Effects of Earplugs and Protective Headgear on Auditory

Effects of Earplugs and Protective Headgear on Auditory
Effects of Earplugs and Protective Headgear on Auditory
Localization Ability in the Horizontal Plane
Nancy L. Vause and D. Wesley Grantham, Vanderbilt University Medical Center,
Nashville, Tennessee
The purpose of this study was to determine how well humans localize sound
sources in the horizontal plane while wearing protective headgear with and
without hearing protection. In a source identification task, a stimulus was presented from 1 of 20 loudspeakers arrayed in a semicircular arc, and participants
stated which loudspeaker emitted the sound. Each participant was tested in 8
conditions involving various combinations of wearing a Kevlar® army helmet
and two types of earplugs. Testing was conducted at each of 2 orientations
(frontal and lateral). In the frontal orientation, overall error was slightly greater
in all protected conditions than in the bare-head control condition. In the lateral
orientation, overall error score in the protected conditions was substantially and
significantly greater than in the bare-head control conditions. Most errors in the
lateral orientation were accounted for by front-back confusions, indicating that
the protective devices disrupted high-frequency spectral cues that are the basis
for discriminating front from back sound sources. The results have practical
implications for the use of protective headgear and earplugs in industrial or military environments where localization of critical sounds is important.
Localization represents a fundamental auditory skill that contributes to survival by indicating the presence and position of mates, prey,
and enemies. Moreover, identification of sound
source positions is an intimate part of modernday orientation and surveillance of the environment. The ability to accurately localize and
quickly identify the position of potential hazards is critical in combat and in many of today’s
work environments.
Military helmets are designed to protect soldiers from possible head injury during training
and combat operations. Unfortunately, this protective headgear may also hinder the ability of
a soldier to localize sounds. Although Randall
and Holland (1972) reported that earlier helmet
designs (the Hayes-Stewart and M-1 helmets)
did not disrupt localization performance,
Howse and Elfner (1982) reported that tanker
helmets, when worn with various hearingprotective devices (HPDs), did result in an
increase in localization errors compared with
unprotected conditions. The differences in these
results may have been attributable to the different shapes of the helmets studied. Specifically, the tanker helmet occluded the pinnae
more than the earlier two helmet designs. The
effects on localization of the newer Kevlar®
helmet (Gentex Corp., Carbondale, PA; Figure
1), which is currently in use in the army, have
not been investigated.
Additionally, it is well documented that
HPDs (e.g., earmuffs and earplugs) designed
to protect soldiers or employees from noise
hazards and the possibility of noise-induced
hearing loss result in significant localization
difficulties (Abel & Hay, 1996; Atherly &
Noble, 1970; Noble, Murray, & Waugh, 1990;
Noble & Russell, 1972). However, it is not yet
clear if the primary factor underlying decreased
Address correspondence to D. Wesley Grantham, Department of Hearing and Speech Sciences, Vanderbilt Bill Wilkerson
Center for Otolaryngology and Communication Sciences, 1114 19th Avenue South, Nashville, TN 37212; d.wesley.
[email protected] HUMAN FACTORS, Vol. 41, No. 2, June 1999, pp. 282–294. Copyright © 1999, Human
Factors and Ergonomics Society. All rights reserved.
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
Figure 1. Kevlar army helmet employed in this study.
localization ability with HPDs is simply the
overall attenuation provided by HPDs, or if
the altered shape of the spectrum produced
by HPDs also contributes to poorer performance. The former notion, called the attenuation hypothesis, received some support from
studies indicating that as the attenuation of an
HPD increased, so did the detrimental effect
on localization performance (Mershon & Lin,
1987; Noble et al., 1990; Noble & Russell,
1972). However, given that these studies did
not measure the frequency-specific attenuation characteristics of the various HPDs, it is
not possible to know how disruption of spectral shape may have also contributed to the
decreased localization performance.
Etymotic Research (Elk Grove Village, IL)
has recently developed the ER25 “musician’s”
earplug, which is designed to attenuate evenly
across the frequency spectrum (Killion, DeVilbiss, & Stewart, 1988; see Figure 2b). (Since
this study was initially conceived, other manufacturers have developed “flat-response” HPDs,
such as the 9000 earmuff and the “high-fi”
earplug manufactured by Aearo Company,
Indianapolis, IN.) By comparing performance
with this earplug to performance obtained in a
control condition in which no earplugs are
worn but the source signal is attenuated by
25dB (the nominal specified attenuation for the
ER25 earplug), it should be possible to determine whether such “flat” earplugs offer a sig-
nificant advantage over conventional plugs in a
localization task.
The purpose of the present study was to
examine the effect of wearing the Kevlar helmet, one of two types of earplugs, or both the
helmet and earplugs, on localization ability.
Two sets of questions were addressed:
1. What are the effects, individually, of the Kevlar
helmet, the E-A-R earplug (Aearo Co.), and the
Etymotic ER25 musician’s earplug on localization performance in the horizontal plane? If
performance is degraded with the musician’s
earplug (compared with a bareheaded listening
condition), can the degradation be accounted
for simply by the attenuation provided?
2. What are the combined effects of the Kevlar
helmet and earplugs on horizontal-plane localization performance? In particular, does an
HPD affect performance for a person wearing a
These questions are each addressed for
localization of sound sources in two orientations with respect to the participant: (a) in the
frontal hemifield and (b) in the lateral hemifield on the participant’s left side.
The participants in the study were six adults
with normal hearing bilaterally (American
National Standards Institute, 1989) who were
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
paid volunteers. Five of the participants were
active duty or reserve soldiers and Reserve
Officers’ Training Corps cadets experienced in
small-squad maneuvers requiring localization
skills; one was a university student. All were
fitted with the proper size Kevlar helmet (small,
medium, or large) according to Army issue
guidelines. Additionally, a pair of custom earmolds for use with the ER25 earplug filters
was made for each participant. None of the participants had prior experience with laboratorycontrolled localization tasks.
Participants each took part in four to five
sessions per week for 2 to 3 weeks. Each session lasted about 2 h. Frequent breaks were
allowed within a session to avoid fatigue.
Figure 2. The types of earplugs employed in this
study: (a) E-A-R plug; (b) Etymotic ER25 plug,
illustrating the earmold and detachable filter; and
(c) manufacturer’s attenuation characteristics for EA-R and ER25 earplugs.
Testing Environment and Stimuli
Participants were tested individually in a
lighted anechoic chamber (6 m × 6 m × 6 m), for
which the low-frequency cutoff was 125 Hz.
During a session, a participant was seated in the
center of the room and instructed to maintain an
upright, forward orientation. There were 43 stationary loudspeakers positioned in a horizontal
arc at ear level and 1.8 m in front of the listener
(Figure 3). The loudspeakers were separated by
4° of arc and extended from –80° to +80°
azimuth. Although 43 loudspeakers were available, only 20 loudspeakers were used for this
experiment (shown as the darkened speakers in
Figure 3). Each of the 20 loudspeakers was
separated from its active neighbor by 8° of arc.
The stimulus employed was a digitized recording of the cocking of an M16 rifle (Human
Engineering Laboratory, Aberdeen Proving
Ground, Maryland). This signal was digitally filtered individually for each of the 20 loudspeakers in order to compensate for small differences
in the frequency responses of the loudspeakers.
Signals were output at a rate of 50 kHz and
were bandpass filtered from 100 Hz to 10 kHz.
The M16 cocking was chosen as the stimulus
because it is a brief sound (approximately 100
ms) that is quite common in combat. The waveform and spectrum are displayed in Figure 4.
The stimulus level was 60 dB SPL in all conditions except for the reduced-level control condition (bare head, no earplugs), which employed a
stimulus level of 35 dB SPL.
To verify that high-frequency information
from the M16 stimulus was available to participants, a 4.0-kHz high-pass filtered version
of the lower level (35 dB SPL) signal was analyzed. This filtered signal was determined to
be at least 10 dB above the ambient noise
level between 4.0 and 7.0 kHz (overall level of
the filtered signal was 26 dBA; overall ambient noise level was 17 dBA). It was clearly
audible from the participant’s position when
presented from any of the loudspeakers
employed in the experiment.
Three types of wearable devices were used:
the Kevlar Army combat helmet (Figure 1), the
E-A-R earplug with a noise reduction rating of
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
Figure 3. Loudspeaker configuration in the anechoic chamber. Symbols indicate the positions of the 43 loudspeakers (separated from each other by about 4°) in participants’ response set. The darkened symbols indicate the 20 loudspeakers in the stimulus set.
29 dB (Figure 2a), and the custom Etymotic
musician’s earplug (ER25) with a noise reduction rating of 25 dB (Figure 2b).
The ER25 earplugs required custom earmolds. The experimenter, after taking a custom
ear impression for each participant, ordered
custom-fitted earmolds for each participant
within the first week of testing. The E-A-R
earplug, a widely used disposable slow-recovery
foam earplug, was also utilized. Each participant was provided with a new pair of E-A-R
earplugs before testing.
For all earplug conditions, the experimenter
inserted the earplugs for participants in order
to ensure consistent and optimal placement.
Although individual ear attenuation measures
were not made, there was no evidence from
the data of anomalies or asymmetries (e.g.,
localization bias that might be expected with
asymmetric attenuation) in earplug placement.
After the entire experiment was completed,
actual attenuation provided by the ER25 earplugs was measured for each participant by
determining the absolute threshold for the M16
cocking stimulus with and without the plugs.
The difference between the two thresholds
reflects the attenuation of the protector and
should be correlated with its noise reduction
rating (NRR). Attenuation values for the six
participants are shown in Table 1. Because of
time limitations, similar measurements were
not made for the E-A-R plugs.
For three of the six participants (BR, KP,
and MD), measured attenuation approximated
specifications; for two other participants (BM
and KG), the attenuation was lower than
desired, but the performance of these two participants in the localization tasks did not systematically differ from the performance of the
first three participants. ER25 attenuation measured for the sixth participant (ME) was
markedly different from specifications. This
participant’s performance in the localization
tasks was also poorer than that of the other
participants in most conditions. A second
measurement was obtained from this participant after replacing the ER25 filters (the custom mold was not changed); this time the
attenuation was 21 dB (shown as the second
value in the table for ME), a value similar to
those obtained with the other participants.
Thus the variability of measurement seemed
to be caused not by an ill-fitting custom earmold but instead by the filter set.
A source identification task was employed
(Rakerd & Hartmann, 1985). The participant
was seated in the center of the room in one of
two orientations: (a) facing the center of the
loudspeaker array, as shown in Figure 3 (the
frontal orientation); or (b) turned 90° to the
right, so that the center of the array was opposite the participant’s left ear (the lateral orien-
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
Figure 4. The M16 cocking stimulus employed in all conditions: (a) waveform and
(b) magnitude spectrum.
tation). A small fixation light indicated to the
participant which direction to face.
A run consisted of the presentation of the
stimulus 80 times (4 times from each of the 20
loudspeakers employed) in random order.
Prior to each stimulus presentation, the participant directed his or her gaze at the fixation
light and pushed a button on a lap-held response box when ready. After the stimulus presentation, the participant responded by calling
out the loudspeaker number he or she believed
produced the sound, swiveling in the chair (if
necessary) to read the loudspeaker label. (A
pilot study indicated that loudspeaker cone
vibration could not have provided a visual cue
for detection.) The investigator, monitoring via
intercom from the adjacent control room,
typed the response into the computer. Feed-
back was not provided. To facilitate responding, each of the 43 loudspeakers was clearly
labeled with a number between 1 and 43.
Participants were unaware that only 20 of the
loudspeakers were employed.
During each run of 80 trials, the orientation
(frontal or lateral) was held constant. Each run
lasted approximately 10 min. When a run was
completed, the participant was given a minimum 1- to 2-min rest, after which a new run
began in a different orientation. In all, 3 runs
were obtained for each orientation, yielding a
total of 6 runs completed for each earplug/helmet condition (subsequently described). These
6 runs per condition were typically completed
in 2 consecutive sessions (on different days);
the order of the 6 runs was randomized for
each participant, with the constraint that the 3
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
TABLE 1: Measured Attenuation Provided by the
Etymotic ER25 Earplug for the Six Participants
ER25 Attenuation (dB)
14 (21)a
Despite efforts to train participants to optimum performance prior to data collection, it is
possible that long-term practice effects can occur
over the course of the main experiment. A comparison of performance in the initial and final
control conditions enabled an assessment of any
long-term practice effects for each participant.
Note: Values were computed by taking the difference between the absolute threshold for the
M16 cocking stimulus measured with and without the ER25 earplugs inserted.
Second measurement, made with the same earmold but a different filter set.
replications for a given orientation were not all
collected on the same day.
Eight experimental conditions were evaluated: (a) bare head/open ears (BH); (b) bare
head/E-A-R earplugs (EAR); (c) bare head/
ER25 earplugs (ER25); (d) Kevlar helmet/
open ears (K); (e) Kevlar helmet/E-A-R earplugs (K+EAR); (f) Kevlar helmet/ER25 earplugs (K+ER25); (g) bare head/open ears at a
lower level (BL; 35 dB HL); and (h) bare head
2nd/open ears (BH2).
The order of conditions was randomized and
was different for each participant, with the
exception of the first and last conditions (BH
and BH2), which were completed first and last
for all participants. Each of the other six helmetearplug combinations was completed (in the twosession protocol, as described earlier) before
moving to a new condition. This sequence was
chosen so that participants could become accustomed to each helmet-earplug combination
during testing rather than continually switch
among conditions.
Practice. Two sessions (4 h) were devoted
to practice before data collection began. During
these practice sessions, participants were exposed to both orientations and to several
earplug-helmet conditions. In addition, once
data collection began, a 20-trial warm-up (1
per loudspeaker) was provided at the beginning of each session and condition.
Data from one representative participant
are shown in Figures 5–7. These three figures
plot results from three of the eight conditions
(BH, K, and EAR), selected to give an illustration of the pattern of responses in both unprotected and protected conditions. In each
figure, the upper panels display results for
frontal orientations and the lower panels display results for lateral orientations. In each
panel, the mean azimuthal response (and standard deviation) is plotted as a function of
stimulus azimuth. Perfect performance is represented along the diagonal. Note that for the
frontal orientation, 0° on the axis represents
straight ahead of the participant, and for the
lateral orientation, 0° represents directly opposite the participant’s left ear.
In many of the lateral-orientation conditions, participants had front-back confusions
(e.g., a stimulus at +80° near the participant’s
front midline would be localized near –80°,
almost directly behind him or her). Because
these confusions were so prevalent, they were
kept separate from the nonconfused responses.
All responses that were nearer to the reflection
of the stimulus azimuth about the interaural
axis than they were to the actual stimulus
azimuth were averaged separately. These
means (with standard deviations) are displayed as squares in the lower panels of the
figures; the numbers beside each of the data
points indicate the number of confused responses (out of 12 presentations) that made
up each average. Note that front-back confusions are represented by the squares in the
lower right quadrants of the panels, whereas
back-front confusions are represented in the
upper left quadrants.
To obtain an overall representative error
score for each condition, the root-mean-square
(RMS) error D was computed as:
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
Figure 5. Data from participant BR in the BH condition. Mean azimuthal responses (with standard deviations) are plotted as a function of source position.
Standard deviations smaller than the size of the data
point are omitted. Mean error score (D) is indicated
on each graph. See text for further description.
Σ(p – r)2
where p is the source location (in degrees), r is
the response (in degrees), and N is the number
of responses (12). When averaged across the
20 loudspeaker positions, this measure is represented asD (see Rakerd & Hartmann, 1985).
This summary measure is given in each panel
Figure 6. Data from participant BR in K condition.
See caption for Figure 5.
in Figures 5–7 and is one of the primary
dependent measures in the analyses that follow.
Given that we consider the front-back confusions to be as practically significant as other
localization errors, no special treatment was
given to the confusions in the computation
ofD for the lateral orientations (i.e., we did
not resolve the confusions; a response of –80°
to a stimulus of +80° was scored as a raw error
of 160°). Separate analyses were conducted on
the percentage of confusions committed.
A cursory inspection of Figures 5–7 indicates that participant BR’s performance in the
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
Overall Analyses
Figure 7. Data from participant BR in EAR condition. See caption for Figure 5.
frontal orientation (upper panels) did not
change appreciably across conditions (across
all eight conditionsD varied between 6.2° and
10.5°; there were few severe departures from
the diagonal). Performance in the lateral orientation, on the other hand, varied over a
wide range across conditions (lower panels;D
varied across the eight conditions from 7.8° to
53.2°). Most of the departure from the positive diagonals could be accounted for by frontback confusions (shown as squares along the
negative diagonals). These effects will be described and discussed in the sections that follow.
Error scores. A two-factor (Listening Condition × Orientation) repeated-measures analysis
of variance (ANOVA) was conducted to analyze the data withD scores as the dependent
variable. As shown in Figure 8, results indicated significant main effects for both condition,
F(2, 13) = 14.14, p < .01, and orientation,
F(1, 5) = 64.04, p < .01, as well as a significant
interaction, F(3, 14) = 10.64, p < .01. Note that
the reported degrees of freedom reflect the
application of the Greenhouse-Geisser correction for nonhomogeneity of variance (Winer,
1971). The main effect of condition is reflected
in the consistently greater error scores in the
protected conditions than in the unprotected
conditions. Follow-up tests of simple effects
indicated that listening condition was a significant effect in the lateral orientation, F(3, 14) =
12.63, p <.01, but not in the frontal orientation, F(2, 9) = 3.44, p > .05. This difference
across orientations accounts for the significant
Orientation × Condition interaction.
Front-back confusions. As indicated previously, a large proportion of the error scores in
the lateral orientation shown in Figure 8 can
be accounted for by front-back confusions
(see the squares in the lower panels of Figures
5–7). Figure 9 plots the percentage of confusions (front-back and back-front confusions
are depicted separately) for the eight listening
conditions in the present investigation. A singlefactor, repeated-measures ANOVA indicated
that listening condition was a significant factor,
F(2, 10) = 16.33, p < .01 (degrees of freedom
reflect application of the Greenhouse-Geisser
correction for nonhomogeneity of variance).
The remaining discussion focuses on various
follow-up, pairwise comparisons from the data
displayed in Figures 8 and 9 to address the
research questions posed in the introduction.
Practice effects (BH vs. BH2). Comparisons
of the BH and BH2 conditions revealed that
practice did not significantly affect performance
in either orientation (see Table 2, section A).
The confusions committed in the lateral orientation (3.0% and 1.5%, respectively) are similar to values reported by Oldfield and Parker
(1984a, 1984b, 1986) and Wightman and Kistler
(1989). Additionally, this outcome supports the
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
Figure 8. Mean error score (D) across the six participants for the eight conditions tested in this study. Error bars
indicate 1 standard deviation, open bars represent frontal orientation, and solid bars represent lateral orientation.
premise that as listeners gain experience, confusions tend to diminish, as reported by Asano,
Suzuki, and Sone (1990) and Wenzel, Arruda,
Kistler, and Wightman (1993).
Effects of Helmet or Earplugs Worn
Effect of Kevlar helmet (BH vs. K). Surprisingly, the Kevlar protective helmet did not significantly disrupt localization compared with
the bare-head condition (Table 2, section B).
In the frontal orientation, the error score increased from 7.7° (BH) to 9.8° (K); in the lat-
eral orientation, the error score increased from
15.8° (BH) to 19.9° (K). It appears that listener
performance with only the Kevlar helmet (which
partially occludes the pinnae) is consistent
with performance with the earlier prototype
helmets (which did not occlude the pinnae) as
measured by Randall and Holland (1972).
Additionally, no significant difference was
found between confusion rates in the bare-head
condition (3.0%) compared with the Kevlar
condition (4.2%).
The lack of effect of the Kevlar helmet may
be related to the low-frequency emphasis in the
Figure 9. Mean percentage of front-back confusions in the lateral orientation for the eight conditions investigated.
Solid portions of bars represent front-back confusions; open portions of bars represent back-front confusions.
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
TABLE 2: Comparisons between Selected Pairs of Conditions, Showing Means (and Standard Deviations)
and the Results of the Statistical Tests
A: BH vs. BH2
8.0°(3.7 )
B: BH vs. K
C: BH vs. EAR
< .001
D1: BH vs. ER25
D2: BH vs. BL
D3: BL vs. ER25
E: K vs. K+EAR
< .01
F: K vs. K+ER25
Note: For each pair of conditions, comparisons are made for frontalDs, lateralDs, and percentages of
confusions in the lateral condition.
M16 stimulus employed (see Figure 4b). To the
extent that performance is governed by lowfrequency information, perhaps little disruptive
effect of acoustic cues by the helmet would be
expected (because of the relatively long wavelengths of the low-frequency sounds). On the
other hand, a stimulus with relatively more
high-frequency energy may have produced
more severely disrupted acoustic cues based on
the interaction of the shorter wavelengths with
the helmet structure. Whether such disruption
occurs with high-frequency stimuli, and if so,
whether it would lead to performance deficits
under helmet conditions, remains to be tested.
Effect of E-A-R plug (BH vs. EAR). When
the E-A-R plug was employed, performance was
disrupted to some degree in the frontal orientation, but the difference was not statistically
significant (Table 2, section C). On the other
hand, performance was dramatically worse in
the lateral orientation in the EAR condition
(D = 52.0°) than in the BH condition (D = 15.8°).
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
The increased error score was chiefly the
result of the greater percentage of confusions
made in the EAR condition (24.0%) than in
the BH condition (3.0%).
These results are consistent with the conclusions reached by several previous investigators (Abel, 1996; Atherley & Else, 1971;
Atherley & Noble, 1970; Mershon & Lin,
1987; Noble et al., 1990; Noble & Russell,
1972; Russell, 1976), indicating that HPDs
disrupt localization even for participants with
normal hearing. The increased high-frequency
attenuation provided by the E-A-R earplug,
although providing protection, appears to disrupt the auditory system’s ability to detect or
discriminate the high-frequency information
required to distinguish front from back sound
Effect of ER25 (BH vs. ER25; BH vs BL;
BL vs. ER25). A complete test of the effects of
the ER25 earplug on localization involves a
three-way comparison among the conditions
BH, BL, and ER25. The BL condition was included as a second comparison condition for
the ER25 condition, chosen such that the
stimulus level arriving at the participants’
eardrums would be approximately the same as
that received when the normal stimulus level
was presented and participants were wearing
the ER25 earplugs. The comparisons among
these three conditions are shown in Table 2,
sections D1, D2, and D3.
Section D2 of the table shows that there
were no significant differences between BH
and BL; thus, simply attenuating the signal
source by 25 dB had no significant effect on
any of the measures of performance. On the
other hand, there were some significant differences between ER25 and BH in both frontal
and lateral orientations (Table 2, section D1)
and between ER25 and BL in the frontal orientation (section D3).
These results indicate that in terms of localization precision, using the ER25 earplug is
not always equivalent to simply attenuating the
signal by 25 dB. Although performance in the
lateral orientation was not significantly different in the BL and ER25 conditions (suggesting
that the frequency response in the ear canal
with the ER25 plug may have been somewhat
preserved in the higher frequencies), perfor-
mance in the frontal orientation was significantly better in the BL than in the ER25 condition. The latter result suggests that the ER25
earplug may have disrupted the lower-frequency
interaural temporal cues required to localize
accurately in azimuth.
Effects of the Kevlar Helmet Combined
with Earplugs
In the frontal orientation, the K+EAR combination condition did not result in significantly more errors than the helmet-alone (K)
condition (Table 2, section E). However, the
K+ER25 combination condition did result in a
small but significant increase in error rate over
the helmet-alone condition (Table 2, section F).
This result supports the conclusion from the
previous section that the ER25 earplug may
disrupt the interaural cues required for accurate localization in the frontal orientation.
In the lateral orientation, the mean error
score increased significantly with either combination (K+EAR or K+ER25) over that obtained
with the Kevlar helmet alone (Table 2, sections
E and F). The observed confusion rates in the
combination conditions (K+ER25: 12.5%;
K+EAR: 19.0%) are consistent with those of
Oldfield and Parker (1984a, 1984b, 1986),
who reported confusion rates in pinnae-occluded
conditions of 12.5% and 26%. These results
indicate that compared with wearing the helmet
alone, wearing earplugs plus a helmet disrupts
the cues that are important for discriminating
front from rear positions.
Localization Bias in the Frontal
All participants displayed a small but consistent bias to judge targets in the vicinity of ±40°
to be closer to midline than they actually were.
This bias was evident primarily in the unprotected condition (BH), but it showed up to some
extent in the other conditions as well. (Note the
departure of the data from the diagonals in the
upper panels in Figures 5–7.) It is not clear why
this bias occurred; however, Sandel, Teas,
Feddersen, and Jeffress (1955) reported a similar bias for sources at ±40° when the stimulus
frequency was between 1500 and 5000 Hz. In
the present study the bias decreased as the
source’s distance from the midline increased
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
beyond 40°, and it was no longer evident for
sources at ±80°. It should be reiterated that any
bias attributable to end effects (potentially a factor at extreme target positions) was presumably
minimized in this study, because the response
set of targets extended beyond the stimulus set
(see Figure 3).
When the Kevlar protective helmet was worn
alone, there was a small (about 25%) increase
in horizontal-plane localization errors compared with the bare-head conditions in both
the frontal and lateral orientations. These differences were not statistically significant.
When the earplugs were worn alone, there
was an increase in horizontal-plane localization errors compared with the bare-head conditions. In the frontal orientation, error scores
increased about 25% when either the E-A-R
or the Etymotic ER25 plug was used compared with the bare-head (baseline) condition.
This difference was statistically significant
only for the ER25 plug. The pattern of results
suggests that the use of either type of earplug
may have some disruptive effects on the interaural cues that underlie azimuthal localization
in the frontal hemifield; however, these disruptive effects are apparently greater for the
ER25 than for the E-A-R plug.
In the lateral orientation, error scores increased about threefold for the E-A-R plug and
about twofold for the ER25 plug. These differences were highly significant. Most of the increases in error scores could be accounted for
by the increase in the percentage of front-back
confusions. The confusions presumably occurred
because of elimination or disruption of highfrequency spectral cues caused by the differential
attenuation provided by the earplugs. Whether
the ER25 plug disrupted lateral orientation performance more than could be accounted for by a
simple 25-dB attenuation of the signal could not
be answered conclusively from our data.
The combination of the Kevlar helmet and
either type of earplug degraded localization
performance when compared with the Kevlar
helmet alone. This effect was small for the
frontal orientation (an increase in error rate of
4%–20%), but was substantial for the lateral
orientation (an increase in error rate of 60%–
110%). The increase in errors in the lateral
orientation was accounted for primarily by an
increase in the percentage of front-back confusions (from 4.2% for Kevlar alone to 19.0%
for K+EAR and 12.5% for K+ER25).
Overall, the results suggest that military
and industrial employees should carefully
weigh the advantages and necessity of head
and hearing protection for workers against the
risk associated with increased incidence of frontback localization errors. Further research is
needed to investigate the detrimental effects of
various helmets and HPDs on localization performance and to determine whether wearing
such protection over extended time periods
might reduce error scores. Results from such
future studies should enable the development
and use of protection devices that offer adequate protection while not severely compromising a worker’s auditory localization ability.
The authors thank William Howell, Editor of
Human Factors, and an anonymous reviewer
for many constructive comments on an earlier
version of this manuscript. The authors also
thank Daniel H. Ashmead for assistance in statistical design and interpretation. This research
was supported by grants from the National
Institute on Deafness and Other Communication Disorders and from the National
Organization for Hearing Research. Portions of
this research were presented at the 131st
Meeting of the Acoustical Society of America
(Vause & Grantham, 1996).
Abel, S. M. (1996). Combined effects of hearing loss and hearing protection on sound localization: Implications for worker safety. In
A. Axelsson, H. Borchgrevink, R. P. Hamernik, P.-A. Hellstrom,
D. Henderson, & R. J. Salvi (Eds.), Scientific basis of noise-induced
hearing loss (pp. 286–295). New York: Thieme Medical Publishers.
Abel, S. M., & Hay, V. H. (1996). Sound localization: The interaction of aging, hearing loss, and hearing protection. Scandinavian Audiology, 25, 3–12.
American National Standards Institute. (1989). ANSI S3.6-1989.
Specifications for audiometers. New York: Author.
Asano, F., Suzuki, Y., & Sone, T. (1990). Role of spectral cues in
median plane localization. Journal of the Acoustical Society of
America, 88, 159–168.
Atherley, G. R., & Else, D. (1971). Effect of earmuffs on the localization of sound under reverberant conditions. Proceedings of
the Royal Society of Medicine, 64, 203–205.
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
June 1999 – Human Factors
Atherley, G. R., & Noble, W. G. (1970). Effect of ear defenders
(earmuffs) on the localization of sound. British Journal of
Industrial Medicine, 27, 260–265.
Howse, W. R., & Elfner, L. P. (1982). Identification of sound
source azimuth with active and passive hearing protectors
(Interim Report Ad-A168542). Frederick, MD: Fort Detrick,
U.S. Army Medical Research and Development Command.
Killion, M. C., DeVilbiss, E. L., & Stewart, J. (1988). An earplug with
uniform 15-dB attenuation. The Hearing Journal, 41, 14–17.
Mershon, D. H., & Lin, L.-J. (1987). Directional localization in
high ambient noise with and without the use of hearing protectors. Ergonomics, 30, 1161–1173.
Noble, W. G., Murray, N., & Waugh, R. (1990). The effect of various hearing protectors on sound localization in the horizontal
and vertical planes. American Industrial Hygiene Association
Journal, 51, 370–377.
Noble, W. G., & Russell, G. (1972). Theoretical and practical
implications of the effects of hearing protection devices on
localization ability. Acta Otolaryngology, 74, 29–36.
Oldfield, S. R., & Parker, S. P. (1984a). Acuity of sound localisation: A topography of auditory space. I. Normal hearing conditions. Perception, 13, 581–600.
Oldfield, S. R., & Parker, S. P. (1984b). Acuity of sound localisation: A topography of auditory space. II. Pinna cues absent.
Perception, 13, 601–617.
Oldfield, S. R., & Parker, S. P. (1986). Acuity of sound localisation: A topography of auditory space. III. Monaural hearing
conditions. Perception, 15, 67–81.
Rakerd, B., & Hartmann, W. M. (1985). Localization of sound in
rooms. II: The effects of a single reflecting surface. Journal of
the Acoustical Society of America, 78, 524–533.
Randall, B. R., & Holland, H. H. (1972). The effect of helmet form on
hearing: Free field thresholds (Tech. Note 5-72). Aberdeen Proving
Ground, MD: U.S. Army Human Engineering Laboratory.
Russell, G. (1976). Effects of earmuffs and earplugs on azimuthal
changes in spectral patterns: Implication for theories of sound
localization. Journal of Auditory Research, 16, 193–207.
Sandel, T. T., Teas, D. C., Feddersen, W. E., & Jeffress, L. A.
(1955). Localization of sound from single and paired sources.
Journal of the Acoustical Society of America, 27, 842–852.
Vause, N. L., & Grantham, D. W. (1996). Effects of musician’s
earplugs and protective headgear on localization ability in the
horizontal plane. Journal of the Acoustical Society of America,
99, 2597.
Wenzel, E. M., Arruda, M., Kistler, D. J., & Wightman F. L.
(1993). Localization using nonindividualized head related
transfer functions. Journal of the Acoustical Society of
America, 94, 111–123.
Wightman, F. L., & Kistler, D. J. (1989). Headphone simulation of
free-field listening: II. Psychophysical validation. Journal of
the Acoustical Society of America, 85, 868–878.
Winer, B. J. (1971). Statistical principles in experimental design
(2nd ed.). New York: McGraw-Hill.
Nancy L. Vause received a Ph.D. in hearing and
speech sciences in 1997 from Vanderbilt University.
She is an experimental research audiologist at the
U.S. Army Research Laboratory in Aberdeen Proving
Ground, Maryland.
D. Wesley Grantham received a Ph.D. in experimental psychology in 1975 from Indiana University. He
is the Director of Research and a professor in the
Department of Hearing and Speech Sciences in the
Vanderbilt Bill Wilkerson Center for Otolaryngology
and Communication Sciences at the Vanderbilt University Medical Center.
Date received: August 11, 1997
Date accepted: June 26, 1998
Downloaded from hfs.sagepub.com at PENNSYLVANIA STATE UNIV on May 12, 2016
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