notebook pages and data sheets. Doyle (1992) argues that curriculum intentions are always transformed through pedagogical action, which is in turn influenced by factors of interpersonal dynamics between teacher and students, classroom management of events, materials, transitions, and often subtle negotiations and decisions from both teacher and students on “what counts” as acceptable work for a student product. In order to interpret Jorge’s sense making in his written work, then, it was essential for me to capture and analyze the features of classroom discourse and to assume that work products were an embedded, systematic production within a larger sociocultural system of classroom events. These transcripts captured verbatim the introduction the teacher gave for the task, and indicated when and with what purpose individuals added to the whole-class discussion, or talked among themselves in small groups. This level of transcription focused on the teacher’s actual words and actions as she

108 interpreted the driving questions for science inquiry and written lesson plans I had written up after our meetings.

Each student produced up to six work artifacts during the course of the unit. Five of these were designed as formative and summative assessments. The six artifact consists of the student's entire science notebook, a wire-bound lined paper journal interspersed with data sheets. I took photographs of all 28 students' notebooks and work artifacts at the end of the unit. For the visual artifacts, the photograph became the data record. I transcribed work verbatim and cross-referenced these transcripts to any visual records. I transcribed all written text verbatim, using students' original spelling with my interpretation in [brackets]. Any visual information on the page was noted and interpreted semiotically to interpret additional layers of meaning.


Analysis Level 1: Phenomenographic Semiotic Analysis (Emic Perspective)

In the first pass of data analysis, I interpreted Jorge’s drawings, diagrams, systems representations, and other visual information were interpreted based on Kress and Van Leeuwen

(2006) and informed by Lemke (1998). As argued in the conceptual framework, the visual or other non-linguistic representations children construct both mediate and communicate their unique perspectives and meanings. My goal in this analysis was to interpret Jorge’s perceptions and meanings from his perspective in relation to the content domain and academic task. The four ecological concepts and synthesis concept of “ecosystem” were used as organizing categories to code artifacts and transcripts. Within each of these categories, Jorge’s own language, perceptions, and meanings

(themes) emerged. This semiotic analysis was both descriptive and interpretive simultaneously. Due to the tight timeline for access to the students, I was not able to interview Jorge about his artifacts, which is a limitation to my interpretation and the study that I would address in future work.

The close analysis of each artifact looked both for what was present and not present in

109 relation to the ecological concepts, although I assumed the presence or absence of an idea in an artifact did not necessarily indicate understanding or lack of understanding for the child. A section below provides a sample analysis of the first of Jorge's artifacts, the pre-assessment task. This example demonstrates how the artifact was first situated in relation to the academic task as presented by the teacher, and then interpreted semiotically. This level of analysis included questions, uncertainties and conjectures that arose for me about Jorge's sense making about ecosystems that I returned to after conducting the second (etic) level of analysis. The semiotic data were also used in the larger corpus of data used in the second level of analysis.

Example of the semiotic analysis process

Academic Task 1 (April 10, Day 1): Assessing prior knowledge. This is the first task that students completed on the day after everyone’s consent was confirmed. The teacher begins the class with a review of her expectations on their work effort using a poster at the front of the room titled

“Perceived effort.” This is to reinforce both of our earlier introductions that the work students do as part of this project will be “counted” as part of their regular school tasks and will be factored into their assessment (grades) for the quarter. She introduces the task with an emphasis on “making thinking visible:”

Teacher: All right. When people get jobs, work on problems, first thing they do is get together to

figure out what they know about ecosystems, and what they don’t yet know. How are we going to get evidence of what you already know? This is what J— knows? How can we make what’s in your mind visible to you. Discuss at your table. What’s one way? How can you show me what you know

about ecosystems? What can you do?

[She gives them less than a minute to talk at their tables, then elicits ideas. Various students suggest:]

Student 1: Words, numbers and pictures.

Student 2: Show some diagrams.

Student 3: Mind map. [This is the name students use for a concept map].

Student 4: Text box…[a specific form the class has learned that includes an illustration with a text box, label, captions].

Student 4: You can just write it down on paper!

Teacher: You mean just write down like a story, sentences, paragraph? What do you want me to

write here [on the board for all to see]?

Student 4 : Paragraphs.

Teacher: [writes]. Ok, so that’s what we’re going to do. You’re going to get a large piece of paper.

The first thing you want to do is put your name and today’s date in the upper right hand corner. Do

that now. Upper right hand corner. [I pass out paper]. [30 second pause]. 5, 4, 3, 2, you’re

done with that, look up here…[3 second pause]. OK. You need to remember everything you’ve

learned about ecosystems, [2 second pause] think about questions that you have about ecosystems,

and think about possible places for us to look for answers about those questions [speeding up]

because remember your job? [1 second pause, gestures vaguely towards materials at the left side of the room.] You’re gonna build an ecosystem in a bottle. Hopefully a successful one.

So right now, we don’t know yet what all of you know. We’ve learned a lot, but I don’t know what

you really know. So on one side of the page, you’re gonna draw what you know [points to board].

This is like the diagram, labels, pictures part. [underlines these words on board]. The mind

mapping part. [underlines]. So use pictures, words, diagrams, mind mapping to show everything

that you know about ecosystems [writes “ecosystems” on board at top]. [To me] Is there

anything else you want me to add to that?

Elisabeth: [To students] Just a suggestion that as you’re doing this you might want to start by

closing your eyes a minute and think about ecosystems with all of your senses. What would you see,



what would you hear, how would it feel, how would it smell, how would it taste?

Teacher: [Nodding] Yeah. You have been in ecosystems. [1 second pause] This morning. Um

hm. You have been in different ecosystems on field trips. So you want to make it as detailed as

possible. [writes as she talks] Smell, taste, what you saw, what you felt…

Elisabeth: What you heard…

Teacher: What you heard. OK? We’ll give you about…[shrugs, 1 second pause] we’ll see how

10 minutes does for you. Ten minutes. Get started.

Note: Students work on this representation with further instructions over about 48 minutes.

Figure 2: Jorge’s pre-assessment work product

Jorge has divided his paper into five parts by drawing four boxes (frames). I discuss these from top left to bottom right.

Frame A. The box in the upper left corner is composed of images and labels. He draws a

112 dog (labeled twice, once below and once on the body), possibly a cat (not labeled), a bug (labeled) on the cat. They are standing under and looking up at a tree (labeled) with a wide trunk and scribbles for leaves. Short vertical hash marks around the tree are labeled “rain.” A horizontal line anchors the tree and animals about halfway down the space in the frame. This seems to represent the surface of land, but there is no label. Below the horizontal line under the tree, about 16 extended vertical lines with a more curved shape radiate out from under the base of the tree trunk and extend down and even a bit beyond the bottom of the frame. These are not labeled. There appear to be two points of view illustrated: what a person would see above ground, and what a person might see if the area of the ground was cut away or cross-sectioned. I am intrigued by how Jorge has divided this frame to show not only visible but also non-visible elements, and that the non-visible elements comprise almost half the frame. Yet nothing below the ground line is labeled.

Ecosystems/systems. All of the elements in the drawing relate to each other, as they would in a natural setting. The composition includes both representational and conceptual elements (these being the verbal labels as well as the cutaway view of roots or water underneath the tree). Jorge also illustrates not only what would be visible to a person observing at ground level, but what is typically not visible, the roots or water in the ground. The living components he illustrates are: an insect, two mammals, and a tree. The non-living components are rain (water), and ground (surface and cutaway view). He labels the tree, rain, dog, and bug, but not the cat, ground surface, or roots/water below the ground line. I notice he does not label the leaves of the tree, nor does he include the sun in this part of the page. In labeling objects, Jorge focuses on the whole organism: animal, tree, bug, rather than parts of organisms (leaves, roots). The objects in the scene depict fairly accurate scale relationships (the tree and underground are larger than the animals, the dog is larger than the cat, the bug is smaller than the cat yet slightly larger than life (unless it is a very large beetle or bee).

Energy. The cat looks up at the tree, and the dog, standing behind, looks at the back of the

113 cat where the bug has landed. Does Jorge mean to imply the dog will eat the bug (or the cat!), and that this arrangement represents a food chain? He does not include the sun in this frame.

Materials. Jorge specifically shows and labels rain. He draws but does not label ground/soil as a surface and an area with a cross-sectional view. Air or atmosphere is not labeled or indicated.

Interactions. The cat looks at the tree, the bug is on the cat’s body, the dog looks at the cat/bug, The lines below the ground could represent roots of the tree or rain soaking into the ground, or both. Because many of these lines radiate from directly under the tree, I infer they are roots, but the lines could also indicate water in the soil. The roots of the tree and/or water appear to interact with the ground, not at the surface, but underground. Jorge’s drawing of roots or water underground suggest that he may be thinking about how water moves through soil, or how the tree roots are distributed to collect water from the soil. I notice that the root/water lines take up slightly more area on the page than the tree itself. I wonder if Jorge understands that healthy tree roots typically do extend to the height and drip line of a mature tree, or if the roots are more important to him than the other parts of the ecosystem?

Change over time. If Jorge is thinking about processes in the system, he does not label them as such. He labels rain, but because all his other labels represent things, I would not infer he thinks of rain as a process. The illustration is in a frame, which may imply a “snapshot” or one-point-in-time perspective. On the other hand, the cutaway view of the soil with lines extending down from the tree could suggest interaction of tree, water, and soil over time, implying growth and change. Since all other parts of the drawing appear to represent a moment in time, change over time is, at most, suggested.

Analysis level 2: Jorge’s Expressions in Relation to the Content (Etic Perspective)

The second level of analysis consisted of using the qualitative software analysis tool

HyperResearch to create a coded network built on both visual and verbal artifacts of Jorge’s sense

114 making to identify both domain-specific ecological concepts and domain-general concepts related to systems. The HyperResearch software enables coding of text, video, and graphics. The purpose of this analysis was not to compare Jorge’s sense making about ecosystems to a scientifically “correct” interpretation, but more pragmatically, to enable me to locate Jorge’s sense making in relation to content in order to track the consistency of his expressed ideas or explanations across tasks and through time.

Example of the coding process

Again, based on my theoretical framework, the coding categories were constructed to capture both the conceptual content under investigation and Jorge’s individual interpretations and meanings related to that content. The codes therefore consisted of two levels. The first was the content level category, and the second qualified the Jorge’s idea using his language whenever possible (sometimes linguistic, often visual). His ideas about the role of water and water cycle were subsumed under the “materials cycling” category. For example, a full code would read, “abiotic materials—water—rainwater soaks into the soil near tree roots” from the visual artifact analysis example described above. Each code was linked in the HyperResearch database to an artifact of Jorge’s work, both visual and verbal, across situations. This enabled me to identify situations in which his sense making and explanations were consistent or varied across learning episodes, which I define as bounded by an academic task. The analysis also enabled me to identify which structures and functions Jorge described most frequently, and the level of specificity he used in his narratives or explanations of relationships and functions.

Analysis Level 3: Role of Water in Ecosystems

As evidenced in the literature reviewed, learning about complex systems via modeling practices requires that a learner shift perception and explanation from surface description of system parts (e.g., all ecosystems contain living and non-living components) to a synthetic perception of

115 structure (how components are interrelated) and function (how these structures contribute to a system’s processes and behavior over time). The two previous levels of analysis enabled me to identify the frequency and type of interactions or relationships Jorge described in relation to ecosystems and his BioBottle across the unit. Using a modification of a systems thinking tool called a “Connection Circle” (Quaden & Ticotsky, 2004) to construct a visual network of components and interactions based on the emic and etic coding of all of Jorge’s artifacts, both visual and verbal.

From this level of analysis I determined that Jorge focused primarily on four structure-function relationships involving sunlight, plants, water, and soil. His descriptions of the role or function of water in relation to organisms or the ecosystem became increasingly detailed and focused on microscopic interactions over the course of the project. At the same time, his sense-making about water appeared vary noticeably depending on the context of explaining water in the BioBottle vs. explaining water based on his experience.

I conjectured that Jorge’s focus on water across the project was influenced by the contingency of merging the curriculum of the FOSS Water unit with the Bottle Biology Project, as described above. This case therefore provided a unique opportunity to study how an elementary student might make sense of the role of water and the water cycle in relation to ecosystems, as water is typically taught from the disciplinary perspective of earth or physical sciences (Roseman & Stern,

2003). From my own experience as a curriculum coordinator for a local school district, I remembered too an extended argument held by elementary teachers who criticized several curriculum units that taught water and the water cycle as not including living organisms. The FOSS unit taught in this study had been amended by the district to include “water is necessary to support all living things” in the conceptual storyline.

The confluence of curriculum contingency and data emerging from my analysis provided an opportunity to discover where and how Jorge made (or did not make) connections among his

116 understandings of water in geophysical contexts (the FOSS unit) and ecological contexts (the

BioBottle project). Therefore, while the initial analysis addressed Jorge’s sense making in terms of five ecological “big ideas” (energy flow, materials cycles, interactions, behavior and scale), I further bounded the case analysis by the role of water in order to investigate how elementary children make sense of water across abiotic and biotic subsystems of an ecosystem.

Methodological Trustworthiness, Credibility, and Limitations

This study is a naturalistic inquiry (Guba & Lincoln, 1994) in the tradition of qualitative research. In qualitative approaches to research, the criteria for judging a study rely on the researcher's integrity, transparency, and awareness of personal biases. As Patton (2002) emphasizes:

"Any given design inevitably reflects some imperfect interplay of resources, capabilities, purposes, possibilities, creativity, and personal judgments by the people involved" (p. 3). This design most certainly does. Strategies for increasing trustworthiness (validity) and credibility (reliability) that have been used in this study to address possible biases include the use of multiple data sources and analysis lenses, which allow for triangulation of findings, systematic and rigorous approaches to coding based on the theoretical framework, and the provision of detailed data to support claims so that readers may make their own judgments about the assertions made (Creswell, 2003; Merriam,

1998). Qualitative studies are not "generalizable" in the sense of being readily applied to other contexts or used as a blueprint for "scaling up" an intervention.


As in many qualitative studies, decisions made about methods changed as the study proceeded and new ways of “seeing” the data emerged. The methods also were adapted to pragmatic concerns of time, money, and the emergence of the role of water and the water cycle as findings that could contribute to the literature on curriculum and instruction related to elementary students and the domain of ecosystems.




This chapter presents findings of my analysis of Jorge’s sense making within and across learning situations. The overall question addressed by the study was:

Within the context of a culminating class project to design a functioning closed ecosystem, how did the interactions between a learner, the academic tasks, and modeling practices influence what he perceived and communicated about the structures and functions of an ecosystem?

A second, question that further bounds this case emerged from the contingencies of designing the project and findings from the initial analysis:

In particular, how did this student perceive and communicate his sense of the form and function of water in the microcosm and the interaction of water with other living and non-living components of an ecosystem?

I first analyze the ways that Jorge expressed ideas about water in ecosystems in the curriculum tasks leading up to designing the BioBottle and deciding which variables to measure or track to describe changes to the model ecosystem over time. Jorge and his partner’s work on designing and observing the BioBottle, and the construction of their scientific poster are analyzed next. I then follow with an extended comparison of Jorge’s talk about the water cycle in the context of the sealed BioBottle versus his discourse with me during the semi-structured interview.

Academic Tasks

As a reminder, Figure 3 below outlines the flow of the curriculum tasks and the ecological concepts that each was designed to teach for the 12-week unit.


Figure 3: Academic tasks across the study

Figure 4 below shows which ecosystem components were a focus of the major activities of the project, either by curriculum design or by Jorge’s interest in the open-ended pre- and postassessments. While many of the curriculum tasks integrated content from the FOSS Water curriculum unit, Jorge also included ideas about water in his pre- and post-assessments, and in our semi-structured interview as well. The only task in which he did not communicate about water was in Assessment 4, which focused on energy flow.


Energy Sun Water Air Soil



/ rotting food

Assessment 1 4/9: What features and interactions of ecosystems do you know?

Decomposition Bottle 4/16

How does sun energy affect water in the decomposition models? 4/19

Assessment 2

4/19: What is the role of water in ecosystems

How do types of soil interact with water? 4/20

Assessment 3 4/22: Interactions in an ecosystem

What will you include in the BioBottle


What variables in the BioBottle will you observe?


Assessment 4 5/10: How does energy flow through an ecosystem?

Scientific poster


Post assessment 5/22: What do you now know about ecosystems?

Interview 5/25

Figure 4: Jorge’s references to ecosystem structure-function across academic tasks

Plant Animal

Task 1: Pre-Assessment

Task and context

The pre-assessment task was given a day after I explained the purpose of working with the class and presented the goal of having them design a model ecosystem. The teacher elicited ideas from the class on strategies they could use to “get evidence of what you already know [about ecosystems].” Students and the teacher brainstormed options such as “words, numbers, pictures” as well as concept maps, which the class and teacher call “mind maps.” Then she instructed students as follows:

You need to remember everything you’ve learned about ecosystems, [2 sec pause]. Think about


questions that you have about ecosystems, and think about possible places for us to look for answers

about those questions [speeding up] because remember your job? [1 sec, gesture] You’re gonna

build an ecosystem in a bottle. Hopefully a successful one. So right now, we don’t know yet what all of you know. We’ve learned a lot, but I don’t know what you really know. So on one side of the

page, you’re gonna draw what you know. [points to board]. This is like the diagram, labels,

pictures part. [underlines words on board]. The mind mapping part. [underlines]. So use

pictures, words, diagrams, mind mapping to show everything that you know about ecosystems

[writes “ecosystems” on board at top] (transcript, 4-9-10).


Students were given a large sheet of manila paper and choice of regular or colored pencils. I asked students to work from what they remembered rather than look back at their science notebooks from the Ecosystems unit so that the pre-assessment would capture their most salient and accessible ideas.

Interactions of child and task

Figure 5 below shows Jorge’s work product. The ecosystem Jorge chose to illustrate (in the upper left corner, Frame A) was a stereotypical backyard scene with a dog, cat, tree, earth and a worm in the soil. Water in the ecosystem is represented as rain. This ecosystem example is not representative of the desert southwest ecosystem he lives in, but may very well be a representation of his actual backyard. Jorge’s illustration of the structures in an ecosystem include all of the visible biotic and abiotic components of ecosystems: sun, water as rain, and soil for the abiotic components, and producers, consumers, detritivores (worms) and a generic category of “bug” or insect for biotic components.


Figure 5: Jorge’s pre-assessment work product

Soon into the task, however, he encountered a problem: how do I draw an interaction? Jorge calls the teacher over and asks a question I cannot hear.

T: That’s a good question. Can you show two things interacting?

J: A dog?

T: Interacting with…what? What might a dog interact with?

J: A cat?

S: Ok, so would that be a living or nonliving interaction?

J: [Shrugs].

The teacher shifts the intent of Jorge’s question away from ‘how do I draw an interaction’ to what kinds of interactions he might show, followed up by a question that introduces more formal science vocabulary from the Ecosystems unit (living or nonliving interaction). Jorge appears to interpret this as a hint that he should include an explanation of living/non-living interaction in his

122 work. What he does in Frame B is to reason from his own human experience and creates a dramatic and emotional illustration of this concept by showing a living person crying a puddle of tears over a dead person. This is not the concept the teacher (nor I) was expecting to see in Jorge’s ideas about ecosystems, yet it is a very personal and meaningful interpretation from his perspective.

As the class continues working, the teacher interrupts about 10 minutes into the task:

T: [to whole class] What’s more important in a system? The…parts? or the connections between

the parts?

S: Connections…

T: Connections are more important in a system. So make sure in your mind map to start showing

connections. [Holding up someone’s work at front of room]. So this is one way that you can

show connections, with arrows, and lines…[swaps to another paper] This is another way you

can show connections, with systems diagrams [walks up to each table in front row to show more closely, about 20 seconds for them to look, then swaps to paper 3]…this is

another way you can show connections [walks across front of room holding paper]. This is

another way you can show connections [4th paper]. They’ve got labels, arrows. So what we want

you to do is continue with your ideas, but you should be connecting them. Because connections are the

most important. OK? Keep workin’!.

Jorge appears to interpret these instructions to show connections or interactions more explicitly, which he did in Frame C and Frame D of his work. He again used an illustration of humans, waving “hi,” which he labels “interact.” He also labeled an arrow from the sun to the tree and the worm to the tree roots “interact.” In Frame C he merged his human life world experience of

“interact” with the more ecological interactions between sun and plant, worm and root. The teacher’s emphasis on using “systems diagrams” or arrows to show connections may also have sparked Jorge to create the diagram in Frame D.


Role of water

A close inspection of the illustration in Frame A shows that Jorge depicts the rain as slashes in the air and these slashes also appear beneath the surface of the soil, suggesting an idea that rain is absorbed into the soil. Despite the teacher’s emphasis on “showing interactions” Jorge does not label the water-soil interaction, and water does not appear in any other frames of Jorge’s preassessment. In the second part of the pre-assessment, students complete a KWL chart for themselves; Jorge does not include any knowledge or questions about water, perhaps assuming water as a given, something he is already familiar with or not germane to his particular learning interests

(Notebook, 4-9-10).

Practices of modeling

Analyzing this artifact is where I first noticed Jorge’s tendency to become so focused on some details of what he is trying to communicate about the ecosystem as well as respond to the instructions of the teacher that structures or interactions in ecosystems (e.g., the sun’s light or water) fall out of his perceptual and cognitive view. The multiple structures (components and interactions) in an ecosystem come in and out of focus as Jorge attempts to capture all that he knows or is making sense of during an act of communication, which is a task difficult even for experts and adults (Meadows, 2008). This dynamic of Jorge’s communication reflects the challenge of relating all parts of a system with each other, while also relating the parts to the whole. At the same time, Jorge has constructed a multi-layered set of interrelated representations that show he makes sense of ecosystems using resources from his perceptions of the real world (his backyard), his experiences as a human (living/non-living, and interacting), and more scientifically by presenting a model of a food chain and/or materials cycle similar to the “expert models” the class learned in the Ecosystems unit.

Despite Jorge’s lack of formal instruction in modeling practices, Frame A in this example is a hybrid of an illustration and a model, due to the “cutaway” perspective he uses to show his

124 knowledge of how tree roots interact with the soil under the surface. He also shows rain seeping into the soil near the tree roots. In fact, he uses this inscriptional convention of showing the roots of a tree or plant in the soil for all but one of his visual representations across the project. The cutaway view approaches the practice of modeling by explicitly identifying unseen dimensions of a system.

The plant-root-soil-water network is itself a subsystem that exemplifies the interdependence of biotic and abiotic components of an ecosystem. Its appearance in multiple representations and narratives suggests this network is perceived by Jorge as a significant feature of ecosystems.

The other model Jorge uses in this task is the cycle diagram shown in Frame D. This appears to be an explanation of feeding relationships in the local desert ecosystem. Jorge implicitly indicates the role of the vulture as a detritivore by specifying a comparison between the living coyote eating a snake and the dead coyote being eaten by a vulture. This model may also suggest an explanation for the cycling of materials from and to the soil. The arrows are not labeled so their meaning is ambiguous.

Task 2: Decomposition Bottles

Task and context

The decomposition bottle learning experience was designed to introduce students to the process and products of decomposition, and the trophic level of detritivores, which were not included in the original Ecosystems unit. This was the first instance in which we used the bottle model in class.

To put this task in context I need to provide the following background. The teacher and I originally planned to have students construct their own compost bottle models and observe changes over two weeks, but a delay in the start of the project limited our timeline. To compensate, I videotaped myself at home constructing six compost bottles, using everyday language to describe the contents and process: green weeds, dead brown leaves and twigs, old food from my refrigerator,

125 including six pieces of a half-rotted apple, six sprouting potatoes, coffee grounds, tea bags, and pieces of lemon. I also added different amounts of a liquid “secret ingredient” (compost tea, to introduce microorganisms) and encouraged students (in the video) to think about what that ingredient might be. Figure 6 below is a frame grab from the video that shows the six bottles 24 hours after they were assembled. The different amounts of liquid have filtered through the materials and gathered at the bottom of the bottle, except for Bottles 1 and 5 which did not receive any added liquid. The video next shows the bottles two weeks later. I “zoom in” (a term students are familiar with from work with the Private Eye jewelers’ loupes in science observations) to show Bottle 3 more closely. Condensation can be seen on the inside of the bag “lid,” but I do not draw attention to that in my narration. Instead, I take off the lid and show a close-up of the bottle in natural light, which shows a variety of mold and fungi growths (Figure 7 below). I deliberately chose to not use scientific vocabulary to name these as “mold” or “fungi” until the children had the opportunity to see them for themselves and develop a sense of curiosity about what might be going on.

Figure 6: Compost bottle models at 24 hours


Figure7: Bottle 3 showing mold and fungal growths

By the following week when I was able to have time with the class, the molds and fungi had completed their life cycle. The compost had become a thick brown goo with a few recognizable lumps, and all six bottles now had brown liquid in their bottom reservoir.

During the introduction of the compost bottle models, the teacher asked the class, “What’s a model? How do scientists use models?” She gives students about 7 minutes to write in response to these questions.

During an extended whole class discussion the teacher tries different strategies to shift students’ answers from the model-is-a-smaller-copy-of-the-real-thing and towards using models as a way of exploring behavior in a system, or as she says, “things working together”. She then says, “I’m going to change this [erases “model” on the blackboard] to this [writes, “modeling”], from a noun to a VERB. We know as systems thinkers we care about how systems are working.” She uses the model globe on her desk to emphasize that it doesn’t just show where things are but it is “modeling how the earth rotates.”

Jorge’s notebook page is shown in Figure 8, with a glossed transcript of his writing.


Figure 8: Jorge’s notebook entry in response to “What is a model?”

“A model is some thing, like a thing you, let’s say a car. A car is heavy to pick up but you can have a model that you can pick it up. In [science] we always use model[s] to understand what is happening. Like [arrow pointing to the model of the earth]. On the side he has drawn a person with a balloon labeled “Brain but it’s a balloon.”

The teacher then introduced students to the actual decomposition bottles placed on their

128 desks and encouraged them to “think about what’s happening inside. What are the interactions? What’s living,

non-living?” (video transcript 4/16/10). The teacher emphasized that these are the same bottles in the video that are now on their desk, which surprised several students. The teacher instructed students to first observe the bottle without opening it, while they complete a notebook page entitled “I notice…I wonder,” a standard observation and recording strategy used in the district. While they observed and took notes the teacher circulated and repeats instructions:

T: I want you to go beyond what you’re seeing to behavior. What are the interactions?...Tell me,

this is happening because…I want to hear about behavior. I want to see VERBS. [After about eight minutes she elicits some responses from students, starting with questions about what is living and non-living, and then asks about interactions.]

S: Our group saw a worm moving through a tunnel and interacting with the soil.

T: How was it interacting?

S: By crawling.

S2: Some liquid seems to have evaporated and it goes to the top and then rains, kinda.

[The teacher elicits the vocabulary “rising” to the top and “condensation” on the lid “like the steamy bathroom mirror”.]

Interaction of child and task

This task did not appear to engage Jorge’s interest at the time as evidenced in his lack of questions and bare minimum written notes in his notebook. However, as I will show in his later work designing and observing the bottle, he gleaned quite a bit of information and ideas about fungi and bacteria, as well as the role of detritus in conditioning soil.

Throughout these parts of the lesson, Jorge listens and takes notes but does not raise his hand to be called on. Jorge wrote that he noticed: “Bug, maybe plant, water, eggs, air, food, dirt,

129 grub.” He got the word grub from his table mate. He wondered: “Gas?” After they opened the bottles (a rollicking good time for almost everyone to shout “gross!” at some point), Jorge wrote in his notebook for interactions: “Bug and worm, fly and [classmate], worm and the dirt, water, mud.”

He added additional sensory observations: “Dark, cold, clods [clumps?], egg, dirt, worm, food stinks!

Dirty water! Stick.”

In his observation, Jorge “noticed” air in the bottle, a component of earth’s systems and ecosystems that he did not include explicitly in his pre-assessment. I was not at his table at the time to observe whether he came up with the idea of air in the bottle, or whether this came from a member of his group. If the idea did come from Jorge, could it be possible that the “sealed” bottle model suggested the notion of a container, and that he knew air can be captured in containers? I made a record of this question in my field notes. I knew that Jorge was familiar with the idea of a gas: previously he had a conversation with his teacher during the pre-assessment during which a fellow student asked if she should include “greenhouse gas” in her mind-map. Also, on his “I wonder” page, Jorge wrote “Gas?” which together with the previous interaction suggests he is familiar with the idea of a gas and therefore, air. It could be that because he mostly drew his ideas in the pre-assessment, and as air is invisible, he did not think to explicitly label air as a component of an ecosystem, but took it for granted as being present. In this task, the bottle model may have afforded Jorge the opportunity to think about air being contained in the closed system.

In this task, the video of the model decomposition bottles provided Jorge the affordance of being able to observe the process of decomposition over time, as well as the changes of living material to detritus, and the release of liquids during decomposition. Jorge was also able to observe the end products of decomposition with his senses, which he did intently (the video shows that he controlled the bottle at his table for most of the open exploration time, and he dug materials out of the bottle to examine them more closely. See Figure 9).


Figure 9: Jorge and his table partner explore the contents of the decomposition model

Role of water

The constraints of watching the decomposition process on video limited Jorge’s opportunity to experience decomposition in real time, to feel and talk about the presence of water in the fresh leaves and food, or to observe water collecting in the bottom reservoir as the water in the leaves was released. Several students noticed and commented on the fact that the present-day decomposition models all had water condensate at the top of their “lids.” Both the teacher and I realized afterwards that we had missed an opportunity to draw students’ attention to the liquid at the bottom of all six bottles, and ask, if there was no liquid added to two of the bottles at the start, where did that liquid come from at the end? Jorge took the water he observed in the finished decomposition pile as a given and did not make a connection between that water and the ideas that all living things need water, or that water comprises a large proportion of mass in organisms.

Modeling practices

The introduction of the practice of models and modeling elicited ideas from many students

131 that a model as a type of thing, an object that replicates an object in the real world. Jorge mirrors the teacher’s language in his statement “In science we always use a model to understand what is happening” but without any explanation or indication he makes sense of the functional purpose of a model. I was surprised by the teacher’s emphasis on models as things despite our work in systems thinking that emphasizes inscriptional models s such as feedback loops, connection circles, and behavior over time graphs. We had repeatedly used these inscriptional tools to “make sense of” or

“explain” what might be going on in a natural system and looked at student work during professional development time. While Jorge did not use most varieties of systems dynamics models in his work, other students in the class did; many in their pre-assessments. Yet no students in the class identified these as types of models in the whole class discussion. Neither the teacher nor the students identified these forms of inscriptions as “models,” confirming findings from research studies discussed in the literature review.

Task 3: How does sun energy affect water in a bottle model?

Task and context

Between the decomposition bottles task and this one, students completed two Water unit activities on evaporation and phase change. For the next set of ecosystem investigations we re-used the decomposition bottles in an experimental activity designed to focus the students on the interaction of the sun’s energy and water in a closed and open ecosystem model (and by connection to the water cycle, to the Earth’s system). We also included a second experiment to help students recognize that there was air in the bottle by investigating how energy from the sun affected air temperature, again in a closed vs. open system. Each was planned as a jigsaw activity for two separate extended sessions, but the teacher’s absence on one day pushed both activities into one period. We also planned to use the energy of the late spring desert sunshine at noon. This was, of course, the only day of the semester it rained. We quickly gathered some desk lamps and attempted

132 to use the lamps to model the energy of the sun for both conditions, but there were not enough lamps, and their wattage was too low to generate enough light energy to affect the air temperature models in the time we had.

Interactions of child and task

Students did get some results with the open and closed models of energy and water interactions. Jorge correctly predicted that the water would evaporate in both the open and closed systems, but in the closed system “it can’t get out” and will “move to the top.” His observation confirmed his prediction: the water became “water vapor.” He did not complete the section of the data sheet asking him to explain what caused the change. He was “not sure” what would happen to water in a closed and open system in the shade (Notebook, 4-19-10).

Role of water

The activity may not have changed Jorge’s thinking about the effect of energy on the phases of water in the decomposition bottle, but he indicated that he at least knew that evaporation would take place in both systems, and the liquid would become “water vapor.” What the task was not able to help Jorge articulate was whether there was air inside the bottle and whether “water vapor” is a phenomenon of water condensing on the inside of the bottle or being distributed through the air in a gas state.

Immediately following the experiment, the class completed a second “mind map” assessment on the role of water in an ecosystem. The experiment had run over time, so the time for the assessment task was cut to just over 13 minutes. The lack of time and the teacher’s directions resulted in Jorge producing a product that reflected the instructions verbatim while missing the overarching idea of the role of water in an ecosystem.

T: OK…here’s our last task for the day….You’re going to work on your mind map for five

minutes. On your mind map, your big idea is ecosystems. I want to see ideas with a bubble


around it [illustrates on the board] and labeled connections. Put down the ideas and we’ll

link them together. The labels are really important. OK. What have you just been studying?

Can you use your notebook? Can you use the papers we just recently wrote on? [children nod, she nods]. Go.

T to Jorge: Now, you know a lot more, you remember a lot more about ecosystems…what are you

going to put on your map? OK, the ideas linked together are the important part, show how the ideas are linked together. Water cycle…water…she put water, what are you gonna put down?

What have we been studying?

J: Evaporation?

T: Yes. Well, what did we work on today?

J: Water.

T: Water, put down water. What else? Well, look in your notes, what did you write down today?

J: Fog.

T: Fog, put down fog, does that connect to water?...Water evaporates. [Talking with Jorge about three states of matter]…Air, put that down. But how are they connected by

evaporation? How did water become fog? You told me it evaporates, so you need to put evaporate on the line. I know I’m rushing you, put down your last best ideas, labeling your


Jorge’s resulting “mind map” is shown in Figure 10 below. The result of the close direction by the teacher combined with what is likely Jorge’s lack of mastery of the conventions of a concept map or mind map, is a work product that captures scientific vocabulary about the phases of water tacked onto a few ideas Jorge connects to the concept of “ecosystem.” The map suggests Jorge has a grasp of the definitions of the vocabulary and a notion that states of matter are somehow related to water and water phases, but only gas is connected to the idea of evaporation and the water cycle. He

is essentially following the teacher's directions to "put that down" on his paper. Neither of the chains related to water cycle or states of matter are connected to the chain with the sun and living organisms, despite having just had two direct experiences with modeled sunlight energy.


Figure 10. Jorge’s assessment work product for “What is the importance of water in an ecosystem?”

As Perkins (1986) and other researchers have noted, we cannot assume that what is on or not on his map reflects Jorge’s sense making about water in an ecosystem. Given just 13 minutes to complete his map, he put down what the teacher indicated she wanted to see. Because I was not able to interview him about the representation or his intentions, I can only speculate that he was focused on representing what he "knew" or remembered about the concepts of evaporation and states of matter taught in the days previous to this task, before we had begun the Bottle Biology project.


His representation reflects not only the teacher's directive coaching, but the curriculum itself, in which the water cycle is taught as a unit focused on water in the geophysical environment, while

"ecosystems" were taught in a separate unit at the beginning of the school year. At this point in the

Bottle Biology project, Jorge has had little opportunity to observe water in either real or modeled ecosystems. He has had only one opportunity to make sense of how water behaves in an ecosystem— either how it changes over time or how water causes changes in living organisms. He has begun to develop an understanding of air and water vapor as a "gas" but it is not clear he understands gas at molecular, or micro-level scale, nor how phase changes of water cause or correlate with other changes in an ecosystem. His main take-away sense is indicated by the two connections that [an] ecosystem “needs” the water cycle, and “needs” the three states of matter.

Practices of modeling

Any educator looking at this work without knowing the particulars of the context would have a difficult time inferring what Jorge was “doing” in this representation, or what sense he was making of either ecosystems or dynamics of the water cycle. Both a “concept map” (Novak &

Gowin, 1984), and a “mind map” (Buzan, 1996) are types of models that express relationships in terms of concepts and propositions that link concepts. Jorge appears to be unfamiliar with or have difficulties with the conventions of these forms of inscription, which presents a serious constraint to his expression of ideas. In this task, the teacher also is less concerned with how Jorge is actually thinking and more concerned with his producing a work product (in this case, an assessment artifact for my research). The goal of producing a product focused on the "right answers” short-circuited the goal of engaging students in modeling practices.

Likewise the students’ work with the physical model was constrained by the mangle of practice experienced everyday in classrooms: limited materials, uncooperative weather, lack of time.

While the processes of evaporation and condensation and phase change of water do play important

136 roles in the function of both organisms and ecosystems, this was not a successful use of either a physical or inscriptional model in our project to create an opportunity for students to make sense of these ideas.

Task 4: How Does Water Interact with Different Soils?

Task and context

This next task was designed to focus students’ observations on two abiotic interactions—soil and water. In ecosystems, the type of soil and the amount of water available are limiting factors that constrain what types of plant populations and plant communities will evolve, which in turn affects the communities of organisms in a given ecosystem.

This task was designed as reductionist and experimental, as opposed to a holistic investigation of the entire ecosystem. There were two parts to the investigation: an initial observation of the six soil samples to visually identify their properties, and then an investigation of soil-water interactions. The goals of the lesson were to focus students on variation in soil structures found in their local ecosystems, natural and human-amended, and the interaction of water with various soils. Students worked with six sample of soil from local ecosystems, including the local environmental center the children had visited, as well as from under a tree in my yard that included a great deal of detritus, and another soil sample from the class worm bin, which was very rich in organic matter and already quite moist. Students worked with their tables in a jigsaw structure in which each table would work with two soil samples, ensuring we would have experimental replication of all samples for comparison.. For the interaction investigation, students used soil samples in clear plastic nested cups. The top cup contained a soil sample and had a small hole in the bottom for the water to drain through into another plastic cup.

In her questions during the whole class discussion, the teacher focused on how fast the water moved through the soil and what color the water was afterwards. She also emphasized how

137 much water ‘is still in the soil,” which seemed to be a more difficult idea for many students to grasp.

Interactions of child and task

Jorge worked with his group at their science table. Each first created an “I Know/ I

Wonder” t-chart about soil in ecosystems in their notebooks. In this individual warm-up, Jorge makes statements and asks questions not just about soil, but about the interaction of soil, plants, and water, the subsystem he consistently refers to across the project. Jorge’s focus on these details and his notation about soil looking like his backyard suggests that he has a particular interest in the relationship of plants and soils, has some extended experience with growing plants, and “sees” this interaction of plant roots and soil as a salient and important structure-function in an ecosystem.

Jorge wrote: “Know: Plants need soil. Wet, brown, non-living. Smells good when wet. Wonder: How does it help the plant to grow?” Jorge’s first ideas about soil include the property “wet,” and an affective response: “smells good when wet.” I made a field note that it was interesting that a desertdwelling child would think of soil as wet, and wondered if his family gardened (He makes other statements and refers to prior life experiences further in the unit that support this inference).

Jorge and his table mates then observed each of the dry soil samples with a jewelers’ loupe.

He drew a second t-chart to record and compare the properties of his table’s two soil samples.

Comparing the properties of the two soils, he noted for Sample 3 (sample from a desert wash):

“Shell, acorn, desert sand, not dark, smelling like peanuts, rock.” For Sample 4 (soil from under the tree in my yard) he notes: “Looks like the forest and my backyard. Dark. Rock. Plant dead.”

These notations Jorge makes about soil are more detailed than those he made about the decomposition bottle, and he makes explicit connections to the task of designing a BioBottle that will support a plant. He knows that “plants need soil” and he wonders “how does it [soil] help the plant grow?” He connects his observations to life experience: “[It] looks like the forest and my backyard” for the one sample, names the second sample “desert sand,” and “knows” that wet soil

138 smells good. This attention to detail about soil and its role in helping plants grow again reflects his interests or prior knowledge.

In his prediction about what will happen when his group adds an equal amount of water to each soil sample, Jorge wrote: “When we add the same amount of H


O to both samples 3 and 4 I think it will become darker.” Due to his use of the deictic “it,” I cannot tell if he means the soil or the water will become darker. While this is all he writes, his conversation with me during the activity reveals he has some sophisticated ideas about the structure of soils that contribute to the interaction of soil with water.

In the video excerpt I talk with Jorge and his table mates during the experiment. I express surprise that almost no water has moved through the soil sample with detritus, and a girl explains they could not even get all the water in. In Figure 11 below, Jorge explains to me that in Soil Sample

3 (desert sand) “the water gets through because it [soil] has bigger pieces, they can’t get all smushed together, they’re too big, so there’s more room for the water to get through.” Jorge points to a sample at another table that is mostly clay and says “That one over there, it’s more compressed…[I ask, what do you mean by that?] He responds: “It sticks together, like clay. I think it’s Indian clay.”

Jorge’s confident demeanor in this exchange and the detailed explanation he offers about soil structure points to the synergy that occurs when a child’s interests and prior experience connect to and are elicited by an academic task. The task affords Jorge the opportunity to express his ideas in discourse with me and the teacher as well as his table mates. His reluctance to talk during whole class discussions, which shows up again at the end of this lesson, unfortunately results in his insights staying within his group and not being made available to the class at large.


Figure 11: Jorge explains how soil with bigger pieces lets the water drain through

On his observation sheet (Figure 12, below) Jorge responds to three questions the teacher writes on the board: How did the water interact with my soil? About how much water stayed in and about how much drained out? If there’s water at the bottom, what does it look like? The teacher gives them “30 seconds” to write the answers, which are the only notation in his notebook describing results from the experiment.


Figure 12: Jorge’s notes on results of the water-soil interaction

At the time Jorge writes these observations, he has been working with the soil samples for over 45 minutes, first examining the soil dry, using a jeweler’s loupe, and then conducting the water experiment. He has been fully engaged in the activity, and uses what he learned in part 1 (examining dry soil) to theorize about why water moves through soil with “bigger pieces and more spaces” more quickly than through soil with detritus “all smushed together.” The use of the jeweler’s loupe to

“zoom in” and observe the textures and contents of the different soil samples provided an affordance for thinking about how soil structures differ, and when the water is added, he can use his observations of the soil to explain the difference in how water interacts with each sample. He and his classmates are working with a visible, human scale component of an ecosystem, the soil, but at the same time, beginning to think about the structure of the soil, which is a phenomenon that is partly microscopic.

For Jorge, the task enabled him to focus on two critical abiotic components of the

141 ecosystem, and one interaction. This episode suggests that there may be a good pedagogical argument that not all investigations involving ecosystems need to be holistic, as long as the results of the investigation are linked back to the idea of a system. In this task, students also observed and explained the concept of “interaction” in more depth and more “authentically” in that the phenomenon they were observing was itself a process or behavior. The teacher did not need to prompt students to look for behavior or “use verbs;” these discourse practices emerged naturally through the affordances of the task and the focus of children’s attention on one type of interaction.

Role of water

Jorge’s phenomenological response to water in this activity is that it makes soil “smell good when wet.” He appears more interested in the structure of different soils and how water moves through different soils than about properties of water per se. He indicates that water may change color once it interacts with soil, and observes that water moves through different soils at different rates. He also implicitly expresses the idea that water changes the properties of soil in his comment about the soil sample possibly being Indian clay because it becomes sticky when wet. However, his first statement is that “plants need soil” followed by “how does it help the plant grow?” Jorge’s explanatory focus of the interaction is on the soil rather than the water. The task did not provide him opportunity to address his questions about how soil or water helps the plant grow, nor did it integrate what students had previously learned about the phase changes of water, especially evaporation and resulting humidity. Like most of the students in the class, Jorge cannot or does not express any ideas about how much water different types of soil can absorb, which has implications for the humidity of the ecosystem and the types of plants adapted to different soil-water conditions.

Yet in his end of the project interview, he expressed several ideas related to these structure-function relationships.


Practices of modeling

The model used in this investigation was designed to enable students to focus on one interaction among many in an ecosystem. Using an “open system” model for this task also allowed them to use all of their senses to explore the wet and dry soil; Jorge and several others continually touched or poked their wet soil samples to see if they felt “wet or dry.” Due to limited time available for the activity, the teacher and I limited inscriptional practices to written t-charts, which do not capture much useful information about the phenomenon. A more fruitful modeling practice might have been to encourage Jorge and students to create and share inscriptions modeling how soil properties influence the rate of water absorption and percolation, and to explain why water changed color after it percolated through. This also could have led to a bigger question that is current in ecological research of whether or not soil could be considered a system in itself, and the functions of soil as a subsystem in different ecosystems and its co-variation with different plant communities.

Jorge’s questions and observations during the investigation suggest he and other students would be interested in and developmentally ready for this next level of modeling an ecosystem.

Task 5: The Connection Circle

Task and context

This task was designed jointly by the teacher and myself to pull all of the structures and functions of the biotic and abiotic components of the ecosystem together in a single systems representation, with the focus on what they might need to put in the BioBottle model. The following transcription captures the teacher’s introduction and whole class discussion around the purpose of the task.

T: Today we’re going to shift from talking about parts to talking about the system. What’s the

difference, between parts and a heap and a system? Only Jorge has an idea…willing to take a

risk? Discuss at your table…[about 60 seconds go by].


T: OK [calls class to attention.] What’s the difference between parts and a system?

S: A heap is parts, a system is parts working together.

T: OK so think about an ecosystem. The parts that are working together, are they connected or

not connected? [1 second pause] They are connected. [Writes on board, circles the word “connected.”]. You know a systems tool that can help show that. This is a big fat

hint…Discuss at your table, what tool might you use to show they are connected?

Students at table: Jorge talking, gesturing, points to the Hoberman sphere on the bookshelf, S2 nods, makes a gesture with her hands as a sphere opening and closing and opening. Jorge and S2 explain and sign to S3: “Each of those white things, white points, is an idea, and the colored things connect the ideas. If one of the colored things breaks you can still trace around to get from one idea to another.”

T: [Brings class together]. Jorge, what did your table discuss?

J: [points to the sphere, says “the thing” and gestures with his hands to show the sphere growing bigger and smaller.]

T: This is a physical model. Can I draw a picture that shows…like this? What do I call that

picture, K—?

S5: A connection circle?

T: Yes! So we will be drawing a connection circle to model how the different parts of the ecosystem

in your bottle will be connected. OK? Help me come up with some nonliving components of the

ecosystem and some living components. 15 seconds at your table. Discuss.

The teacher proceeds to draw a big circle on a poster (see Figure 13 below). She labels the left upper corner “nonliving” and the right upper corner “living.” She then calls on various students to provide one living or non-living component that they might include in their BioBottle ecosystem,

144 emphasizing whether each is living or non-living. At the end of 30 minutes she has a complete circle with labeled components:

Figure 13: The teacher points at the whole-class-constructed Connection Circle

She then instructs each student to get a large piece of paper and reproduce the Connection

Circle as it is shown on the board. Their task is then to construct connections, or interactions, between the components, and label the connections. She has a student come up and model constructing an interaction: Carnivores eat herbivores. Students are given about 20 minutes to construct all the connections.

Interaction of child and task

Jorge’s finished work is shown in Figure 14 on the next page. A transcription of what he wrote follows underneath. The first challenge Jorge faced was that the whole class discussion resulted in the “living components” of the ecosystem being a combination of individual organisms


(bacteria, worms) and trophic level categories (herbivores, carnivores). The second challenge for him

(and readers of his work) was that he tended to start drawing a connection before he thought about what he wanted to connect to and why. For example, notice the line constructed between “secret ingredient” in the top left corner and “detritus” two steps counterclockwise. It looks as if he starts drawing the connection from “secret ingredient” toward the categories of living things, then changes his mind abruptly and without picking up the pencil moves the line over to “detritus” and labels the connection “breaks down.” He also appeared to soon forget the convention of using an arrow to show the direction of a connection, and frequently labeled these arrow lines backwards, as show on the lower right where he linked “plant” to “herbivores” with an arrow but then labels the connection “eats.”

Figure 14: Jorge’s finished Connection Circle work product


This task, which is meant to help students identify all the structures and functions in an ecosystem, overwhelms Jorge’s capacity to construct all the necessary connections, especially in the time allowed. He reverts to using simple verbs: “eats,” “helps,” “lives on.” Most of the connections he constructs among living components of the system relate to eating, and most of the connections from the abiotic to biotic components of the system refer in general to “needs” or “helps.”

Label upper left: "Nonliving"

Label upper right: "Living"

[Starting from top, moving counterclockwise]

"Secret ingredient"

"Air helps plant"

Detritus brakes down secret ingredient" OR "secret ingredient brakes down detritus"

"Water (arrow toward top of list of living things on right side of circle, labeled

"worm and"

"Water helps grow plant"

"Soil" (arrow toward plant and worms, no label on connection line)

"Energy helps save insects,"

"Energy helps detritus"

"Energy" [arrow points in general direction of plant(s), worms]

"Worms" [no outward connections to anything]

"Plant" [arrow to] "herbivores" labeled "eats"

"Insects" [long wandering line connects to] "omnivore," labeled "eats"

"Bacteria" [line to plant, labeled] "live"

"Bacteria [line connects to line from detritivores labeled] "eats detritus"


"Bacteria same (as) fungi same as omnivore"

"Carnivore eats herbivores" [line shows arrow from carnivore towards herbivore]

"Fungi same as bacteria"

"Fungi live in plants" [by inference from fungi same as bacteria]

"Omnivore same fungi" [I believe this means, bacteria, fungi are omnivores]

[Partial sentence on bottom of page] "Energy is the most important thing about eco…"

There are several possible explanations for the vagueness of connections Jorge represents.

One is that he may not have had enough time or metacognitive experience to systematically check each component and ensure he had made all the possible connections. It could be an effect of working memory, with Miller’s (19xx) “seven plus or minus two” as the number of elements people can manage to hold in working memory without using the strategy of “chunking.” A related constraint would be time to complete the task of drawing/labeling all the connections possible, let alone just one connection per component.

Another question Jorge’s work raises is whether and how the class or the curriculum unit defined the concept of “connection” or “interaction” in ecological terms of transforming energy, exchanging/cycling materials, or even more experientially in terms of communities, habitat, and niche. One clue can be found in the curriculum teacher’s guide, which says—

The term interact refers to one component of a system (an ecosystem in this case) acting upon, or influencing, another component of the system. For example, a pelican might interact with a fish or insect because it needs food. A pelican also might interact with a bushy plant because it needs shelter from the weather. For example, a deer interacts with water when it drinks, and a turtle, sunning itself on a rock, interacts with both the rock and the sun (BSCS, 1999/2006, p. 72)


It is only in the Lesson Extension, Information for the Teacher section that the text refers to the “result” of interactions as energy flow through food webs or materials cycling (p. 77). The language in the student reading emphasizes how animals meet their needs by interacting with the abiotic components of the ecosystem. Jorge’s resulting model reflects this language and emphasis on the interactions of living organisms with their environment.

Role of water

Jorge constructs two connections between water and living organisms: water helps plants grow, and a partially labeled arrow in the direction of animals that says “worm and,” which suggests in this context of what should be included in the BioBottle, Jorge considers water an important component but does not have the time to detail how it functions beyond meeting the needs of living things. More notable is that despite the investigation of the previous day, Jorge does not construct a connection between soil and water. He also does not connect water to air, although he has learned about evaporation and previously expressed that water evaporates into the air. The task constraints described above provide one explanation; another is considered below.

Practices of modeling

Jorge’s confidently expressed ideas about water and soil in the soil-water activity contrast with the those he was able to represent in this task, the Connection Circle, which occurred on the following day. Despite the task constraints, Jorge manages to construct all key connections needed for the design of his BioBottle. His vague labeling and lack of consistency using the conventions of the Connection Circle again suggest that students’ fluency with inscriptional forms constrain the ideas they are able to express. From a different perspective, Jorge may have been using a more sophisticated strategy of parsimony in representation, making just enough connections to express ideas he considered most salient.


Task 6: Designing and Observing a Sealed Ecosystem Model: The BioBottle

Task and context

A brief recap of the narrative frame of the project I presented to the class at the beginning of our work establishes the purpose and context of the project. Before this study, I had worked with this class as a volunteer during their experiences at the local environmental education center where most experienced a three-day program called Earthkeepers (Johnson & van Matre, 19xx). Jorge and the other fourth and fifth grade students knew me by my “nature name:” Otter, and that I was, like them, a fellow Earthkeeper. My introduction went something like this:

The reason I’m in your class and working with you for the next semester is because we are Earthkeepers. I am also a teacher and a scientist, and I want to understand how people learn about Earth’s ecosystems. I’m inviting you to join me in this research project as fellow Earthkeepers and scientists.

One of the problems we face in the desert is: Where does our food come from?

Growing food in the desert takes lots of energy and materials. A friend and fellow scientist I know was one of the people who lived inside the closed system of

Biosphere2 for two years. She designed the food ecosystems in Biosphere2 and had to figure out what kinds of food, how much of each kind, what energy and materials were needed. Some of her garden system worked, but parts of it didn’t. She learned a lot about how to grow plants in a sealed system. So that’s how I came up with this problem.

I want to know how students like you might think about a similar problem. We can’t build another Biosphere2, but we can make a model system in a soda bottle that might support a plant through its life cycle. The teacher and I would like to invite all of you to do this project and help us think about this.


On Day 6 of our working together, I briefly reminded students about the purpose of the

BioBottle design project. For the remainder of the project, students worked with a partner selected by the teacher. Their task was to design, observe, and explain changes to the BioBottle system and the plant inside. A constraint on their design was a lack of choice about how much energy their system would need, because the Wisconsin Fast Plants are genetically engineered to grow under fluorescent lights in laboratory conditions. The teacher pointed this out and showed students the light structures set up on the side of the room. The teacher explained all BioBottles would receive the same amount of energy from the lights. A second constraint was that the students could not collect any living things from local ecosystems except insects, an ethical and practical decision the teacher and I made. Students asked if they could use worms from the worm bin. After a brief class discussion about pro’s and con’s, the class agreed they could include worms.

Interactions of child and task

Designing the BioBottle. Jorge and his partner, Rosa, shared a table with another pair of students. They were “Team I” – a letter used to identify their bottle from the others. Jorge was the older student; Rosa was a third-grade student. They worked together to construct a design for their bottle. They also received a line drawing of the BioBottle set-up that was close to life size, as well as actual BioBottle parts so they could think about relative space and amounts of materials to include.

Figures 16 and 17 show their design plans.

Figure 16: Jorge and Rosa’s written plan for the design of their BioBottle model

Figure 17: Jorge and Rosa’s illustrated plan for the design of their BioBottle model



Their design is basic but complete: it includes the necessary biotic and abiotic components for an ecosystem other than energy, which was a given, and air, which was likely assumed since it would already be in the bottle and was not something they had to add. The first four items are

Jorge’s most salient ecosystem components: worm, water, soil, and the plant. “Scrap” refers to detritus or food waste as I had modeled in the decomposition bottles. They also wanted to include bugs or a bee, which Rosa explained to me was for pollinating the flower, a topic we had not discussed in class, and not what Jorge wrote on the paper (field notes, 4/27/10). I asked them what their plan was for collecting a bug or a bee in the next day before they built the bottles; they looked at each other and then crossed out this option.

Their rationales for including different components in the system show a varying amount of detail for explanation. Water and soil are included “to help them [plants] grow,” a reiteration of

Jorge’s explanatory labels in the Connection Circle task. The worm, which is first on the list, has the most detailed justification: “to make soil and room” [for plant roots] (field notes, 4/27/10). A detail in the illustrated plan shows “water vapor” at the top of the model system (Figure 4.x) written in

Jorge’s handwriting. This detail suggests Jorge is not distinguishing between a plan and a prediction, or perhaps an expectation that he will observe water vapor at the top once the system is closed.

Building and sealing the BioBottles. A whole-class discussion before the students build their bottles brings forward an essential ecological idea that also would influence the water and materials cycling: Whether or not the BioBottle should be a sealed system. Throughout the project the teacher and I emphasized that it would be a sealed system, but children did not appear to pay attention to this until they realized that they were about to take worms out of the worm bin and seal them inside a BioBottle.

As I talked with the children at Jorge’s table, a girl in the other pair asks “shouldn’t we put air holes in the bottle because we have living creatures in there?” The teacher revoices this problem

153 as “changing the structure of the system, which will change the behavior of the system. So discuss at your table why that would be a good thing or why that would be a bad thing.” Students discuss among themselves for about five minutes and then share out. Students who argue in favor of poking holes provide a similar argument: that the plants and other living things need air to breathe, and if the system is sealed they will die. Students who argue against poking holes give evidence that the decomposition bottles were sealed for weeks but insects stayed alive, or that the water would evaporate and the plants would die from lack of water. Only one student argues that plants “take in carbon dioxide.” Another boy adds that if plant health decreases, insect health decreases “because plants give off oxygen that the insects need.”

The teacher cuts the discussion short “because we need to build our bottles.” She leaves it up to the individual groups whether they choose to poke holes in the bottle or not. Jorge does not participate in this discussion, although he listens and pays attention throughout. In the end, Jorge and Rosa decide to seal their bottle and not poke holes. When I ask, they say they do not know if the worms will live or die, or if the plant will live or die, but they “hope it lives.”

Role of water

Initially, Jorge's understanding of the location and function of water in ecosystems was tacit and based primarily in lived experience, as evidenced by his pre-assessment naturalistic illustration of a backyard ecosystem with domestic animals, a tree with roots in the soil, and rain falling from the sky, perhaps soaking into the soil. Up to this point in the project, Jorge has had five opportunities to express and/or construct understanding of the role of water and the water cycle in the functioning of ecosystems. In the practice of these tasks, Jorge recognized water as a component of the decomposition bottle, as well as a substance that can transform from liquid to gas, and physically move to different parts of the model system. His explanation for the process of how water moves shows a tenuous grasp of the idea of evaporation: he uses this correct terminology to describe the

154 behavior of water in the open system, which is analogous to earlier investigations students did with open cups of water. He did not use the term "evaporate" for the closed system in either sunlight or shade; instead he predicted water will "move to" the top. His description of "water vapor" implies a change of state, but does not correctly explain that what he observed is the condensation of water vapor on the inner surface of the bottle. That he thinks water will change state under contrasting conditions of sunlight/no sunlight suggests he does not make the connection that the energy in a system determines the phase of water (solid, liquid, gas).

In the design and building of the bottles, Jorge does not express any ideas about water explicitly except that it is needed to help the plant grow. During the whole class discussion about whether to seal the bottles, the teacher asked the class whether the amount of liquid in the bottle would increase, decrease, or stay the same. The teacher emphasized they were not going to be opening the system to let water evaporate out, or to add new water. The class was divided in their opinions of whether the amount of liquid would change in a sealed system. One girl suggested that the amount of water might not change but the location of the water might change. Jorge did not participate in the discussion.

Practices of modeling

It appears as if Jorge and Rosa designed their physical microcosm (the BioBottle) based on everyday familiarity with gardens, soil, worms and plants, as well as shrewdly picking up on "clues" provided by myself in the introductory activity and discussion of the decomposition bottle. The focus of the task was to design a system that would support the growth of the plant, so their rationales focus on worms, water, and soil as "what the plant needs." The only mention of ecosystem is made in conjunction with including the "scrap" (detritus from the earlier bottles and/or bits of leaves and food each group chose to add). In contrast to other studies in the literature, Rosa and Jorge are not concerned with making a physical model that resembles their school garden, with

155 its plant stakes, plant labels, trash, or compost from the compost bin. They focus on designing the model for function: to support the life cycle of the Fast Plant. This mirrors findings from the literature that elementary students can and will focus on function over form if the academic task emphasizes this purpose. However this focus appears to be implicit; in their visual model of their plan they carefully label the parts but not any interactions.

Jorge and Rosa do not appear to have explicitly considered two "big ideas" of ecosystems

(materials cycles, food chains, energy flows) to justify what they include. (The design of the bottle task is also completed before the teacher presents a lesson on energy in ecosystems.) I was surprised, however, that the idea of a food web or even "what-eats-what" plays no role in their design planning or rationale, despite heavy emphasis on food webs in the ecosystems unit, and Jorge's illustration of the food-materials web-cycle in his first assessment, as well as his detailed explanations of how worms and insects eat detritus and improve the soil in his later interview. The likely explanation is that their design focused on supporting the plant’s growth, and the limitation of other organisms they could include focused their attention away from the idea of food webs.

The quantification of amounts of water, soil and scraps (abiotic components) turned out to be challenging not only to Jorge and Rosa, but many students in the class. They only had one experience of observing water-soil interactions, and the cups we used were not quantified measures.

Students did not have an opportunity to test out their ideas about the ratios of water and soil. They also lacked opportunity to observe Wisconsin Fast Plants grow in normal test conditions or altered experimental conditions before constructing an entire system. Yet all felt confident they knew what they wanted to include in their BioBottles, and could provide rationale for their design in basic terms of “need” or “help.”

The teacher addressed the quantification on the morning students assembled their bottles by leading a quick whole class discussion to give students a reality check about some of the amounts of

156 soil, water and detritus they planned to add to the BioBottle. She then gave students a chance to modify their design proposals before setting them loose with materials. Most students revised the proportion of water in the system, many lowering the amount of water they originally planned to add. Jorge and Rosa decided to increase the amount of water they added in their final plan, from ½ to 1 cup, to 1½ cups of soil.

Task 7: Observing the BioBottles

Task and context

In order to help students make more systematic observations and link those observations to the big ideas of energy flow, materials cycles, and interactions, The teacher and I designed a new data sheet that students would complete as they observed their bottles. The teacher led a whole class discussion emphasizing that the next phase of the project was to observe what changes in the

BioBottle systems. She introduces the idea of a variable as “something that changes” and draws ideas from students about what variables they can measure to observe change in the bottle and determine the health of the plant. Students were instructed to write their variables in their notebooks and discuss what tools they could use to measure change. Two popular responses are to measure the height of the plant and temperature inside the bottle. Several students point out that it will be difficult to measure plant height or temperature given that the bottles are now sealed. Students invent a variety of ways to measure height including taping a ruler to the outside of the bottle.

Another student suggests they can add a thermometer to one of the unsealed bottles, then seal it, to use as a comparison.

Interactions of child and task

In his notebook after the bottle is assembled, Jorge constructs two behavior-over-time graphs (BOTG, pronounced Bot-G) for number of worms and plant height. He does not include water in the list of variables he and Rosa will observe.


Jorge and Rosa recorded a total of six observations from the day they assembled the bottle

(4/30/10) to 5/19/10, a duration of 20 days, 15 of which were school days. The lack of daily observations and the lack of systematic observations was due to a decreased amount of time available for the project because the teacher had made a commitment to involve students in a social studies project with another class. It was also the final three weeks of the school year, with the usual interruptions of special visitors, assemblies and celebrations.

A sample of Jorge’s data organizer from Day 2 (actually, Day 4 of the system, the Monday after the bottles were assembled) is show below in Figure 18. Here he makes an observation that the water in the bottom well of the bottle decreased, and that a substance he labels “fungi” has appeared in the bottle on the surface of the soil/detritus mixture near the plant. His drawing is shown in

Figure 19. His visual notation on the data sheet is not that the water decreased one inch, but by three “tick lines” on the ruler, which is actually in centimeters.

Figure 18: Example of a data sheet from Jorge’s observation of the BioBottle Model on Day 4


Figure 19: Jorge’s first BioBottle observation drawing

Role of water

Figure 20 below summarizes the observation data sheets from Jorge and Rosa to show the consistent attention to water and the less consistent attention to other components of the ecosystem, including the plant. Note: at times both of them write their observations on the same sheet, and their observations do not always concur. Water is observed five of six times.









Where is the energy in the system? How are the living things in the system changing?

The plant grow. 1 ½ inch. Plant 6 leafs.

Water on the bottom.

Water vapor on the top




Plant is dead.

How are materials

(non-living things) in the system changing?

What interactions do you see or infer?

What else has changed since your last observation?

Water decrease from day

1 (11/2 inches) minus 3 centimeters

Water became dirty

Plant interact with water, soil, and sunlight.

Plant interact with water vapor

Blank Water [arrow pointing up and down] 1 inch 3 cm. Became more darker than the last.

Detritus 3 inch

Water still 11/2 – 3

So the rot is the same


I use a ruler and it is still

3 and a ½. It hasn’t change [at] all. 1 inch and 6 cm for the water.


Fungi kill plant? Dieing. Blank


A stem

I have fog on the top


Water is darker.

More fungi, like 10 more

Smell bad. On day one it smell good but now its more like rotten milk. We smell it.

Figure 20: Summary of Jorge and Rosa’s BioBottle observations

On 5/13/10 Jorge’s data notes include: “Water still 1/12” -3. So the rot is the same. Water is dark brown.” This is a new connection he constructs between the level of the water and the amount of rot. It suggests again that Jorge was listening closely to class discussions about how as the materials in the decomposition bottle rotted, the amount of water in the reservoir increased, an idea that the teacher re-emphasized during the discussion about whether to poke holes in the model and how that might affect the amount of water in the system.

His attention to the water level suggests Jorge was listening closely to the class discussion about whether to seal the system, and is now perceiving as salient how the water moves in a closed system, and whether the amount of water increases or decreases. On his observation drawing

(Figure 19 above) he has used a light scrolling scribble line to show something at the top of the

BioBottle. When I asked him what it represents, he tells me “water vapor.” I then ask him, “So where do you think this water vapor came from?” and he points to the water reservoir: “Down

160 here.” He then points to the observation data sheet and says “see, it [the water level in the reservoir] decreased” (field notes, 5-3-10). He also notes that the plant interacts with sunlight, soil, water, and

“water vapor” but does not provide any description or explanation about the processes of interaction.

Practices of modeling

The sample data sheet shown in Figure 4.13 is typical of all Jorge’s observations. He writes a few words and occasionally adds a diagram, like the physical measurement lines. He drew two

BOTGs for his prediction about the number of worms and height of the plant, but then abandons that form of inscription. The size and layout of the data sheet may have constrained his use of the graph and other forms of inscription. The language of the questions also implied that a short, correct answer is expected.

Interlude 1: Water Cycle Lesson

On May 5 and 6, the class’s work on the BioBottles was interrupted by a visit from a VIP systems thinker. The teacher, Ellie, had become well-known in the district for her use of systems thinking strategies in her classroom, and she chose a science lesson on energy interacting with the plant as the venue for this VIP’s visit. I mention this interlude for two reasons. One, this distracted the teacher, myself, and the class from the focus on the BioBottle as a system and how the system structure-functions all afford or constrain the growth of the plant. Two, we had originally planned to teach the plant-energy interaction and life cycle before the students constructed their BioBottle, but decided to swap the energy lesson to coincide with the visit. Based on my field notes and Jorge’s energy assessment artifact, the energy lesson and assessment appears to have shifted Jorge’s focus away from considering the role of energy in plant growth, because he made no more observations of the plant after the first two days until noting it was “dead” on the last observation.

An unexpected benefit happened on the following day, however, when I came into the

161 classroom to collect some unused equipment. I arrived right as the teacher was giving a lesson on the water cycle, which I did not know she had planned. I include this episode because the content of this lesson pertains to Jorge’s communication about the role of water and the water cycle in ecosystems in his later work.

The lesson had started when I enter the room. The teacher was working directly from the

FOSS Teacher’s Guide while she drew an illustration of the water cycle on a large sheet of poster paper. As I joined the class she had already talked about condensation and precipitation; these are illustrated and labeled on the poster along with what looks like a river flowing into a larger body of water. An arrow from the body of water points to a plant and soil showing roots. Another arrow points to the cartoon character “Woodstock” (see Figure 21 below).

Figure 21: Teacher Illustration of the Water Cycle during her lecture

She had just finished explaining that some of the precipitation goes into the ground, where it is

“taken up” by the roots of the plant. She then says:

T: It’s moved up through the plant and then, what’s that fancy word?

S1: Transpiration.

S2: Like transport.

T: What’s this word mean?

S3: How water moves through the plant.

T: Right. Then there’s a couple of choices. Woodstock could eat the plant [laughter]. If

Woodstock eats the plant, where does the water speck go?

S4: Woodstock.

T: Into Woodstock. And hopefully nobody eats Woodstock, or maybe they do, but let’s say

Woodstock lives a long and happy life, and at the end, Woodstock dies, where does the water


S4: Up.

T: Back into the system. Right?

S4: Wait, how—

T: —Let’s say Woodstock runs around and gets really sweaty. What do we call that? [gives further hints]

S5: Perspiration.


So maybe Woodstock could give off some water, or when dogs pant or when people talk…

Have you ever seen Jorge or [another student] playing basketball? Do they give off water?



Yeah! Which is good. If you couldn’t that would be bad, because when the perspiration evaporates that’s how you cool off. You might overheat and die. That’s one of the ways the

body helps cool itself.

[At this point the teacher picked up a small mirror and had several children breathe on it to show students that exhaled air contains water, which condenses on cold surfaces.]


So Woodstock can either get sweaty, or when he’s talking to Snoopy, water comes out of his

mouth. Then how does it move from here? Three people know? Oh come on!

S7: It goes into the ground?

T: Already went into the ground. When an animal sweats it out, what happens?

S8: It evaporates.

T: It evaporates. Remember that book you read, Did a Dinosaur Drink my Water? Yeah?

We don’t get more water from Walmart. All the water on Earth is already here. That’s.

All. We. Get. It’s held sometimes in icebergs, sometimes in the ocean, sometimes in the

aquifer, sometimes in the clouds, sometimes in the air, but it’s always moving.

The teacher moved on following the guide to compare the desert to the rainforest. She continued to ask closed-ended questions such as “Does the desert have more or less plants than the rainforest?” She eventually makes the point that the water cycle happens “more slowly” in the desert because there are fewer water specks and have a difficult time coming together to form clouds. She also asks students if they think the water cycle happens “faster or slower” in the desert or the rainforest, and tell them it happens more slowly in the desert. The duration of this lecture, investigation with the mirror, and class discussion was just under 8 minutes. She closed the lesson by having students jointly construct “the big idea in ten words.”


This example is not meant to critique the teacher, but to illustrate the way in which the water cycle continued to be taught in this class as a separate unit and a primarily geological phenomenon, even when our explicit goal was to merge this unit with the BioBottle project. The most significant point about this exchange is that neither the teacher nor the students make any references to their

BioBottle models, despite that several models are sitting on student’s tables. In the lesson, the teacher includes a plant and cartoon character in the water cycle and talks of water “passing through” the organism, but does not mention in her list of where water is stored that organisms retain a large percentage of water. Neither she nor students make the connection that energy from the sun plays a role in “moving” the water or transforming water from solid to liquid to gas. The lesson structure provided no opportunity for making connections about the role of water in ecosystems.

Task 8: Scientific Posters

Task and context

The class met to work on the BioBottle project again the following week. The teacher wanted students to produce a product that summarized their work on the BioBottle to show other students in the school and parents. We decided on having them create a scientific poster using a format from the Science Fair. Students worked on their poster drafts while they continued to observe their BioBottles, saving the conclusions section to write just before Family Fun night on

May 20. They had approximately 5 hours from the first “sloppy copy” to the final draft, which was not enough time for most students to review their data, synthesize their ideas, write their poster sections, and construct explanations for what changed in the BioBottle and why.

Interactions of child and task

The time and energy students put into their posters took away from their opportunity to observe, discuss, and reflect on their observations. On the other hand, the poster task did provide an

165 affordance for Jorge and the other students to organize and reflect on their experience, and pushed them to make sense of their observations and construct an explanation for a public audience.

Figures 22 and 23 on the following pages shows Jorge and Rosa’s finished poster. The handwriting appears to be Jorge’s. The introduction is short: "Class 5 is trying to grow a plant in a bottle and Elizabeth is seeing how class 5 learns. "They present the task as multilayered: what they are doing and what I am doing. This is the only place they mention my role in the project. Under

"Purpose" they write, "To see if we can build a ecosystem. We also we try to grow a plant in the bottle." The phrase "we also" suggests that they did not connect the task of growing a plant in a bottle as part of building an ecosystem. Yet in the conclusion section they write: "Ecosetem did not work [to] support the plant." This could be another artifact created by writing the two sections on two different days, and with a different level of attention to the purpose.

The results section is structured as both narrative (the plant died) and explanation (we think the fungi killed the plant). They begin by showing the increase and decrease of the plant

(presumably, the height of the plant, given that Jorge's data sheets record measurements. Their assertion that the plant died is supported by data of plant height shown as a BOTG. The assertion that the fungi killed the plant is accompanied by a series of five frames illustrating the plant at five points in time, but these illustrations do not support the assertion. The drawings seem instead to be a more concrete representation of how the plant looked on each of five days they observed. Since neither Jorge or Rosa have these illustrations on their data sheets or notebooks, these seem to be drawn from memory and have the purpose of emphasizing the progression of the plant's growth and “death.”


Figure 22: Jorge and Rosa’s Bottle Biology Poster

Their second claim is that the worm died. They support this with two BOTG's, one showing an increase in "bad smell" and the other showing increase in "dark water." They do not explain why the dark water might be an indication of the worm dying, or even if they connect the darkening water to the death of the worm. Again there seems to be a disconnect between the observations they made on their data sheets, the information in the poster, and later, Jorge's explanations in our interview. The annotated pictures of the bottle at the bottom of the poster were their idea of another way to present their data.

Their conclusion is straightforward: the ecosystem did not work to support the plant. They restate the claim that fungi killed the plant, but again give no evidence to support this. Their next

167 statement is that if they used compost tea, they would have a living plant. This statement is repeated again. They give no explanation of their thinking about why or how the compost tea would have made a difference and kept the plant alive. They do state this as tentative, "if we add compost tea maybe it will work." The last sentence in the conclusion makes little sense to me: "If we put the dirt on the bottom not the top becuze the plant grows on the top not the bottem."

No text on the poster refers explicitly to the "big ideas" of energy flow or materials cycles.

The only interaction is implied by the claim "the fungi killed the plants." They show change over time using three different BOTG's and one diagrammatic sequence. It makes sense that the poster emphasizes how different parts of the system changed over time, as that was the emphasis of the predictions and observations. That their conclusion is succinct and generally correct suggests that

Jorge at least understands the gist of the task and the role of an ecosystem working together to support the plant. They offer a reasonable explanation for why the plant died, supported by evidence of fungi (or mold) in the bottle in proximity to the plant.

The added-on methods section is show below. It appears that their original methods section was intended to be the illustration and text in the lower right corner of the poster listing four

“steps.” The redone methods section lists five components, described slightly differently than in their plan: “worms, rotten stuff, plant, water, soil.” Then there is a note, “We mix the [rotten} with the soil…now we don’t think it a good idea.” The sentences below the line were added after the teacher insisted that everyone include an explanation in their methods section of why they included each component. She gave them a sentence structure to complete: We added ______ because


Their rationales here are also slightly more detailed than their plan. Here they write “We added water because the plant and the worm needs water.”


Figure 23: Jorge and Rosa’s methods section of their poster

Role of water

Jorge and Rosa’s poster reflect the two main ideas in their design plan: that water was needed by the plant and by the worm. They also conjecture that perhaps adding compost tea might have helped, but are not able to explain why or how. Their reasoning may be based on the idea that compost tea might have been important because I used it in the decomposition bottle and because I

169 included it in the materials they could use to assemble their system. Jorge and Rosa’s poster only refers to the water getting darker, which they cite as evidence that the worm had died. They make no mention of the height of the water in the reservoir dropping from Observation 1 to Observation 2 or water vapor collecting at the top of the model. Jorge’s obvious interest in the amount of water decreasing in the reservoir and becoming water vapor at the top of the bottle during his observations is not included in the poster. This is probably explained by the focus of the poster task on describing the goal of the project to “make a plant grow in the bottle” and their success or lack of success regarding the goal. In that narrative, the height of the water or where the water is in the ecosystem is not a salient detail.

Practices of modeling

While introducing the poster task and basic structure, the teacher reminded students that they needed to show their ideas as “words, numbers, pictures.” Jorge and Rosa write about the

BioBottle model, and include pictures of the model with annotations, a new inscriptional strategy for them and the class. They use the data from their observation sheets to construct an accurate BOTG showing the height of the plant to illustrate growth and “death,” of the plant. Their claim about the death of the plant is also reinforced with a separate inscription the class called a “storyboard,” that shows “snapshots” of change at regular intervals of time. This is representation based on Tufte’s

(1990) “small multiples” strategy for illustrating processes of change. However, when these snapshots are compared to actual photos of the plant over time, this inscription is based on either

Jorge or Rosa’s idea about how the plant changed rather than on empirical data.

The second BOTG included is a linear model of increasing “smell” and increasing darkness of the water, which is not supported by data from their observations. In this classroom, qualitative

BOTGs are used as part of the practice of introducing systems thinking ideas to younger children as a way of expressing their ideas about possible directions of change without having to specify change

170 quantitatively. It appears Jorge and Rosa were using these qualitative models to support a conjecture rather than construct a scientific explanation. This suggests that classrooms where these inscriptional tools of systems thinking are taught will need to give children the opportunity to explicitly consider the role of qualitative graphs based on data versus the role of qualitative graphs to make conjectures or predictions.

As this was the students’ first experience with constructing scientific posters, it is not surprising that they have not mastered the conventions of constructing a scientific argument or explanation based on empirical data. Their use of multiple and creative forms of representation does suggest their readiness and interest in expanding their inscriptional repertoires in a science context.

Interlude 2: Setting Free the Worms

The day after Family Fun Night, the task was supposed to be for students to line up all 14 bottles, examine the designs and posters, and have a discussion about factors that may have played a role in the changes in their plant and BioBottle. It quickly became clear that students were exhausted from the previous night and most were too antsy to focus on the task. One girl argued that it was important that they open the bottles because if some of the worms had died as claimed, it was their responsibility to save the rest. At this point the teacher asked for a class vote. All but one girl wanted to open their bottles and “set free the worms.” No one expressed concern about the plants. In the last half of the session we went outside to the garden and helped students open the bottles. A few students planted their plants in the garden, but most were more intrigued by the fact that many bottles had far more worms at the end of the project than the number they had put in! “Setting free the worms” became the catch phrase for students, and this turned into a celebration of their learning and work, which was a lovely, if chaotic, ending for the project.


Task 9: Written Post-Assessment

Task and context

The last task that students completed for the project was to create a final “mind map.” I also conducted a post-project interview with Jorge for this study. The teacher introduced the postassessment as “show us evidence of what you’ve learned from the project.” She elicited ideas from students on what should be included. One student, not Jorge, mentions “water cycle.” The teacher responded: “Yeah. That better be on there because that’s a big part of what happened in our bottle.”

She also instructs them to include “energy flow and materials cycling in your bottle.”

Interactions of child and task

Jorge covered both sides of his paper with representations that were generally unconnected to each other, much like in his pre-assessment. I discuss the frames in which he mentions water or the water cycle. Figure 24 shows the main frame on his first page, a “mind-map” representation. The main idea, ‘Bottle Biology’ is linked to “Biosphere” which in turn is linked to the two ideas “living” and “non-living.” Water is linked to “non-living” and in two large red links to “plant” and “worms.”

While many of the living components are linked to one another, Jorge does not link an abiotic component to another abiotic component. This suggests he still “sees” the function of abiotic components in an ecosystem as supporting living organisms and does not see water as playing a role in other parts of the ecosystem.

Likewise, in his illustration of the BioBottle (Figure 25 below), he indicates water in blue located in two places: the bottom reservoir and the top without labels or explanation. He does not indicate water in the soil/detritus system, nor in the plant.


Figure 24: Jorge post-assessment mind-map


Figure 25: Jorge’s post-assessment representation of water in the BioBottle


In a third representation (Figure 26), Jorge illustrates from memory the life cycle diagram of a Fast Plant the students received when they were learning about their plants. Water plays an undefined role in the life cycle between the seed and root, but he does not “see” or know how to represent the role of water in the remaining stages of the plant’s life cycle, including transpiration.

We did not teach photosynthesis, so I would not expect Jorge to have included that process.

Figure 26: Jorge’s Post-Assessment diagram of the life cycle of a plant


Lastly, in a diagram of the water cycle (Figure 27 below), Jorge again produces the representation from memory (he did not take notes during the teacher’s lesson). Note that in this diagram, Jorge does not include any living organisms, although much of the teacher’s lesson focused on how living organisms are part of the water cycle. He also does not situate the water cycle inside the BioBottle or inside any bounded system; this is a diagram lost in space.

Figure 27: Jorge’s Post-Assessment Representation of the Water Cycle

Role of water

Across these inscriptions, Jorge expresses four ideas about the role of water. First, water is a non-living resource that is connected specifically to two living things: a worm and a plant. He does not connect water to the more general category of “living.” In the second representation, he illustrates that water was located in the basin and at the top of the BioBottle. He does not show water in the soil or in relation to the plant. In Figure 25, he indicates that water interacts in some way with a plant seed and root, perhaps suggesting with the arrow that the interaction of water with

176 the seed causes roots to grow. He does not indicate the role of water in any other stage of the plant’s life cycle. Lastly, Figure 27 appears to be a memorized re-presentation of a water cycle diagram from the teacher’s water cycle lesson, but Jorge omits living things.

These four representations not only fail to capture interactions taught in the project, such as soil and water, but also do not include the subsystem so salient to Jorge at the beginning of the project and evident through many of his representations: the mutual interaction of plant-roots-soilwater. He either does not perceive or does not know how to represent the movement of water in the

BioBottle nor through the structures of the plant. The last two diagrams appear to be reproduced from his memory and the water cycle diagram is likely included because of the teacher’s imperative that “it better be there.”

Practices of modeling

In Jorge’s mind-map (Figure 24), he makes a connection: that the BioBottle and the

Biosphere “are the same thing.” He is likely referring not to the Earth’s biosphere but to Biosphere

2, the self-contained model that I used as an example to help frame the project of modeling, and that we also visited the day before he completed this task. I regret not having the opportunity to have interviewed him about this representation to better understand what he meant. In the rest of the mind-map, Jorge still struggles with some of the representational conventions while continuing to use one of his own invented convention: that a connection line may link to another connection line as well as to a concept “bubble.” He has also added the use of color and a key to distinguish between non-living and living components of an ecosystem, as well as to indicate the cycle of materials returning from living to non-living, an idea carried through from his pre-assessment.

The two models re-presented from memory (plant life cycle and the water cycle) suggest the challenges of introducing students to pre-constructed models that are to be taken-as-given. Neither of these accurately express ideas that Jorge has expressed elsewhere or that he would express in his

177 final interview. His misuse of each representation echoes difficulties described in the literature where children can reproduce or correctly “read” representations such as a food web but cannot use the representation as a model to support their reasoning about form and function.

What is most striking about Jorge’s final assessment is that he chooses to construct a “heap” of unrelated representations that express ideas inconsistent with and generally less detailed than those he has expressed in other contexts. This supports findings from other studies that caution not to include models or representations in science instruction for the sake of reproducing canonical knowledge, or with an emphasis on quantity over quality.

Summary Across Tasks

The above nine examples of the interaction of Jorge with tasks in this project and his resulting sense making (or lack of sense making) about the role of water provide evidence for how the context or situation created by the task is negotiated by a child who constructs meaning not only through content presented by the curriculum, but also his life experience, prior knowledge, curiosity, and interest in particular interactions of an ecosystem (e.g., soil and plants).

In each task, the role of water in organisms or the ecosystem was foregrounded or backgrounded by the curriculum design (not all tasks included water). It was also brought in or out of focus by the curriculum enactment. For example, in the decomposition bottle, there were opportunities for students to notice and investigate the increase in water as the vegetation decomposed, and while Jorge did notice water in the system, the fascination and “gross factor” of opening and exploring the bottles brought other aspects of decomposition to the foreground. The task exploring interaction of different soils with water engaged Jorge and gave him an opportunity to connect his knowledge and observations of the different soil structures to explain why water might drain through more quickly or slowly. Yet the next day, he failed to make this interaction a connection on this “Connection Circle” diagram.


Jorge’s consistent yet vague connection that “water helps the plant grow” is correct but not necessarily scientifically correct; it is a “black box” connection in terms of an abiotic component of an ecosystem providing a resource for a biotic component. Connections are the most important part of any system, as the teacher emphasized, but it is not clear from his work in the project whether

Jorge has made sense of enough of the relevant connections to successfully design, observe, and explain changes to a BioBottle or describe the role of water in an ecosystem or ecosystem model.

Fortunately, the opportunities I had to talk with Jorge during the project and in the postproject interview provide evidence that Jorge has perceived and made sense of more about the role of water in an ecosystem than were apparent in his class participation and work products. The next section of this chapter analyzes how he expressed those ideas in our one-on-one interview, and as with the lesson tasks, how he expresses different ideas about the role of water in ecosystems when he is working with the BioBottle versus reasoning from experience.

Water in Experience, Water in a Model System

Water in Experience

From a number of things he says during class and in our interview, Jorge uses his experiential knowledge about growing plants and about our local ecosystem to reason about the interaction of water and plants. For example, when I ask him in our interview to imagine what his

BioBottle model might look like a year from now, he imagines the plant would keep growing and the roots would be all bunched up "Like if you go to some stores and like they [the roots] come out of the...black...things from the bottom, for the root, when you take them out they're all like [makes bunched up gesture with his fist] (Interview lines 599-603).

“The roots suck up the water”

In their design of the bottle model, Jorge and Rosa try to estimate how much water they should add to their model. They settle on a ratio of 1 cup of water to 1.5 cups of soil, with no

179 rationale other than each with "help them [plant] grow" (Jorge and Rosa BioBottle Design Proposal,

4-27-10). I originally concluded from my analysis of his talk and representations during project tasks that this idea of water “helps the plant grow” was a “black box” explanation of how biotic organisms require abiotic resources. But during our interview, when I ask Jorge if there is any other way that water can move through the system, Jorge responds by describing the flow of water through the plant using everyday language (Interview, lines 321-333):


OK. And is there any other way that the water can move through the system? Any other

place that you might find water in the system?

J: The roots.

E: In the roots.

J: They suck it up and then um, those somethings, I don’t I don’t know what they throw out, but they do throw out from the leaves something?


They throw out something from their leaves. That’s actually water vapor they throw out from

their leaves. They have pores.

J: I think that’s how they got out.

E: Do you remember the word for that? We talked about it last week.

J: [2 sec pause] No, I don’t remember.

E: Human beings perspire…and plants transpire?

J: I uh [pause] I forgot.

"This is a moister area plant"

Jorge’s understanding about the role of water in ecosystems becomes even more evident when I invite him to imagine designing another BioBottle for a more familiar kind of plant

(Interview lines 658-686):

E: OK. So what if I asked you to design another BioBottle, but this time we’re going to put a

desert plant in there and try and try to help a desert plant, like a cactus, to grow. What would you do differently? What would you put in there?

J: Ok, well I would, I would really add more water? Uh maybe? And then um, put in a different kind of soil/

E: Ok, you would add /more? Water?

J: Maybe. A little bit more.

E: Maybe a little bit more. OK

J: No, I don’t think we add more water, cause like here in [our desert], there’s not a lot of water/

E: There’s not a lot of water—

J: —but it depends on the, it depends the on the—what kind of plant. Cause you know the cactus stores water, inside the cactus?

E: Yes it does

J: And then I think I might stores it, for like a month/

E: Oh, it might store it for a month. Let’s…are you thinking of the saguaro? The big cactus with the arms?

J: Yeah. So um, yeah, I thought it depends…yeah, if it was a saguaro cactus, you would need a bigger bottle?

E: You’d need a bigger bottle, absolutely.

J: Yeah. And put in less water.

E: Less water.

J: Cause this [points to drawing of plant in the bottle model] is pretty much like a, like a moister/ area plant?

E: A…moister? area? Is that what you said?



J: Yeah.

Jorge not only brings in his knowledge of the desert plants in our area, he also recognizes that different ecosystems have different amounts of water available, and different species of plants are adapted to different amounts of water in the soil. When he talks about the cactus storing water, he offers a particular and accurate detail about water interacting with organisms. He also uses his knowledge of the desert and familiarity with desert plants to express another important ecological idea: different plants are adapted to different amounts of water in an ecosystem.? He makes the connection through a lesson the teacher had recently taught in social studies that I had observed

(Interview, lines 756-786):

E: Ok. Ok. And you, you said you’d put sandy soil in? Why would you put sandy soil?

J: Cause if I put like a soil from like a moisty area?

E: Uh huh? From a moister area…

J: Cause do you know how much the saguaro holds the water?

E: The saguaro holds the water, uh huh

J: Cause I was thinkin’ if you put like the saguaro in the soil we were using/ which was from a moisty area, I was thinking the cactus would die cause like that’s too much water.

And if we put sandy, that would be pretty much its normal life cycle.

E: Okay, so, so sandy soil would hold less water,

J: Like, can I make a connection?

E: Yeah, sure!

J: Um, like, we were learning about uh, um, this explorer, ca—uh, I forget his name? But he wanted to explore the desert. He thought the desert [was] more like laaakes, like not cactus but like [forest], or like…he thought it there was more like, like over there, like



E: Ooh, the explorer/! Yes. What was his name?...Well, I know who you’re talking about. So he was exploring and he thought the desert was more like the forest in England.

J: And then he came here and he didn’t, he wasn’t ready for like, he didn’t know what the water was like less, so they weren’t, used to it?

E: Right…

J: —they weren’t used to it. So like the cactus is not really used to the soil.

E: Oh, so a cactus would not be used to the soil, that a, a garden plant would grow in.

J: And then if we had that soil, maybe later on?

E: Uh huh

J: Like maybe later on, I don’t know but maybe like nature might change the cactus? For like less water storage?

E: Ok, so the cactus might change over time and evolve to a kind of cactus that doesn’t store as much water, uh huh…

J: —yeah.


“My cousin said, ‘looook, it's a rock.’ I said ‘no, it's my mango.’”

In contrast to the detailed domain knowledge Jorge demonstrates in the above explanations about water-plant interactions, there are other experiential contexts in which he does not make the connection that water is one of the structures in all living organisms.

For example, in our interview, I asked Jorge a question from an earlier study conducted by

Leach et al (1996a), an apple-in-a-sealed system "thought experiment." (See below for the details). In response, Jorge tries to make sense of what changes might happen to the apple in a closed system.

He makes another "connection" to his lived experience, a mango he threw away into the backyard

(interview, lines 342 to 405):

E: OK, this is what we call a “thought experiment.” We’re not going to do it, but I’m going to ask you to think about it? So imagine that we built a new bottle, and we just put soil on the bottom, and I put just an apple on the soil. And we sealed it all up, so you know, it’s closed, it’s a closed system. And it sits there for two years. Soooo, uh, when I put the apple in there, we sealed the bottle, and I weighed the bottle, we put it on a scale. And then two years later I weigh the bottle again. It’s been sealed the whole time. How much do you think the bottle weighs? At the time we weigh it the second time, will it weigh more or less or the same?

J: So can I start from…from the first weighing, can it be like two pounds?

E: Yeah, let’s say it weighs two pounds—

J: I mean—

E: —when we first weigh it

J: —two years later, it might be………uh…I think it depends on the weather?

E: Depends on the weather, ok…

J: I was thinking like if it was summer, the sun was hot, and then there’s a lot of energy flowed from the sun/

E: Ok, a lot of energy from the sun,

J: And I think it makes [more] less

E: And then it reaches? I’m sorry?

J: I think it will make more less. It weighs less.

E: Ok,

J: Less than a pound.

E: Less than a pound? OK, so it will lose/ weight in the summer. OK. And where do you think that weight goes?

J: The weight? [pause] So this thing [shrinks] [long pause] So this thing shrinks…

E: The soil? [I meant to say the apple shrinks, and Jorge repeats what I said without thinking]

J: The soil shrinks—

E: —the soil shrinks

J: so only if you put soil and an apple…

E: Yeah, put an apple on top of the soil

J: I think the mineral gets out.

E: You think the minerals get out. OK.

J: I’m not sure but I think the mineral gets out. That’s what I’m thinkin’.

E: OK, so more minerals get out, especially if there’s more energy from the sun, more minerals are going to get out? And then it weighs less?

J: Yes.

E: OK. Um, ….OK. That’s interesting! OK, ummmm. So we’re going to talk a little bit more about that, too, after we go to Biosphere. Let me ask you, how do you think the apple might change over those two years? If you can imagine—

J: Is the apple more like a ball, or slices?



E: Um, a whole ball. Lets’ imagine we just put a whole apple in there.

J: Um, OK, let me say that it might shrink.

E: It might shrink? OK.

J: Cause I think I have an experience where I experienced it before?


J: Cause I think…one time…I had a… mango

E: A mango, uhuh

J: I think I had a mango, so then I um, I look at it, cause it had this ah, mold? So—

E: It had mold on it, uhuh—

J: —So then the mold is not really good, I guess, well, we eat a mold?

E: Yeah, we eat some molds

J: Some. So I was like, errr. But that was like when I was uh, seven, eight I guess, so then I left it outside, and then I forgot about it and later on I came out/ like that one I think three days later/ I came out/

E: Ok

J: It wasn’t sealed, it was open, and then like my cousin came, any my cousin said like “loooook, I think it’s a rock,” and I said “its not, it’s my mango, I left it outside and it shrunk.

E: It shrunk way down. In an open system it shrunk way down. OK, and what do you think happened when it shrank…where did the stuff that used to be in it go?

J: Down to the detritus.

In this explanation of what happens to the theoretical apple and his actual mango, Jorge describes the condition of the fruit as "it shrinks" and his explanation is "the minerals get out."

While later he explores what he might mean by "minerals" and states that "water has minerals," he

186 does not seem to understand that the majority of the mass of the fresh fruit is water, which evaporates into the air in both the closed and open systems. Even though he indicates he understands the system that the apple was in was "closed" and the mango system was "open," he claims that the system with the apple in it would weigh less over time. He does however tie the change of the fruit to the amount of energy coming from the sun, which may suggest he is tacitly including his emerging ideas about evaporation, but cannot express them in scientific terms.

A challenge in analyzing these data is Jorge's use of the deictic, "it" to refer to both the apple/mango and the system/event of change: "I think it will make more less. It weighs less."

Making a connection from the closed system with the apple to the experience with the mango may actually further confuse his thinking, because he has shifted the context of the problem to the earth system, where it would be true that the mango "shrinks" and therefore weighs less due to water loss, and he has no way of calculating the weight of the entire earth system, or tracing the evaporation of the water from the mango into the atmosphere.

Water in a Model System

"I'm guessing like two drops of water…”

The activities in the FOSS Water Cycle kit focus primarily on water in an open system, that is, one that has no visible boundaries for the children to perceive. However, in a two-liter closed

BioBottle model, the water cycle is made visible and salient to Jorge in ways that either increase his understandings about the water cycle or enable him to express his understandings through talk and drawing. Figure 28 below comes from Jorge's representation of his and Rosa's sketch of what they plan their BioBottle will contain. After all of the "ingredients" of the ecosystem were added and arranged by the children, each group sealed their bottle system with clear mailing tape around the circumference at the connection where the "funnel" bottle meets the "base bottle." This created a materially closed system, one in which all materials must cycle within the boundaries of the bottle.


Figure 28: Jorge and Rosa’s design of the closed system BioBottle

As Jorge's illustration shows, the funnel design of the model allows excess water to both drain out of the soil into the reservoir underneath at the bottom, and also allows evaporated water to collect as vapor or condensation at the top of the bottle. The bottle "cap" at the bottom of the funnel actually has holes in it covered by a net and connected to the reservoir with a string, to allow water to drain out or be wicked up into the soil from the reservoir. I begin by asking him to talk more about the nonliving parts of the system and he offers a fairly complex explanation/think aloud about how water "moves" in the bottle system (Interview, lines 268-306):

E: OK, um…so let’s talk a little bit more about the water, and the nonliving parts of the system.

What, what are the nonliving parts that you know are in your system?—

J: —water, soil, [every] thing…yeah. Water and soil.

E: Water and soil, …[encouraging]

J: And then detritus.

E: And detritus, right, that’s stuff from…that’s dead bodies/

J: And then this thing [water vapor] exactly can’t connect into this water

[condensation at the top of the bottle]…the water vapor [now recalling the term he wants to use to indicate water condensate]?

E: Yeah…?

J: Cannot get to the water [in the reservoir]. Um,

E: You don’t think this, the water vapor [at the top], can get to the water [in the reservoir]?

J: Nah, I don’t think so.

E: No?

J: No, I think unless if you shake it, you know, when you shake the water vapor, um…yeah, when you shake it, they uh, come together in uh, groups/ of j—of water?

E: Um, oh, ok—

J: —Water drops?

E: Right—…it turns into water drops/

J: and I then I think it goes down, I don’t think it goes down, I don’t think its going to go completely a hundred percent down but maybe its these right here, um…[pause]

E: ok so you think when the water the conden—, when this condenses, it falls onto the soil, then it mostly stays in the soil, so where did the water down here [below the soil] come from?

J: [Pause] We put the water in there, right?



E: Yeah. So you think that that’s when you put the water in first, it poured through?

J: Yeah.

E: OK, and can you tell me about how the water, ah, does any of this water [in the reservoir] get up to the air?

J: The only, like, I’m guessing like two drops of water.

E: Two drops of water! OK...

J: Cause the the, the um, I wish I brought my bottle, but um but on the cap there are like that, that much [holds fingers about half a centimeter apart]

E: OK, so on the cap, there was like just a little patch of water vapor…

J: No I like pretty much just estimated like two drops of water

E: OK, cause that’s what you saw?

J: That’s what I was [thinkin’ of]

E: Ok. Ok, so you have your evidence for why you think just a very little bit.

J: Yeah, and um, …what was the question again?

E: How did the water—did any water get from here up to here, and if so, how?

J: I think that when it dried out, I think it have like, it opens spaces and the water vapor came out in the top.

Later in the interview Jorge returns to the idea that the amount of water that decreased in the bottle reservoir under the soil is equal to the amount of water vapor that increased at the top of the bottle (Interview, lines 710 - 730):

J: You know, I’m not sure how much our water really increased, I mean, decreased.

E: No?

J: No, it only decreased by, 3 centimeters.

E: When you say your water decreased by three centimeters, can you show me? What you mean?


Do you mean it decreased here [in the reservoir]?

J: —Yes, the water right here. It didn’t, it didn’t go up, it went down from three centimeters.

E: Oh, ok, so, it started at a certain level, and then it dropped three centimeters. Where do you think that water went?

J: Water vapor.

E: Water vapor.

J: I think, no actually, so like…[goes quiet…drawing] a drop of this

……………maybe it was… .. … … … … … … … I think [might be like] like two drops of water vapor?

E: Uh huh

J: I thinks its…three right now, cause it, I just [did it], cause um, one drop, it goes like, no. One [unintelligible]… it goes… …

E: One centimeter? OK, if one drop of water equals one centimeter, and it drops by three—

J: —that’s what I was thinkin of—

E: —centimeters, then you have three drops of water! OK! So you’re doing the math on that. Ok.

So…you think three centimeters went up into the water vapor, or into three drops—

J: —yes.

E: Ok.

In Figure 29 below, Jorge drew an illustration while he was talking with me about the water vapor. Note on the lower right the notation he made, which is what I am reading in the second to last line of the transcript.

Jorge is engaged in trying to make sense of where the water in the bottle was when they first closed the system, and now three weeks later. He grapples with sorting out where the water vapor or

191 condensation at the top of the bottle came from—the water held in the moist soil, or from the water in the reservoir below. Yet he has a very interesting idea, one that came up previously in a conversation I had with him in the classroom on a day he was making observations: he conjectures there is a quantifiable relationship between the amount of water in the reservoir and the amount of water vapor at the top: "I'm guessing like two drops of water."

Figure 29: Jorge’s explanation of how water is conserved in the closed BioBottle


What I find interesting about this exchange and Jorge's idea is his explanation that three centimeters of water in the reservoir might equal three drops of water condensation at the top of the bottle. The notion of equivalence suggests that he may be constructing an understanding of conservation of matter because he knows the system is closed. He refers often to the "cycle of materials" throughout the project. Thus he observes (or imagines) there must be a steady amount of water in the system and quantifiable relationship between the amounts of water in different parts of the system.

Jorge is able to make this connection because the physical boundary of the system is concrete and tangible as the edge of the bottle he can see, and the tape he and his partner used to seal it. As he indicated in the earlier activity with how energy interacts with water and air in the bottle, Jorge understands that water in an open container will evaporate into the air outside the container, and that if the container is closed, all water must stay in the bottle somewhere.

Observations of his and other BioBottle models over time enable him to solidify this understanding and even to construct a quantitative explanation of how water moves from one reservoir in an ecosystem to another. His explanation that water is conserved in a closed system may set the stage for later understanding that all matter within the earth system is conserved, including molecular scale compounds such as carbon and carbon dioxide that affect not only an ecosystem but the earth's climate.


The contrast of Jorge's thinking about the water in the bottle system being conserved, and the weight of the sealed apple system not being conserved raises many questions for educators, as understanding of the conservation of matter and energy in the universe is considered to be one of the pinnacles of scientific or "model-based" reasoning, and one that research has found not many children or adults can grasp (cf. Windshitl & Thompson, 2013).


In the field of systems dynamics, criteria for modeling a system or a subsystem are that the model must be internally consistent, reflect data from current scientific understandings, and include all of the important components and specify interactions among the components (Few, 1996). These criteria align with recent theory and research on the highest level of a learning progression for water in socio-ecological systems, descried as "qualitative model-based accounts [that] include driving forces and constraining factors to explain or predict where water and substances in water move in given situations" (Gunckel, Covitt, Salinas, and Anderson, 2012, p. 843). Gunckel et al. (p. 849) specify criteria or elements for model-based reasoning (understanding) about water as follows:

1. Recognize and explain how water and the substances it carries can move along multiple pathways through connected systems;

2. Describe the movement of water at multiple scales, from atomic-molecular (micro) to landscape (macro);

3. Explain the behavior of water using scientific principles such as gravity, pressure, and conservation of matter;

4. Use representations as models to reason about locations and flows of water through the environment;

5) Recognize human dependency on environmental systems to provide fresh (potable) water.

Jorge is able to reason about water in the bottle system to in ways that meet criteria 1-4 above. As noted by learning progressions researchers such as Gunckel et al. (2012), the development of a learning progression can take multiple forms; levels of achievement are defined partly from theories of human development and cognition, partly from empirical research data. There is still much to be learned about the "messy middle" (Gotwals & Songer, 2010) of learning progressions— the types of understandings students use that fall between the lower anchor of everyday life knowledge children bring to school and the "upper anchor" of a scientifically reasoning citizen.

The findings discussed in this chapter suggest that the task of thinking about and observing the function and behavior of water in the Bottle Model provided an affordance for Jorge to partially

194 integrate the discrete learning outcomes of the Water Cycle activities into a more coherent and model-based account of water in an ecosystem.

In contrast, Jorge’s perception, talk, and representations suggest that the role of water and the water cycle comes in and out of focus as he grapples with how to integrate the multiple demands of the expectations of the academic tasks, the teachers’ directions, his own perceptual attunements, interests, and agency, and the complexity of ecosystem structures, functions, and behaviors. In the pre-assessment, Jorge’s first representation emphasizes water in the form of rain, falling to earth and soaking into the soil where the tree roots grow and worms live. This configuration of plant-rootssoil-water emerges as a theme in his design of the BioBottle, and his narrative explanations of where the water is in the BioBottle system in our interview. Yet he does not perceive that the fresh plants placed in the decomposition bottle contain water, and that this water becomes liquid and mixes with other liquids released from components in the decomposition bottle over time. Perhaps he is distracted by or more interested in the “gross” factor, or takes for granted that there would be water in the system. In the water-soil interaction investigation, Jorge offers a fairly sophisticated and semimicroscopic explanation about how spaces in different soils affect the movement of water through the soil. The following day in his Connection Circle, he does not indicate this as an interaction in an ecosystem. Finally, he not only attunes to the movement of water in the BioBottle system, he quantifies it in a way that suggests he has a notion about conservation of water in a closed system.

However, when reasoning about water in the experienced Earth system, Jorge does not consider that water might be part of the mango that evaporated into the atmosphere. What Jorge is able to “see” about the role of water and the water cycle in ecosystems is itself a emergent property of a complex, dynamic process as he interacts with the system of task, tools, classmates, and researcher.

In Chapter 5, I will consider the implications of these serendipities and inconsistencies for research, curriculum and instruction.



Ada told Ruby that she envied her knowledge of how the world runs. Farming, cookery, wild lore. How do you come to know such things? Ada had asked. Ruby said she had learned about what little she knew in the usual way. A lot of it was grandmother knowledge, got from wandering around the settlement talking to any old woman who would talk back, watching them work and asking questions….Partly, though, she claimed she had just puzzled out in her own mind how the world’s logic works. It was mostly a matter of being attentive.

—You commence by trying to see what likes what, Ruby said. Which Ada interpreted to mean, Observe and understand the workings of affinity in nature.

Charles Frazier, Cold Mountain (1997, pp. 137-138)

Chapter 4 provided an analysis of how Jorge interacted with the tasks of the curriculum to make sense of the role of water and the water cycle in both the experienced world and a model

BioBottle system. This chapter synthesizes the findings in Chapter 4 in light of the conceptual framework to address the research questions. I discuss how the findings of this study relate to or extend the existing literature, and discuss implications of the study in light of the problem framed.

Implications of this study and future research questions are then addressed. Lastly, I reflect on the research process and the limitations of the study, and close with Jorge’s perspective on the experience.

Research Findings

The questions guiding this research were: “Within the context of a culminating class project to design a functioning closed ecosystem, how did the interactions between a learner, the academic tasks, and modeling practices influence what he perceived and communicated about the structures

196 and functions of an ecosystem? In particular, how did this student perceive and communicate his sense of the form and function of water in the microcosm and the interaction of water with other living and non-living components of an ecosystem?

Interactions of Learner, Academic Task, and Modeling Practices

In this section I address the first part of the research question. The conceptual framework outlined in Chapter 1 outlines tenets of situated learning, what Greeno terms “situativity” (1997,

1998), with a focus on the affordances and constraints of both physical and inscriptional models in the context of academic tasks presented by the curriculum as enacted. As posited by Greeno (1998), the frame of situativity theory emphasizes that the agency of both teachers and learners leads to an

“emergent” situation, or ill-structured problem space and shifting of goals. This assertion is supported by Doyle’s (1983) theory of academic task, and Doyle’s argument that curriculum is never a static entity but is always “curriculum in motion” as it is enacted by a classroom of teacher and students through contextually-influenced pedagogical moves and negotiation of work and goals

(Doyle, 1992). The BioBottle project followed this dynamic, adaptive course at the level of curriculum and simultaneously at the level of the learner.

The original goal of the BioBottle project was to design a thought-revealing activity (Lesh,

Hoover, Hole, Kelly & Post, 2006) that would create a situation and affordances for Jorge and the students in his class to bring his prior knowledge and experience of ecological concepts and ecosystems to the design, observation, and explanation of a model ecosystem. As explained in

Chapter 3, the project goal itself shifted to accommodate integration of the FOSS Water curriculum so that the teacher could meet her yearly science standards. This in turn shaped and changed the purposes and goals of both individual lessons and the BioBottle modeling project.

Throughout this 12-week study, Jorge encountered the curriculum “in motion” (Doyle,

1992), in situated contexts that addressed domain knowledge of ecosystem structures and functions,

197 materials, through representational strategies and models designed to afford students’ engagement with the structures and functions of ecology and systems. In the ten learning tasks analyzed in

Chapter 4, as well as in his interviews, Jorge expresses ideas about ecosystems that varied in level of detail, were sometimes related to the ecological investigations of the BioBottle and sometimes not, and contradictory depending on whether he was explaining ideas about water in the context of the

BioBottle or drawing from life experiences. Jorge and all of us traveled a dynamic trajectory of learning within the real-world constraints of an urban elementary school classroom.

For example, in the soil-water activity (Task 5), Jorge shares with me and his table mates a very detailed explanation about the structure of different types of soils, and how those different structures affect the rate that water percolates through the soil. He uses a laboratory tool, the jeweler’s loupe, to examine dry soil carefully and identify variation in properties. After conducting the water investigation, he shares further his ideas about how water affects the properties of different soils. Yet the next day, during the Connection Circle modeling task, he does not construct a connection between soil and water. He reverts to expressing an idea he knows well: plants and worms need water. The academic task frame of the two activities emphasized water in two separate contexts: in the investigation, the purpose was to explore an abiotic interaction studied reductively.

In the Connection Circle task, the purpose was to identify all the structures necessary for the successful functioning of the BioBottle (“successful” defined as enabling the plant to complete its life cycle). This confirms Doyle’s argument that an academic task serves to signal to students the content they are to address, the cognitive processes needed to accomplish academic work, and expectations for acceptable performance (Doyle, 1983). Jorge was good at “studenting,” following his teachers’ instructions and directions to “put down” ideas in his written work whether they made sense to him or not, as described in several episodes such as the role of water in ecosystems assessment, the connection circle, and his decision to decrease the water in his BioBottle system

198 after the teacher cautioned the class to think about the ratios of soil and water, and whether water could escape from a closed system.

On the other hand, like Ruby in Cold Mountain, Jorge actively engaged with and attended to the world he lived in. He brought much prior knowledge about “what goes with what” in an ecosystem from his prior curriculum experiences, and his life-world experiences of working with gardens and plants. Jorge expressed his own ideas most clearly in the pre-assessment task, and maintained a strong interest in the interrelated structures and functions of plant-roots-soil-water. In the context of our interview, Jorge expressed many ideas more fluently and in more detail than in his written work or participation in classroom activities. He expresses two important ecological ideas in this context. He explains during the thought experiment about building a desert BioBottle that different types of plants (e.g., a saguaro cactus vs. the Wisconsin Fast Plant) are “used to” different amounts of soil moisture and type of soil. Because the saguaro stores water, it needs less water in the soil; the Fast Plant is a “more moisty area kind of plant.” He also conjectures that should the amount of water in a given region increase or decrease, “nature might change the plant,” which could suggest either a Lamarckian view that the present organism would change, or an ecological view that the population of cacti would adapt to wetter soil. In our conversation Jorge also shares more detailed knowledge of how the roots of the plant take up water from the soil and that the water travels through the plant and exits through some structure in the leaves. Finally, he implicitly expresses ideas about limits to growth by predicting that if he left a living Fast Plant sealed in a bottle, in two years its leaves would be crowding the top like he has seen roots crowded and growing through the bottom of planters.

Had the classroom been structured more like Brown and Campione’s Community of

Learners model (1994), Jorge’s prior knowledge and emerging ecological ideas could have been identified early and made available as a class resource through their jigsaw model. Yet in the current

199 educational zeitgeist, there were limited opportunities for students in this study to express and explore their own ideas. A project designed to be “thought revealing” was reshaped by the framing of academic tasks and external pressures on the classroom into work that would demonstrate achievement over original thought. I will explore this further in revisiting the challenges for translating design-based modeling studies for everyday classrooms in the implications section below.

Role of the Microcosm

Findings from Chapter 4 suggest that Jorge continued to consider a model as a type of

“thing” or miniature reproduction of the real world (as did his teacher). The consistent emphasis on interactions and functions, along with expecting students to present a rationale for why they included each component in their BioBottle did prompt Jorge (and Rosa) to go beyond description of what happened and propose conjectures or explanations of how or why structures in their

BioBottle changed. However, their observations and measurements remained at a surface level of observation, which can be explained partially by the constraint of having the system closed and unavailable for closer inspection or investigation of events happening beyond what they could perceive. Yet as found by Carlsson (1999) and Helldén (1995), the intervention of closing the system did not prompt Jorge to consider the interrelationships of plants and animals in the system, although other students did and raised this as a concern. This confirms findings from studies such as Metz

(2000) that students will continue to view the system at a surface level unless they have opportunity to develop domain-based knowledge about the system. Several other students in the class did express ideas that the worms in the sealed system would not die because plants would produce oxygen. Even Jorge says in our interview that “fish need oxygen to survive,” and early in the unit he identifies “CO


” as a greenhouse gas, but he does not explicitly explain that oxygen is produced by plants. Metz’s criticisms of elementary curriculum embodying low expectations for students and being conceptually disconnected appears to be well-founded in this instance.


Carlsson (1999) and Helldén (1995) expected that the idea of sealing the ecosystem would make salient the interdependency of animals and plants in terms of the exchange of carbon dioxide and oxygen. The elementary curriculum that the students experienced addressed the dependency of animals on plants in terms of “how living things meet their needs for energy (food), water, shelter, and space” (p. 26), all of which are meso-scale, observable interactions. When the question arose in the class discussion about whether we should poke holes in the bottle because “worms need air,” the focus in the project was on enabling the plant to meet its needs, which led the teacher to shift the discussion towards what would happen to the plant (it would decrease in health) if water evaporated through the holes. Had this discussion been carried further, two other connections could have been made. One, if the water evaporates and the plant dies, would the worms have enough detritus to feed on to survive? Since most students added detritus in addition to soil, they might likely have considered this non-problematic. Two, students’ concern about introducing new air into the bottle to provide what the worms need could have been an opportunity to consider whether air is also a material that recycles in a sealed system, and make the connection back to the decomposition bottles which were sealed for six weeks but still supported living insects.

Until students have experience with modeling and observing the interaction of gas exchange between plants and animals through photosynthesis and respiration, the closed bottle system does not make this interdependency salient. This may be a partial explanation for why the student teachers in Carlsson’s studies and the elementary students in Helldén’s study did not change their thinking about interdependencies in an ecosystem when presented with a sealed system. This also provides further evidence for Eberbach and Crowley’s (2009) argument that the practice of observation in science is complex and dependent on domain knowledge.

The closed system also presents pragmatic constraints as an instrument, in the manner described by Pickering (1993) as “the mangle of practice.” Not all systems remained sealed, due to

201 water seepage through the clear tape holding the top and bottom together. This in turn led some students (though not Jorge) to argue that their worms or plant lived because they got air through these breaches. Jorge and Rosa, on the other hand, thought that the breach in their tape allowed fungi to get into the system, rather than recognizing fungi as microorganisms that existed in either the soil or the detritus they added to their system. A further practical limitation of a closed system was the inability for children to make clear observations; many who tried to observe with the jewelers’ loupe could not see clearly, either because condensation on the inside surface of the bottle blocked a view, or because the item of observation was beyond the focal length of the jewelers’ loupe. Students had already sealed their BioBottles before the discussion of “what variables to measure”; several pairs including Jorge and Rosa wanted to measure air temperature but did not have time to undo their seal. Jessie, a precocious third grader, took me by the hand and over to the two bottles we had left open, dropped in a thermometer and sealed the system, thus providing an option for anyone interested in air temperature.

The limits of the microcosm can also be attributed to the lack of variation in designs and limited time students had to actually observe changes. Lehrer, Schauble, and Lucas (2008) describe how the “crashing” of model aquatic systems that sixth grade students had designed to explore the ecological functioning of local holding ponds prompted students to critique and revise their designs, and also to conduct focused inquiries into the functions and interactions of different parts of the system. Had the Bottle Biology project been started at the beginning of the school year, there may have been opportunity for students in this study to do so as well.

The teacher and I missed opportunities to scaffold the idea of the BioBottle as a microcosm of the Earth’s system, as recommended by the research in introducing students to modeling (Baek,

Schwarz, Chen, Hokayem & Zhan, 2011; Fortus, Shwartz, Weizman, Schwarz & Merrit, 2008;

Lehrer & Schauble, 2006; Schwarz & White, 2005; Zangori, Forbes, & Schwarz, 2014). While the

202 teacher did remind students that all the water on Earth recycles during her water cycle lecture, that reference was not enough to scaffold Jorge’s reasoning about why the apple in the thought experiment or his mango narrative would shrink, and that a large proportion of “stuff” in the apple and mango (as well as the rest of the plant) is actually water. On the other hand, the design and building of the microcosm, and the act of sealing it, did provide an important affordance for Jorge and other students to consider ways that water changes form and moves through the parts of an ecosystem, while the total amount of water is conserved. The act of putting water into the bottle and sealing the bottle reinforced the idea that water in the system was limited, and the visible boundaries of the bottle and its seal reinforced the idea that water would have to recycle. Students may have developed more sophisticated understanding of materials cycles (and the presence of water in plants) if they had been able to construct and observe the decomposition BioBottles as well. Given more time for the project, we could have addressed this need for explicit scaffolding.

Enactment of Modeling Practices

The modeling practices in this study were enacted differently in relation to academic tasks than those of other modeling studies previously reviewed. While the core task of the project was for students to design, build, observe and explain a model ecosystem, the teacher and I decided that we needed to first introduce students to the processes of decomposition, which were not addressed in the BSCS Ecosystems curriculum. We also attempted to integrate activities from the FOSS Water unit.

In hindsight, this was an ambitious undertaking that introduced activities which both contradicted the holistic emphasis on modeling an ecosystem, and limited students’ time to observe and make sense of changes in their completed BioBottles. Lastly, introducing the task of the scientific poster also took time away from student sense-making and introduced a new communicative mode that needed to be learned. While students did have the opportunity to transform their raw data into models and explanations, they had very little time or scaffolding needed to do so successfully.


One important form of scaffolding in modeling practices is to create opportunities for children to extend their visual observation of the “things” in a system to express ideas about the structures (interrelationships and causalities) in a system, and explain how those structures affect function and behavior over time (Hmelo-Silver, Marathe & Liu, 2007; Lehrer & Schauble, 1998,

2004, 2005, 2010; Lehrer, Schauble, Carpenter, & Penner, 2000). It was also clear from the findings discussed in Chapter 4 that Jorge and several of the other children struggled to choose which forms of written representation would best express their ideas on the open-choice assessments. Likewise,

Jorge demonstrated limited (though creative) understanding of the conventions of constructing a mind-map/concept map or Connection Circle model. While researchers such as diSessa and his colleagues (1991, 2004) as well as Lehrer and Schauble (2000) advocate that children need the opportunity to invent and critique their own forms of models and inscriptions before learning the conventions of canonical scientific representations, this study illustrates the challenges of making time and providing teachers professional development in order to provide the strategic scaffolding children need to accomplish this.

In the process of developing and implementing the BioBottle project, neither the teacher nor

I explicitly discussed whether or how students’ work on the project task should or could be connected to activities that involved sustained development of inscriptional practices. I made the decision to leave the choice of representations in their pre- and post-assessments up to the children so that the work would capture their unique ways of seeing and making sense of ecosystems. In other assessments, the teacher encouraged them to use multiple modalities including words, numbers, pictures, diagrams, and “mind maps” (concept maps), as well as systems diagrams (such as the Connection Circle or a feedback loop). As I observed the class, I noticed that the teacher’s preferred modality was to have students talk with each other in small groups, then share out their ideas with the class. Writing in science notebooks was done if I wrote the lesson to include it, but

204 other opportunities for summarizing the class discussion and writing ideas or inscriptions in notebooks did not occur.

In many cases this was a function of our having gone over the time allotted for the science project, and needing to negotiate with the other teacher to “borrow” time from her lessons. As a result, students’ notations from their investigations, such as Jorge’s recording of data from the water-soil interaction, were terse, and important ideas he verbally expressed in talking with me or the teacher or his tablemates, such as the structure of different soils, were not recorded. I was actually surprised by how much Jorge remembered and could narrate about the activities we had done during our interview. Yet the lack of opportunity to share and critique representations for form and meaning constrained students’ opportunity to make sense and further attune their attention to salient dimensions of the model ecosystem.

Jorge’s memory may also have served as a constraint on his ability to observe what was actually happening inside the BioBottle model, vs. what he believed was going on, or to express his ideas about ecosystems in the assessments. For example, in his post assessment and science poster,

Jorge does not draw an accurate illustration of his and Rosa’s BioBottle with fungi taking over and the plant dead. Instead, in the post-assessment he draws a diagram of the bottle that is color-coded to show the different components in the bottle, but it is a conceptual representation, not an observational one. For their poster, Jorge and Rosa rely on drawings of the plant and its demise from memory rather than any inscriptions they made in their notebooks or on their observation sheets. They rely on annotated photographs (a reasonable strategy) to show the “dead plant” and the

“dirty water,” but these are neither inscriptions nor models because they do not provide an explanation of the processes that led to the outcome (Lehrer & Schauble, 2006; Windschitl &

Thompson, 2013). The most telling example of Jorge missing a “big idea” because he relied on memory rather than inscription is in his post-assessment diagram of the life cycle of a plant which

205 only notes water as affecting the roots, and that of the water cycle, in which he forgets to include the notion of groundwater and soil, and also omits living things despite the teachers’ explicit lesson.

Many researchers have documented students’ tendency to “see what they believe” (e.g.,

Chinn & Brewer, 1993; Metz, 2000; Schwarz, Reiser, Davis, Kenyon, Archer, Fortus et al., 2009).

Eberbach and Crowley (2009) argue that “When observations are disconnected from disciplinary contexts, we see but we do not observe” (p. 40). They argue that observation is a complex practice that requires disciplinary knowledge in a domain; students need more instructional scaffolding to develop observational and inscriptional accuracy. Building more time and opportunity into the project for students to construct and critique the meaning of different representations and models may have afforded Jorge the opportunity to construct more “coherency” across the academic tasks and his BioBottle observations (Hammer & Sikorski, 2015; Thompson & Windschitl, 2008).

It could have happened that with just a bit more scaffolding, Jorge and other students could have extended his idea about water cycling through the BioBottle to the idea that all materials in the closed system are conserved. During the discussion about whether to “close” the system or make it open by punching air holes, the teacher asked a student, “What happens to materials in a closed system?” The girl replied with a gesture, her finger going around in a circle. Ellie picked up on that and told the class “I want you to pay attention to what C—just said: do you think [makes gesture] happens to materials in a closed system?” Several nodded. Ellie said “That’s an important idea. Do this with me [makes the cycling gesture].” Had students been given the time to make a notation in their science notebooks about this important idea, they may have further consolidated their claim with a model-based representation (Windschitl, Thompson & Braaten, 2008).

The Role of Water in Ecosystems: Ideas In and Out of Focus

This section addresses the second part of the research question: “In particular, how did this student perceive and communicate his sense of the form and function of water in the microcosm

206 and the interaction of water with other living and non-living components of an ecosystem?As a result of the complex interactions of task, tools, and his own interests and agency, Jorge’s attunement to the role of water in ecosystems comes in and out of focus throughout the unit. Jorge indicated throughout the project that he knew water is an abiotic component of an ecosystem, that all living organisms require water for their survival, and that water changes form from solid to liquid to gas as it moves through the hydrosphere in the water cycle. Yet his work products in each task do not show all that he knows about water; the teacher’s emphasis on students including particular correct scientific ideas in their work products and the limited time made available for completing work led to Jorge including in his work only those ideas presented in the lesson. In his explanation during our interview that water moves from the soil through the roots, stem, and leaves of the plant, he reveals for the first time that he has either learned or thought about the micro-level, invisible interactions of water with structures of the plant. Likewise I would conjecture that if I could have asked him directly how animals and humans take in, use, and excrete water, he certainly could have told me from his own biological experience, if such a discussion were not considered too personal.

Yet in his written or drawn representations, he provides a simple gloss: plants and animals “need” water, which not only satisfies the demands of the task (label your connection line), but also repeats verbatim what he has read in the Ecosystems student guide (BSCS, 1997).

This finding is related to the argument I present below regarding the assumptions about learning progressions in some lines of research that students “progress” through levels from naïve descriptions to model-based reasoning. Hammer and Sikorski (2015) address this challenge in the learning progressions research from the premise that learning itself is a complex, dynamic, and emergent phenomenon. Variations and “idiosyncrasies” in learners’ sense making are an essential feature of learning, which leads to their claim that—

It is critical to recognize that what students experience as coherent along the way

207 may not be coherent in the sense of the target canonical understanding but still be a step in that direction….If learning is complex, then the dynamics of students’ reasoning may often be particular and idiosyncratic…. [C]omplexity affords idiosyncrasy, and it is evident across research on learning, especially in settings that cultivate students’ epistemic agency (Hammer & Sikorski, 2015, pp. 427-428).

Jorge brought his phenomenological attunement to the growing of plants and the relationship of plant-soil-water to the modeling situation, and through it developed nascent elements of the practice (Greeno, 1994) of using a model to make sense of the role of the water and water cycle in the ecosystem. Most significantly for Jorge, the work with the BioBottle introduced a critical systems affordance: the ability to visually and tangibly perceive the boundary of a closed system, and to use that perception of the boundary to make the connection not just that materials in the model cycled, but that, at least for water, matter is conserved. I would not argue that this scientific principle became fully salient to Jorge beyond his observation and explanation. As Eleanor Duckworth (1987) and Anna Sfard (1998) emphasize in their work, Jorge does not “have” the concept of matter conservation, rather he is having a wonderful idea about, and developing an attunement to, an important way that the world works, which is the beginning of a journey into knowing. Viewing learning and learners as complex, dynamic systems with idiosyncratic trajectories extends Greeno’s theory of situativity with a caution that in spite of all educators do to design “effective” learning environments, the learner, like Jorge, will find their own path through.

Broader Implications

In this section I discuss the meanings of this study for teaching and teacher education, and research.

Teaching, Curriculum, and Teacher Education

Findings from this study confirm cautions made by Lehrer and Schauble (2010, 2012) that

208 shifting elementary level teaching about ecosystems in light of the NGSS emphasis on practices of modeling will indeed be challenging, but approachable, for “everyday” classrooms and the current structure of schooling. In biology and ecology, phenomena such as the growth of organisms or decomposition unfold over periods of time that are longer than most science units are now allotted in schools. Patterns and trends are often subtle, and the “seeds” of big ideas such as variation and natural selection require multiple revisiting for students to develop more than a “school science narrative” level of understanding. Ultimately, to achieve understanding of the “big ideas” in ecology such as energy flow, materials cycles, evolution, and affect of humans on Earth systems, teachers and researchers need to collaborate over multiple years and work with the same group of students as much as possible through the grades in order to identify when and how to introduce and develop academic tasks that provide students the opportunity to explore an idea in depth.

This study confirms findings that elementary children can be productively engaged in

“seeing” beyond the immediately obvious, surface features of a model ecosystem to proposing conjectures and explanations about relationships, functions, and how these contribute to change over time. The sealed microcosm afforded Jorge and other students a perception of system boundaries, and concurrently, the cycling of materials in an ecosystem. Further, the increased emphasis on observing the function of water in the ecosystem served as a perceptually salient phenomenon that extended Jorge’s understanding of the role and behavior of an abiotic component of an ecosystem beyond meeting the “needs” of organisms. As discussed above, decomposition processes are equally perceptually salient and accessible to elementary students in a sealed model system, even in this case removed from direct observation to video. These phenomena, coupled with a focus on the autecology of a plant, confirm that such simplified physical models are appropriate starting points for elementary children to consider the multiple variables and interactions in an ecosystem, and the potential variations and changes in such systems over time.


Open questions remain about the most educationally fruitful ways to introduce and scaffold such models. Beginning the project at the beginning of the year, even before introducing the Ecology unit, would give the teacher and students more time to construct models, collect data, propose explanations, and revise their models based on multiple criteria. As we did with the Water unit, concepts and/or activities from the district’s related science kits such as New Plants, Air and Weather,

Earth Materials, The Changing Earth, Ecosystems, Water, and Structures of Life could either be integrated or conceptually related to the BioBottles and Decomposition bottles over the course of one or several years at the elementary level without needing to adopt new curriculum.

The inclusion of microcosm models at the elementary level, however, need to go beyond merely having students build and observe terraria or aquaria. Strategic instruction is also needed in teaching ecology through modeling to scaffold students’ thinking about the multiplicity of interactions in an ecosystem, creating a curriculum path that explicitly guides students (and teachers) to examine individual interactions in depth, and then connect those interactions to other interactions in the system. Additional time and academic tasks are needed to integrate the critical habits of systems thinking and understanding of systems behaviors, such as balancing and reinforcing feedback loops to explain ecosystems phenomena such as predator-prey equilibrium or perturbations in a food web, or explore how the availability of abiotic resources establish limits to growth. Time and policy support for these changes can be found in the Next Generation Science

Standards, which advocate postponing the teaching of energy flow and food chains in ecosystems to the middle grade and high school levels (NGSS Lead States, 2013).

Lehrer and Schauble emphasize in many of their studies the critical role of teacher knowledge and the need for them as researchers to work with a cohort of teachers over several years in a community of practice around the modeling design research in order to achieve the learning goals for students (cf. Lehrer & Schauble, 2004, 2005). Modeling practices of explanation and

210 inscription need to be scaffolded for students, which implies needed professional development and extended learning communities for teachers to consider students ideas, develop bridging activities, and assess student learning.

An extended modeling project such as this one places high demands on teachers, especially at the elementary level. Ecology is an interdisciplinary investigation that entails knowledge of not only biology, but chemistry, physics, and systems as well. While Ellie had two years prior to the study to develop her pedagogical content knowledge as a member of our Systems Thinking in

Middle School Science project, she also relied on me at several points to clarify her understanding of different concepts, particularly energy, and the relationship of energy flow to food chains and vice versa. As the studies reviewed in Chapter 2 emphasize, even undergraduate students and adults have difficulty with reasoning ecologically about materials cycles and energy flow. Few have been exposed to or experienced modeling practices in the ways that the new standards and research programs envision.

Lehrer and Schauble’s successful findings depend on an abundance of time, money, and resources that are not available to a typical teacher such as Ellie. Windschitl and colleagues have designed a framework designed to induct pre-service teachers into “ambitious” teaching practices centered around model-based inquiry for secondary science teachers, and have found that their heuristic, based on progressive disciplinary discourse, has been taken up by some of the teacher participants in their early teaching practice (Windschitl, Thompson & Braaten, 2008).

Assessment and Research

In this study, the decision to encourage students to use their own choices of representation during the pre- and post-assessment did serve the purpose of making some of their thinking visible, but not all of it, as indicated during later conversations during activities or in my interview with

Jorge. I note as a limitation of my research methods that not interviewing Jorge about his

211 representations left me to construct inferences and meanings that he may or may not have intended.

The work products Jorge produced for these assessments did not capture the depth of his thinking or the important insights about the water cycle that he expressed in his conversations with me. This is an ongoing tension of using open-ended and informal or semi-formal assessment tasks during classroom instruction, and not building in time for follow-on discussions with focal students about their work.

The tension between assessing for instruction and assessing for research is critical to keep in mind, especially in classroom studies and action research contexts. The task presented at the pre- and post-assessments was also so broad that it likely gave students too little guidance or scaffolding to know what aspects of their thinking about ecosystems to share. During the pre-assessment, the video shows many students having a difficult time getting started, either not knowing where to begin, or having to recall from memory knowledge and experiences from several months prior. The teacher observed this and began to address it by questioning students and reminding the class about ideas she felt were important for them to “get on your paper.” While her experience helped draw out details and ideas for students, her direction may also have been interpreted by students as what they should be showing. This suggests that while classroom artifacts can be used as research data, if a researcher is interested in students’ ideas, the assessment should take place outside of the classroom context, or the researcher and teacher need to negotiate clear boundaries for what the teacher’s role will be during this form of data collection.

The finding that learning is a complex, dynamic system through which learners construct their own pathways also must apply to assessment. Relying only on student work products to infer student learning obscures so much that students have made sense of, but are either unable or uninterested in articulating, or take what is being assessed as a given and so do not include it. This phenomenon is the basis for the recommendation that researchers triangulate their findings through

212 multiple data sources; in a perfect, well-funded classroom world, teachers would also be able to triangulate data sources when assessing student understanding.

Recommendations for Future Research

Being a systems thinker, I envision dozens of directions that this initial research could go in the future. Following the implications described above, there is much to understand about how to design instruction to support children in making sense of complex systems from the perspective of both parts and whole, extending children’s thinking to the microscopic interactions and processes of energy and materials transformation and the macro processes of energy flow, materials cycling, adaptation, evolution, and maintenance of equilibrium in a living system. How can we design more systematic instructional strategies for introducing students to the “expert” representations of ecologists and systems thinkers so that students develop understandings of the epistemology and purpose of such representations, as in the work of Lehrer and Schauble on distribution and variation? In what ways do children who participate in field work and school gardening develop understandings of the structures and functions of natural or human-designed ecosystems?

Reflections on the Research and Limitations of the Study

This qualitative case study emerged from a shared interest between the teacher, Ellie, and myself, about how students might learn more about systems and systems thinking in a disciplinary domain. But as Yogi Berra is alleged to have said, in theory there’s no difference between theory and practice; in practice there is. Collaborating with a colleague and friend to learn about student thinking in her classroom was not “clean” in terms of my role as a qualitative researcher. While participant observation and action research are two of many strategies under the umbrella of qualitative methodology, each requires a continual attention to open communication, clear intentions, and balance of power between researcher and “subject.” While the focal subject in this study was the student, Jorge, there was no “way in” to talking about Jorge’s actions or ideas without

213 also highlighting the decisions that Ellie and I made together. As the careful reader may note, there were times too where Ellie’s decisions as classroom teacher took the “content on the floor” in a direction different from what I had wished, or her communication of the purpose of the project, such as “successfully grow a plant” vs. design a system and learn what happens, became points of tension in the research. Her decision that she would participate if “you write the content and I’ll deliver it” (January, 2010) reveals a dimension of her beliefs about curriculum as content to be delivered, rather than explored; a belief reinforced by the norms and teacher evaluation criteria of the district. Yet there were many times where her improvisations that led away from the written plan were extremely generative, both for the students and the research. I set an intention at the beginning of the project to collaborate with Ellie as much as she had time and patience for collaboration, and to trust her decisions as an experienced teacher when we did not have time to meet and plan the week. Negotiating our work, personal boundaries, and sometimes different beliefs about learning and instruction became our “mangle of practice” in Pickering’s larger sense of making do with what you’ve got in the time you have.

A second related limitation in this study is that we lacked time during the study for me to conduct participant checks, a valuable form of checking validity and reliability by sharing data and conjectures with Ellie as a research participant (Bogden & Biklen, 1998; Stake, 1995, Merriam, 1998;

Patton 2002). I was able to discuss some early findings and ideas with her, and we have had several conversations related to teaching systems thinking since the study ended, but she has not read this thesis in its current form. Should I have the opportunity to turn this work into a manuscript for publication, Ellie will be second author and will have full input into interpreting or reinterpreting the data presented.


Returning to Orr’s vision of ecological literacy and the really big picture, how can researchers

214 and science educators best work towards rebuilding a resilient, attuned a culture that values the need to preserve the commons, to generate resiliency in ecosystems and human systems, and to engage children in the adventure of developing ecological literacy?

As Jorge’s final comments from our interview attest, children do not lack the interest or imagination to envision a different world than the one we have created.

E: What did you, uh, what did this project most help you learn about? Or think about?

J: Some people think there’s water in space? Well, not in space but in the moon?

E: Yes, some people think there’s water on the moon, uh huh…I think—

J: —and I was thinking that maybe if I was a scientist, maybe if I was older, maybe biosphere scientist, a person who studies about the biosphere, I was thinking of maybe we could have a bottle, and uh, live in there? And then have a fast plant in the thing, in the moon, and then maybe we can have like a cactus that grows up there…]?

E: On the moon?

J: On the moon.

E: What a neat experiment!

J: And maybe we can like test the water up there?

E: Yeah, so we could see if maybe fish could live in that water? If they’re in a system where they have everything else they need?

J: Hm yeah

E: Ok. That’s a really neat idea. So, would you suggest to other kids that they do this project?

J: Well, not on the moon but maybe …

The challenge for us as science educators is to learn more from the curious and imaginative students like Jorge through further research and practical work with them and their teachers.



Adeniyi, E. O. (1985). Misconceptions of selected ecological concepts held by some Nigerian students. Journal of Biological Education, 19(4), 311-316.

Aikenhead, G. (1996). Border crossing into the subculture of science. Studies in Science Education, 27,


Andersson, B. (1991). Students' conceptions of matter and its transformations. Studies in Science

Education, 18(1), 53-85.

Baek, H., Schwarz, C., Chen, J., Hokayem, H., & Zhan, L. (2011). Engaging elementary students in scientific modeling: The MoDeLS 5th grade approach and findings. In M. Khine & I. Saleh

(Eds.), Models and modeling: Cognitive tools for scientific enquiry (pp. 195-220). New York, NY:


Bar, V. (1989). Children's views about the water cycle. Science Education, 73(4), 481-500.

Barak, J., Gorodetsky, M., & Chipman, D. (1997). Understanding of energy in biology and vitalistic conceptions. International Journal of Science Education, 19(1), 21-30.

Barker, M., & Carr, M. (1989). Photosynthesis: Can our pupils see the wood for the trees? Journal of

Biological Education, 23(1), 41-44.

Barker, M., & Carr, M. (1989). Teaching and learning about photosynthesis, Part 1: An assessment in terms of student's prior knowledge. International Journal of Science Education, 11(1), 49-56.

Barker, M., & Carr, M. (1989). Teaching and learning about photosynthesis, Part 2: A generative learning strategy. International Journal of Science Education, 11(2), 141-152.

Barman, C., & Mayer, D. A. (1994). An analysis of high school students' concepts and textbook presentations of food chains and food webs. The American Biology Teacher, 56, 160-163.

Barman, C., Griffiths, A., & Okebukola, P. (1995). High school students' concepts regarding food chains and food webs: A multinational study. International Journal of Science Education, 17(6),


Bauer, M. W., Allum, N., & Miller, S. (2007). What can we learn from 25 years of PUS survey research? Public Understanding of Science, 16(1), 79-95.

Bazerman, C. (1988). Shaping written knowledge: The genre and activity of the experimental article in science.

Madison, WI: University of Wisconsin Press.

Becker, H. S. (1998). Tricks of the trade: How to think about your research while doing it. Chicago, IL: The

University of Chicago Press.

Bell, B. (1985). Students' ideas about plant nutrition: What are they? Journal of Biological Education,

19(3), 213-217.


Bogdan, R., & Biklen, S. K. (1992). Qualitative research for education: An introduction to theory and methods.

Needham Heights, MA: Allyn and Bacon.

Booth Sweeney, L. (2004). Thinking about everyday systems: An empirical investigation of middle school students'

and their teachers' conceptions of natural and social systems. (Ph.D. Dissertation), Harvard University,

Cambridge, MA.

Boschhuizen, R., & Brinkman, F. (1991). Environmental education: The use of pupil's ideas about cycles of health and nature. European Journal of Teacher Education, 1(1), 45-56.

Boulter, C., & Buckley, B. (2000). Constructing a typology of models for science education. In J.K.

Gilbert & C. Boulter (Eds.), Developing models in Science Education (pp. 41-57).

Dordrecht/Boston/London: Kluwer Academic Publishers.

Bowen, G. M., & Roth, W.-M. (2007). The practice of field ecology: Insights for science education.

Research in Science Education, 37, 171-187.

Bowers, C. A. (1996). The cultural dimensions of ecological literacy. Journal of Environmental Education,

27(2), 5-10.

Boyes, E., & Stanisstreet, M. (1990). Pupils' ideas concerning energy sources. International Journal of

Science Education, 12(5), 513-529.

Boyes, E., & Stanisstreet, M. (1991). Misconceptions in first-year undergraduate science students about energy sources for living organisms. Journal of Biological Education, 25(3).

Braaten, M., & Windschitl, M. (2011). Working toward a stronger conceptualization of scientific explanation for science education. Science Education, 95, 639-669.

Bransford, J. D., Brown, A., & Cocking, A. (Eds.). (2001). How People Learn. Washington, DC:

National Academies Press.

Bredo, E. (1994). Reconstructing educational psychology: Situated cognition and Deweyian pragmatism. Educational Psychologist, 29(1), 23-35.

Brown, A. L., & Campione, J. C. (1994). Guided discovery in a community of learners. In K.

McGilly (Ed.), Classroom lessons: Integrating cognitive theory and classroom practice (pp. 229-270).

Cambridge, MA: MIT Press.

Brown, A. L., Collins, & Duguid, P. (1989). Situated cognition and the culture of learning.

Educational Researcher, 18(1), 32-42.

Bruner, J. (1991). The narrative construction of reality. Critical Inquiry, 18, 1-21.

BSCS. (1999). Investigating Ecosystems. Dubuque, IA: Kendall/Hunt Publishing Company.

Buzan, T. (1996). The Mind Map Book: Plume Publishers.

Byrne, J. (2011). Models of micro-organisms: Children's knowledge and understanding of micro-

217 organisms from 7 to 14 years old. International Journal of Science Education, 33(14), 1927-1961.

Calabrese Barton, A., Koch, P., Contento, I., & Hagiwara, S. (2005). From global sustainability to inclusive education: Understanding urban children's ideas about the food system. International

Journal of Science Education, 27(10), 1163-1186.

Canham, C. D., Cole, J., & Lauenroth, W. (Eds.). (2003). Models in ecosystems science. Princeton, NJ:

Princeton University Press.

Capra, F. (1996). The web of life: A new scientific understanding of living systems. New York: Doubleday.

Capra, F. (1999). Ecoliteracy: The challenge for education in the next century. Paper presented at the

Liverpool Schumacher Lectures, Liverpool, UK.

Carlsson, B. (1999). Ecological understanding: A space of variation. (PhD), Luleå University of

Technology, Luleå. (1999:39)

Carlsson, B. (2002a). Ecological understanding 1: Ways of experiencing photosynthesis. International

Journal of Science Education, 24(7), 681-699.

Carlsson, B. (2002b). Ecological understanding 2: Transformation—a key to ecological understanding. International Journal of Science Education, 24(7), 701-715.

Carson, R. (1962/1993). Silent Spring. New York: Houghton Mifflin Company.

Cazden, C. (1988). Classroom discourse: The language of teaching and learning. Cambridge, MA: Harvard

University Press.

Chi, M. H. T. (1992). Conceptual change within and across ontological categories: Examples from learning and discovery in science. In R. Giere (Ed.), Cognitive models of science: Minnesota Studies

in the Philosophy of Science (pp. 129-186). Minneapolis, MN: University of Minnesota Press.

Chi, M. H. T., Slotta, J. D., & de Leeuw, N. (1994). From things to processes: A theory of conceptual change for learning science concepts. Learning and Instruction, 4, 27-43.

Chinn, C. A., & Brewer, W. F. (1993). The role of anomalous data in knowledge acquisition: A theoretical framework and implications for science instruction. Review of Educational Research,

63(1), 1-49.

Clement, J. J. (2000). Model based learning as a key research area for science education. International

Journal of Science Education, 22(9), 1041-1053.

Cobb, P. (1996). Where is the Mind? A coordination of sociocultural and cognitive constructivist perspectives. In C. U. Teachers College (Ed.), Constructivism: Theory, perspectives, and practice (pp.

34-52). New York: Teachers College Press.

Cobb, P., & Bowers, J. (1999). Cognitive and situated learning perspectives in theory and practice.

Educational Researcher, 28(2), 4-15.


Collins, A. (1992). Toward a design science of education. In E. Scanlon & T. O'Shea (Eds.), New

directions in educational technology (pp. 15-22). New York: Springer Verlag.

Corcoran, T., Mosher, F. A., & Rogat, A. (2009). Learning progressions in science: An evidencebased approach to reform. Washington, DC: Consortium for Policy Research in Education.

Covitt, B. A., Gunckel, K. L., & Anderson, C. W. (2009). Students' developing understanding of water in environmental systems. Journal of Environmental Education, 40(3), 37-51.

Creswell, D. (2003). Research design (2nd edition). Thousand Oaks, CA: Sage.

Cruzan, J. (1988). Teaching ecology with microcosms. The American Biology Teacher, 50(4), 226-228.

D'Avanzo, C. (2008). Why is ecology hard to learn? Bulletin of the Ecological Society of American(October

2008), 462-466.

Dewey, J. (1902). The child and the curriculum. Chicago, IL: University of Chicago Press.

Dewey, J. (1938/1963). Experience and education. New York: Macmillan.

Dillon, J., Rickinson, M., Teamey, K., Morris, M., Choi, M. Y., Sanders, D. & Benefield, P. (2006).

The value of outdoor learning: evidence from research in the UK and elsewhere. School

Science Review, 87(320), 107-113. diSessa, A. (1993). Toward an epistemology of physics. Cognition and Instruction, 10(2&3), 105-226. diSessa, A. (2004). Meta-representation: Naive competence and targets for instruction. Cognition and

Instruction, 22(3), 293-332. diSessa, A., Hammer, D., Sherin, B. L., & Kolpakowski, K. (1991). Inventing graphing:

Metarepresentational expertise in children. Journal of Mathematical Behavior, 10(2), 117-160.

Donaldson, M. (1987). Children's Minds. Guildford, UK: Biddles Ltd.

Dove, J. E., Everett, L. A., & Preece, P. F. W. (1999). Exploring a hydrological concept through children's drawings. International Journal of Science Education, 21(5), 485-497.

Doyle, W. (1977). Learning in the classroom environment: An ecological analysis. Journal of Teacher

Education, 28(6), 51-55.

Doyle, W. (1979). Classroom effects. Theory into Practice, 18(3), 138-144.

Doyle, W. (1981). Research on classroom contexts. Journal of Teacher Education, 32(6), 3-6.

Doyle, W. (1983). Academic work. Review of Educational Research, 53(2), 159-199.

Doyle, W. (1986). Content representation in teachers' definitions of academic work. Journal of

Curriculum Studies, 18(4), 365-379.

Doyle, W. (1992). Curriculum and pedagogy. In P. W. Jackson (Ed.), Handbook of research on curriculum


(pp. 486-516). New York: Macmillan.

Driver, R. (1983). The pupil as scientist? Milton Keynes, UK: Open University Press.

Driver, R., & Erickson, G. (1983). Theories-in-action: Some theoretical and empirical issues in the study of students' conceptual frameworks in science. Studies in science education, 10(1), 37-60.

Driver, R., Asoko, H., Leach, J., Mortimer, E., & Scott, P. (1994). Constructing scientific knowledge in the classroom. Educational Researcher, 23(7), 5-12.

Duckworth, E. (1987). The having of wonderful ideas and other essays on teaching and learning. New York:

Teachers College Press.

Eberbach, C., & Crowley, K. (2009). From everyday to scientific observation: How children learn to observe the biologist's world. Review of Educational Research, 79(1), 39-68.

Ecoliteracy, C. f. (2010). Rethinking School Lunch. Berkeley, CA: Center for Ecoliteracy.

Edgerton, F. (2007). Commentary: understanding food chains and food webs, 1700-1970. Bulletin of

the Ecological Society of American, January, 50-69.

Eilam, B. (2002). Strata of comprehending ecology: Looking through the prism of feeding relations.

Science Education, 86, 645-671.

Eilam, B. (2012). System thinking and feeding relations: Learning with a live ecosystem model.

Instructional Science, 40, 213-239.

Eisner, E. W. (1994). Cognition and curriculum reconsidered (2nd edition ed.). New York: Teachers

College Press.

Erickson, F. (1982). Classroom discourse as improvisation: Relationships between academic task structure and social participation structure in lessons. In L. C. Wilkinson (Ed.), Communicating

in the classroom (pp. 153-182). New York: Academic.

Ero-Tolliver, I., Lucas, D., & Schauble, L. (2013). Young children's thinking about decomposition:

Early modeling entrees to complex ideas in science. Research in Science Education, 43, 2137-


Eskilsson, O., & Helldén, G. (2003). A longitudinal study on 10-12-year-old's conceptions of the transformation of matter. Chemistry Education: Research and Practice, 4(3), 291-304.

Few, A. A. (1996). System behavior and system modeling. Sausalito, CA: University Science Books.

Ford, M. J. (2015). Learning progressions and progress: An introduction to our focus on learning progressions. Science Education, 99(3), 407-409.

Ford, M. J., & Forman, E. (2006). Redefining disciplinary learning in classroom contexts. Review of

Research in Education, 30(1), 1-32.


Fusco, D. (2001). Creating relevant science through urban planning and gardening. Journal of Research

in Science Teaching, 38(8), 860-877.

Gallegos, L., Jerezano, M., & Flores, F. (1994). Preconceptions and relations used by children in the construction of food chains. Journal of Research in Science Teaching, 31(3), 259-272.

Gardner, H. (1983/1993). Frames of mind. New York: Basic Books.

Gee, J. (1999). An introduction to discourse analysis: Theory and method. New York: Routledge.

Gibson, J. J. (1977). The theory of affordances. In R. Shaw & J. D. Bransford (Eds.), Perceiving, acting,

and knowing: Toward an ecological psychology. (pp. 67-82). Hillsdale, NJ: Lawrence Erlbaum.

Gilbert, J. K. (1991). Model building and a definition of science. Journal of Research in Science Teaching,

28, 73-79.

Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 606-633.

Goodwin, C., & Goodwin, M. H. (1996). Seeing as situated activity: Formulating Planes. In Y.

Engestrom & D. Middleton (Eds.), Cognition and communication at work (pp. 61-95). Cambridge,

UK: Cambridge University Press.

Gotwals, A. W., & Songer, N. B. (2010). Reasoning up and down a food chain: Using an assessment framework to investigate children's middle knowledge. Science Education, 94(2), 259-281.

Green, D. W. (1997). Explaining and envisaging an ecological phenomenon. British Journal of

Psychology, 88, 199-217.

Greene, M. (1993). Diversity and inclusion: Toward a curriculum for human beings. Teachers' College

Record, 95(2), 211-221.

Greeno, J. G. (1991). Number sense as situated knowing in a conceptual domain. Journal for Research

in Mathematics Education [Teaching], 22, 170-218.

Greeno, J. G. (1994). Gibson's affordances. Psychological Review, 101(2), 336-342.

Greeno, J. G. (1997). On claims that answer the wrong questions. Educational Researcher, 26(1), 5-17.

Greeno, J. G. (1998). The situativity of knowing, learning, and research. American Psychologist, 53(1),


Greeno, J. G. (2009). Learning in activity. In P. Robbins & M. Aydede (Eds.), Cambridge Handbook of

the Learning Sciences (pp. 79-96). New York: Cambridge University Press.

Greeno, J. G., & Hall, R. P. (1997). Practicing representation: Learning with and about representational forms. Phi Delta Kappan, 78(5), 361-368.

Griffiths, A., & Grant, B. (1985). High school students' understanding of food webs: Identification of a learning hierarchy and related misconceptions. Journal of Research in Science Teaching, 22(5),



Grosslight, L., Unger, C., Jay, E., & Smith, C. (1991). Understanding models and their use in science:

Conceptions of middle and high school students and experts. Journal of Research in Science

Teaching, 28(9), 799-822.

Grotzer, T. (1993). Children's understanding of complex causal relationships in natural systems. (PhD

Dissertation), Harvard University, Cambridge, MA.

Grotzer, T. (2002). Causal patterns in ecosystems. Curriculum. Harvard Graduate School of Education

Project Zero. Cambridge, MA.

Grotzer, T. (2003). Learning to understand forms of causality in scientifically accepted explanations.

Studies in science education, 39, 1-74.

Grotzer, T., & Bell Basca, B. (2003). How does grasping the underlying causal structures of ecosystems impact students' understanding? Journal of Biological Education, 38(1), 16-29.

Guba, E. G., & Lincoln, Y. S. (1994). Competing paradigms in qualitative research. In Y. S. Lincoln

& E. G. Guba (Eds.), Handbook of Qualitative Research (pp. 105-117). Thousand Oaks: Sage


Gunckel, K. L., Covitt, B. A., Salinas, I., & Anderson, C. W. (2012). A learning progression for water in socio-ecological systems. Journal of Research in Science Teaching, 49(7), 843-868.

Hammer, D., & Sikorski, T.-R. (2015). Implications of complexity for research on learning progressions. Science Education, 99(3), 424-431.

Hatano, G., & Inagaki, K. (1994). Young children's naive theory of biology. Cognition, 50, 171-188.

Hatano, G., Siegler, R. S., Richards, D. D., Inagaki, K., Stavy, R., & Wax, N. (1993). The development of biological knowledge: A multi-national study. Cognitive Development, 8(1), 47-


Heath, S. B. (1983). Ways with words. New York: Cambridge University Press.

Helldén, G. (1995). Environmental education and pupils' conceptions of matter. Environmental

Education Research, 1(3), 267-277.

Helldén, G. (2003). Longitudinal studies—Providing insight into individual themes in science learning and students' views of their own learning. In D. Psillos, P. Kariotoglou, V. Tselfes,

E. Hatsikraniotis, G. Fassoulopoulos & M. Kallery (Eds.), Science Education Research in the

Knowledge-Based Society (pp. 61-68). Dordrecht, The Netherlands: Kluwer/Springer.

Helldén, G. (2004). A study of some recurring core development features in students' conceptions of some key ecological processes. Canadian Journal of Science, Mathematics, and Technology

Education, 4(1), 59-76.

Hestenes, D. (1992). Modeling games in the Newtonian world. American Journal of Physics, 60(8), 732-



Hmelo, C., Holton, D. L., & Kolodner, J. L. (2000). Designing to learn about complex systems. The

Journal of the Learning Sciences, 9(3), 247-298.

Hmelo-Silver, C., & Pfeffer, M. G. (2003). Comparing expert and novice understanding of a complex system from the perspective of structures, behaviors, and functions. Cognitive Science,

28(1), 127-138.

Hmelo-Silver, C., Marathe, S., & Liu, L. (2007). Fish swim, rocks sit, and lungs breathe: Expertnovice understanding of complex systems. Journal of the Learning Sciences, 16(3), 307-331.

Hogan, K. (2000). Assessing students' systems reasoning in ecology. Journal of Biological Education,

35(1), 22-28.

Hogan, K., & Fisherkeller, J. (1996). Representing students' thinking about nutrient cycling in ecosystems: Bi-dimensional coding of a complex topic. Journal of Research in Science Teaching,

33(9), 941-970.

Hogan, K., & Weathers, K. C. (2003). Psychological and ecological perspectives on the development of systems thinking. In A. R. Berkowitz, K. S. Hollweg & C. H. Nilon (Eds.), Understanding

urban ecosystems: A new frontier for science education (pp. 233-260). New York: Springer-Verlag.

Inagaki, K. (1990a). Young children's use of knowledge in everyday biology. British Journal of

Developmental Psychology, 8, 281-288.

Inagaki, K. (1990b). The effect of raising animals on children's biological knowledge. British Journal of

Developmental Psychology, 8(2), 119-129.

Inagaki, K., & Hatano, G. (1987). Young children's spontaneous personification as analogy. Child

Development, 58, 1013-1020.

Inagaki, K., & Hatano, G. (2002). Young children's thinking about the biological world. Brighton, UK:

Psychology Press.

Ingram, M. (1993). Bottle Biology. Dubuque, IA: Kendall/Hunt Publishing Company.

Jennison, B., & Reiss, M. (1991). Does anyone know what energy is? Journal of Biological Education,

25(3), 173-177.

Jewitt, C. (2008). Multimodality and literacy in school classrooms. Review of Research in Education,

32(1), 241-267.

Jordan, R. C., Singer, F., Vaughan, J., & Berkowitz, A. R. (2009). What should every citizen know about ecology? Frontiers in ecology and the environment, 7(9), 495-500.

Kauffman, D. L. J. (1980). Systems One: An introduction to systems thinking. Minneapolis, MN: S.A.



Kelly, A., & Lesh, R. (Eds.). (2000). Handbook of research design in mathematics and science education.

Mahwah, NJ: Lawrence Erlbaum Associates.

Kesidou, S., & Duit, R. (1993). Students' conceptions of the second law of thermodynamics—an interpretive study. Journal of Research in Science Teaching, 30(1), 85-106.

Kirsh, D. (2009). Problem solving and situated cognition. In P. Robbins & M. Aydede (Eds.), The

Cambridge Handbook of Situated Cognition (pp. 264-306). New York: Cambridge University


Klein, P. D. (2006). The challenges of scientific literacy: From the viewpoint of second-generation cognitive science. International Journal of Science Education, 28(2-3), 143-178.

Kress, G., & van Leeuwen, T. (2006). Reading images (Second ed.). London, UK: Routledge.

Kress, G., Jewitt, C., Ogborn, J., & Tsatsarelis, C. (2001). Multimodal teaching and learning: The rhetoric of

the science classroom. London: Continuum.

Kuhn, T. (1962). Structure of scientific revolutions.

Laszlo, E. (1996). The systems view of the world: A holistic vision for our time. Cresskill, NJ: Hampton Press.

Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Cambridge, MA:

Harvard University Press.

Latour, B. (1990). Drawing things together. In M. Lynch & S. Woolgar (Eds.), Representation in

scientific practice (pp. 19-68). Cambridge, MA: MIT Press.

Latour, B., & Woolgar, S. (1979). Laboratory life: The construction of scientific facts. Princeton, NJ:

Princeton University Press.

Lave, J. (1988). Cognition in practice: Mind, mathematics and culture in everyday life. Cambridge, MA:

Cambridge University Press.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, UK:

Cambridge University Press.

Lave, J., Murtaugh, M., & de la Rocha, O. (1984). The dialectic of arithmetic in grocery shopping. In

B. Rogoff & J. Lave (Eds.), Everyday cognition: Its development in social context (pp. 67-94).

Leach, J., Driver, R., Scott, P., & Wood-Robinson, C. (1995). Children's ideas in ecology 1:

Theoretical background, design and methodology. International Journal of Science Education,

17(6), 721-732.

Leach, J., Driver, R., Scott, P., & Wood-Robinson, C. (1996a). Children's ideas about ecology 2:

Ideas found in children ages 5-16 about the cycling of matter. International Journal of Science

Education, 18, 19-34.

Leach, J., Driver, R., Scott, P., & Wood-Robinson, C. (1996b). Children's ideas about ecology 3:


Ideas found in children aged 5-16 about the interdependency of organisms. International

Journal of Science Education, 18, 19-34.

Lehrer, R., & Schauble, L. (1998). Reasoning about structure and function: Children's conceptions of gears. Journal of Research in Science Teaching, 35(1), 3-25.

Lehrer, R., & Schauble, L. (2000a). Developing model-based reasoning in mathematics and science.

Journal of Applied Developmental Psychology, 21(1), 39-48.

Lehrer, R., & Schauble, L. (2000b). Model-based reasoning in mathematics and science. In H. Doerr

& R. Lesh (Eds.), Beyond constructivism: A models and modeling perspective. Mahwah, NJ: Erlbaum.

Lehrer, R., & Schauble, L. (2000c). Modeling in mathematics and science. In R. Glaser (Ed.),

Advances in instructional psychology (Vol. 5, pp. 101-159). Mahwah, NJ: Erlbaum.

Lehrer, R., & Schauble, L. (2002). Symbolic communication in mathematics and science: Coconstituting inscription and thought. In E. D. Amsel & J. Byrnes (Eds.), The development of

symbolic communication (pp. 167-192). Mahwah, NJ: Erlbaum.

Lehrer, R., & Schauble, L. (2003). Origins and evolution of model-based reasoning in mathematics and science. In R. Lesh (Ed.), Beyond constructivism : models and modeling perspectives on mathematics

problem solving, learning, and teaching (pp. 59-70). Mahwah, NJ: Lawrence Erlbaum Associates.

Lehrer, R., & Schauble, L. (2004). Modeling natural variation through distribution. American

Educational Research Journal, 41(3), 635-679.

Lehrer, R., & Schauble, L. (2005). Developing modeling and argument in elementary grades. In T. A.

Romberg, T. P. Carpenter & F. Dremock (Eds.), Understanding math and science matters (pp. 29-

54). Mahwah, NJ: Lawrence Erlbaum Associates.

Lehrer, R., & Schauble, L. (2006). Cultivating model-based reasoning in science education. In R. K.

Sawyer (Ed.), The Cambridge handbook of the learning sciences (pp. 371-388). New York:

Cambridge University Press.

Lehrer, R., & Schauble, L. (2010). What kind of explanation is a model? In M. Stein (Ed.),

Instructional explanations in the disciplines (pp. 9-22). New York, NY: Springer.

Lehrer, R., & Schauble, L. (2012). Supporting inquiry about the foundations of evolutionary thinking in the elementary grades. In S. M. Carver & J. Shrager (Eds.), The journey from child to scientist:

Integrating cognitive development and the education sciences. (pp. 171-206). Washington, DC:

American Psychological Association.

Lehrer, R., & Schauble, L. (2015). Learning progressions: The whole world is NOT a stage. Science

Education, 99(3), 432-437.

Lehrer, R., Horvath, J. K., & Schauble, L. (1994). Developing model based reasoning. Interactive

Learning Environments, 4(3), 218-232.

Lehrer, R., Schauble, L., & Lucas, D. (2008). Supporting development of the epistemology of

225 inquiry. Cognitive Development, 23(4), 512-529.

Lehrer, R., Schauble, L., Carpenter, & Penner, D. (2000). The interrelated development of inscriptions and conceptual understanding. In P. Cobb, E. Yackel & K. McClain (Eds.),

Symbolizing and communicating in mathematics classrooms: Perspectives on discourse, tools, and instructional

design. (pp. 325-360). Mahwah, NJ: Lawrence Erlbaum.

Lehrer, R., Schauble, L., Strom, D., & Pligge, M. (2001). Similarity of form and substance: Modeling material kind. In S. M. Carver & D. Klahr (Eds.), Cognition and Instruction: Twenty-five years of

progress (pp. 39-74). Mahwah, NJ: Erlbaum.

Leigh, L. S., Burgess, T., Marino, B. D. V., & Wei, Y. D. (1999). Tropical rainforest biome of

Biosphere 2: Structure, composition, and results of the first two years of operation. Ecological

Engineering, 13(1), 65-93.

Lemke, J. (1990). Talking science: Language, learning and values. Norwood, NJ: Able Publishing.

Lemke, J. (1998). Multiplying Meaning: Visual and verbal semiotics in scientific text. In J. R. Martin

& R. Veel (Eds.), Reading Science: Critical and functional perspectives on discourse of science (pp. 87-

113). London: Routledge.

Lennon, K. A., & DeBoer, G. E. (2008). Probing middle school students' understanding of ideas about

interdependence in living systems through content aligned assessment. Paper presented at the National

Association for Research in Science Teaching, Baltimore, MD.

Lesh, R., & Doerr, H. (2003). Foundations of a models and modeling perspective on mathematics teaching, learning, and problem solving. In R. Lesh & H. Doerr (Eds.), Beyond constructivism:

Models and modeling perspectives on mathematical problem solving, learning, and teaching (pp. 3-33).

Mahwah, NJ: Lawrence Erlbaum.

Lesh, R., Hoover, M., Hole, B., Kelly, A., & Post, T. (2006). Principles for developing thoughtrevealing activities for students and teachers. In A. Kelly & R. Lesh (Eds.), Handbook of

research design in mathematics and science education. Mahwah, NJ: Lawrence Erlbaum Associates.

Lin, C.-Y., & Hu, R. (2003). Students' understanding of energy flow and matter cycling in the context of the food chain, photosynthesis, and respiration. International Journal of Science

Education, 25(12), 1529-1544.

Lofgren, L., & Helldén, G. (2008). Following young students' understanding of three phenomena in which transformations of matter occur. International Journal of Science and Mathematics Education,

6, 481-504.

Louv, R. (2005). Last child in the woods. New York, NY: Algonquin Books.

Magntorn, O., & Helldén, G. (2005). Student-teachers' ability to read nature: reflections on their own learning in ecology. International Journal of Science Education, 27(10), 1229-1254.

Magntorn, O., & Helldén, G. (2007). Reading nature from a 'bottom-up' perspective. Journal of

Biological Education, 41(2), 68-75.


Magntorn, O., & Helldén, G. (2007). Reading new environments: Students' ability to generalise their understanding between different ecosystems. International Journal of Science Education, 29(1), 67-


Manz, E. (2012). Understanding the codevelopment of modeling practice and ecological knowledge.

Science Education, 96(6), 1071-1105.

Marton, F. (1986). Phenomenography—A research approach to investigating different understandings of reality. Journal of Thought, 21(3), 28-49.

McNeill, K. L., Lizotte, D. J., Krajcik, J. S., & Marx, R. W. (2006). Supporting students' construction of scientific explanations by fading scaffolds in instructional design. The Journal of the Learning

Sciences, 15(2), 153-191.

Meadows, D. H. (2008). Thinking in systems: A primer. White River Junction, VT: Chelsea Green


Merriam, S. B. (1998). Qualitative research and case study applications in education. San Francisco, CA:

Jossey-Bass Publishers.

Metz, K. E. (1995). Reassessment of developmental constraints on children's science instruction.

Review of Educational Research, 65(2), 93-127.

Metz, K. E. (2000). Young children's inquiry in biology: Building the knowledge bases to empower independent inquiry. In J.A. Minstrell & E.H. Van Zee (Ed.), Inquiring into inquiry learning and

teaching in science (pp. 371-404). Washington, DC: American Association for the Advancement of Science.

Miles, M. B., & Huberman, A. M. (1994). Qualitative Data Analysis. Thousand Oaks, CA: Sage


Mohan, L., Chen, J., & Anderson, C. W. (2009). Developing a K-12 Learning Progression for

Carbon cycling in socio-ecological systems. Journal of Research in Science Teaching, 46(6), 675-


Munson, B. H. (1994). Ecological misconceptions. Journal of Environmental Education, 25(4), 30-34.

Norman, D. A. (1999). Affordance, conventions and design. Interactions, 6(3), 38-43.

National Research Council. (1995). National Science Education Standards. Washington, DC: The

National Academies Press.

National Research Council. (2007). Taking science to school: Learning and teaching science in grades K-8.

Washington, DC: The National Academies Press.

National Research Council. (2012). A framework for K-12 science education: Practices, crosscutting concepts,

and core ideas. Washington, DC: The National Academies Press.

NGSS Lead States, N. L. (2013). Next Generation Science Standards: For States, By States. Washington,


DC: The National Academies Press Retrieved from http://www.nextgenscience.org.

O'Connor, M. C. & Michaels, S. (1996). Shifting participant frameworks: Orchestrating thinking practices in group discussion. In D. Hicks (Ed.), Discourse, Learning, and Schooling (pp. 63 -

103). New York: Cambridge University Press.

Odum, H. T. (1968). Energy flow in ecosystems: A historical review. American Zoologist, 8(1), 11-18.

Orr, D. W. (1992). Ecological Literacy. Albany, NY: State University of New York Press.

Palmer, D. H. (1997). Students' application of the concept of interdependence to the issue of preservation of species: Observations on the ability to generalize. Journal of Research in Science

Teaching, 34(8), 837-850.

Patrick, P., & Tunnicliffe, S. D. (2011). What plants and animals do early childhood and primary students name? Where do they see them? Journal of Science Education & Technology, 20, 630-642.

Patton, M. Q. (2002). Qualitative research and evaluation methods (Third ed.). Thousand Oaks, CA: Sage


Penner, D., Giles, N. D., Lehrer, R., & Schauble, L. (1997). Building functional models: Designing an elbow. Journal of Research in Science Teaching, 34(2), 125-143.

Penner, D., Lehrer, R., & Schauble, L. (1998). From physical models to biomechanics: A designbased modeling approach. Journal of the Learning Sciences, 7(3 & 4), 429-449.

Perkins, D., & Grotzer, T. (2005). Dimensions of causal understanding: The role of complex causal models in students' understanding of science. Studies in science education.

Pfundt, H., & Duit, R. (1994). Bibliography: Students' alternative frameworks and science education (4th edition ed.). Kiel, Germany: Institute for Science Education.

Pickering, A. (1993). The mangle of practice: Agency and emergence in the sociology of science. The

American Journal of Sociology, 99(3), 559-589.

Pickett, S. T. A., Kolasa, J., & Jones, C. G. (2007). Ecological understanding: The nature of theory and the

theory of nature (Second ed.). San Diego, CA: Academic Press.

Polis, G. A., & Strong, D. R. (1996). Food web complexity and community dynamics. The American

Naturalist, 147(5), 813-846.

Prigogine, I., & Stengers, I. (1984). Order out of chaos: Man's new dialogue with nature. New York: Bantam


Quaden, R., Ticotsky, A., & Lyneis, D. (2004). The shape of change. Acton, MA: Creative Learning


Reiner, M., & Eilam, B. (2001). Conceptual classroom environment: A system view of learning.

International Journal of Science Education, 23(6), 551-568.


Reiss, M., Tunnicliffe, D. S., Andersen, A. M., Bartoszeck, A., Carvalho, G. S., Chen, S.-Y., Jarman,

R., Jonsson, S., Manokore, V., Marchenko, N. Mulemwa, J, Novikova, T, Otuka, J., Teppa,

S. & Van Roy, W. (2002). An international study of young peoples' drawings of what is inside themselves. Journal of Biological Education, 36(2), 58-64.

Resnick, M. (1995). Turtles, termites, and traffic jams. Cambridge, MA: MIT Press.

Richmond, B. (1993). Systems thinking: Critical thinking skills for the 1990's and beyond. System

Dynamics Review, 9(2), 113-133.

Roberts, E., & Johnson, B. (2005, April). Reasoning up and down the food chain: Fifth graders' understandings

of energy flow after an Earth Education immersion experience. Paper presented at the National

Association for Research in Science Teaching, Dallas, TX.

Roth, W.-M. (1996). Art and artifact of children's designing: A situated cognition perspective. The

Journal of the Learning Sciences, 5(2), 129-166.

Roth, W.-M. (2001). Situating cognition. The Journal of the Learning Sciences, 10(1&2), 27-61.

Roth, W.-M., & McGinn, M. (1998). Inscriptions: Toward a theory of representing as social practice.

Review of Educational Research, 68(1), 35-59.

Rowlands, M. (2004). What do children think happens to the food they eat? Journal of Biological

Education, 38(4), 167-171.

Salomon, G., & Perkins, D. (1998). Individual and social aspects of learning. In P. D. Pearson & A.

Iran-Jejad (Eds.), Review of Research in Education (Vol. 23, pp. 1-24).

Schauble, L. (2008). Commentary: Three questions about development. In R. Duschl & R. Grandy

(Eds.), Teaching scientific inquiry: Recommendations for research and implementation (pp. 50-56).

Rotterdam: Sense Publishers.

Schwarz, C. V., Reiser, B. J., Davis, E. A., Kenyon, L., Acher, A., Fortus, D., & Krajcik, J. (2009).

Developing a learning progression for scientific modeling: Making scientific modeling accessible and meaningful for learners. Journal of Research in Science Teaching, 46(632-654), 632-


Schwarz, C., & White, B. (2005). Meta-modeling knowledge: Developing students' understanding of scientific modeling. Cognition and Instruction, 23(2), 165-205.

Senge, P. (1990). The fifth discipline: The art and practice of the learning organization. New York:


Sfard, A. (1998). On two metaphors for learning and the dangers of choosing just one. Educational

Researcher, 27(2), 4-13.

Shepardson, D., & Britsch, S. J. (2001). The role of children's journals in elementary school science activities. Journal of Research in Science Teaching, 38(1), 43-69.


Slingsby, D., & Barker, S. (2003). Making connections: Biology, environmental education and education for sustainable development. Journal of Biological Education, 38(1), 4-7.

Smith, E. L., & Anderson, C. W. (1986). Alternative conceptions of matter cycling in ecosystems. Paper presented at the National Association of Research in Science Teaching, San Francisco, CA.

Paper retrieved from

Solomon, J. (1982). How children learn about energy—or: Does the first law come first? School

Science Review, 63, 415-422.

Solomon, J. (1983). Learning about energy: How pupils think in two domains. European Journal of

Science Education, 5(1), 49-59.

Solomon, J. (1983). Messy, contradictory and obstinately persistent: a study on children's out-ofschool ideas about energy. School Science Review, 65, 225-229.

Stake, R. (1995). The art of case study research. Thousand Oaks, CA: Sage Publications.

Starr, C., & Taggart, R. (1984). Biology: The unity and diversity of life. Belmont, CA: Wadsworth

Publishing Company.

Stavy, R., & Wax, N. (1989). Children’s' conceptions of plants as living things. Human Development,

32, 88-94.

Stavy, R., Eisen, Y., & Yaakobi, D. (1987). How students aged 13-15 understand photosynthesis.

International Journal of Science Education, 9(1), 105-115.

Sukhontapatipak, C., & Srikosamatara, S. (2012). The role of field exercises in ecological learning and values education: Action research on the use of campus wetlands. Journal of Biological

Education, 46(1), 36-44.

Tamponnet, C., & Savage, C. (1994). Closed ecological systems. Journal of Biological Education, 28(3),


Tang, K., & Moje, E. (2010). Relating multimodal representations to the literacies of science. Research

in Science Education, 40, 81-85.

Thomashow, M. (2002). Bringing the biosphere home. Cambridge, MA: MIT Press.

Tubiello, F. N., Lin, G., Druitt, J. W., & Marino, B. D. V. (1999). ecosystem-level evapotranspiration and water-use efficiency in the desert biome of Biosphere 2. Ecological Engineering, 13(1), 263-


Tunnicliffe, S. D., & Ueckert, C. (2007). Teaching biology—The great dilemma. Journal of Biological

Education, 41(2), 51-52.

Van Driel, J. H., & Verloop, N. (2002). Experienced teachers' knowledge of teaching and learning of models and modeling in science education. International Journal of Science Education, 24(12),



Van Matre, S. (1979). Sunship Earth: An earth education program for getting to know your place in space.

Martinsville, IN: American Camping Association.

Van Matre, S. (1990). Earth education: A new beginning. Greenville, WV: Institute for Earth Education.

Van Matre, S., & Johnson, B. (1988). Earthkeepers: Four key for helping young people live in harmony with the

earth. Greenville, WV: The Institute for Earth Education. von Bertalanffy, L. (1968). General systems theory: Foundations, development, applications. New York, NY:

George Braziller.

Vygotsky, L. (1978). Mind in Society. Cambridge, MA: Harvard University Press.

Vygotsky, L. (1986). Thought and Language. Cambridge, MA: MIT Press.

Wandersee, J. H. (1983). Students' misconceptions about photosynthesis: A cross-age study. In H.

Helm & J. D. Novak (Eds.), Proceedings of the International Seminar: Misconceptions in Science and

Mathematics (pp. 898-934). Ithaca, NY: Cornell University.

Webb, P., & Boltt, G. (1990). Food chain to food web: A natural progression? Journal of Biological

Education, 24, 187-190.

Webb, P., & Boltt, G. (1991). High school pupils' and first-year university students' responses to questions based on data related to ecosystems. Journal of Biological Education, 25(2), 119-123.

Wertsch, J. V. (1991). Voices of the mind: A sociocultural approach to mediated action. Cambridge, MA:

Harvard University Press.

Wertsch, J. V. (Ed.). (1985). Culture, communication and cognition: Vygotskian perspectives. Cambridge:

Cambridge University Press.

White, P. A. (1997). Naive ecology: Causal judgments about a simple ecosystem. British Journal of

Psychology, 88, 219-233.

Wiggins, G., & McTighe, J. (1998). Understanding by design. Upper Saddle River, NJ: Prentice-Hall.

Wilson, B. B., & Myers, K. M. (2000). Situated cognition in theoretical and practical context. In D.

Jonassen & S. Land (Eds.), Theoretical foundations of learning environments (pp. 57-87). Mahwah,

NJ: Lawrence Erlbaum Associates.

Wilson, C. D., Anderson, C. W., Heidemann, M., Merrill, J. E., Merritt, B. W., Richmond, G., . . .

Parker, J. M. (2006). Assessing students' ability to trace matter in dynamic systems in cell biology. CBE-Life Sciences Education, 5(Winter), 323-331.

Windschitl, M., & Thompson, J. (2006). Transcending simple forms of school science investigation:

Can pre-service instruction foster teachers' understandings of model-based inquiry? American

Educational Research Journal, 43(4), 783-835.

Windschitl, M., Thompson, J., & Braaten, M. (2008). Beyond the scientific method: Model-based

231 inquiry as a new paradigm of preference for school science investigations. Science Education,

92(5), 941-967.

Wood-Robinson, C. (1991). Young people's ideas about plants. Studies in science education, 19, 119-135.

Worster, D. (1977/1988). Nature's Economy: A history of ecological ideas. New York: Cambridge

University Press.

Zangori, L., Forbes, C. T., & Schwarz, C. (2014). Exploring the effect of embedded scaffolding within curricular tasks on third-grade students' model-based explanations about hydrologic cycling. Science and Education, 24, 957-981.

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF