Widexpress 29: Evidence for the Benefits of Binaural Amplification

Widexpress 29: Evidence for the Benefits of Binaural Amplification
WIDEXPRESS
no.29
march 2011
By Robert W. Sweetow, Ph.D.
University of California, San Francisco
Evidence for the Benefits of
Binaural Amplification
scene analysis, improved understanding of speech in
adverse acoustic environments, better management
of bilateral tinnitus, and greater user satisfaction. Yet
despite these advantages, a significant number of
hearing health care professionals still do not, or feel
they cannot, abide by the general consensus that unless a significant asymmetry exists between the ears
in either sensitivity or word recognition ability, the
standard should be trial with binaural amplification.
Furthermore, much of this disparity in usage of binaural amplification and recommendation of two hearing
aids, is geographic specific. For example, in the United
States, approximately 80% of hearing aid users who
have bilateral hearing loss, own two aids1, but the same
can be said for only 40–65% for western European
nations 2 and unconfirmed data estimate binaural usage
to be as low as 30% in certain Asian regions 3.
The benefits of binaural hearing have been known for
decades. They include, but are likely not limited to,
elimination of the head shadow effect, binaural summation, binaural squelch, reduction of central auditory
sensory deprivation, utilization of the unique characteristics of certain neurons to encode both arrival time
and details about the shape of the acoustic input in
order to enhance localization, optimization of auditory
www.widex.com
In this issue of WidexPress, this perplexing matter will
be confronted by 1) exploring reasons, both justifiable
and unjustifiable, for not regularly utilizing binaural
amplification when a bilateral hearing loss exists; 2)
examining evidence supporting binaural superiority;
3) discussing whether hearing aids take advantage of
binaural cues; and 4) briefly describing new strategies designed to maintain those cues for the hearing
impaired listener.
widexpress
1
Reasons why binaural amplification isn’t
universally routine for listeners with
bilateral hearing loss
There are a number of reasons why users may not want
to wear two hearing aids and why professionals may
not recommend binaural amplification. These concerns range from legitimate to questionable to simply
unsupportable. Among the legitimate concerns from
the user perspective, is the fact that there is greater
cost for two hearing aids than for one hearing aid.
Yet cost alone cannot account for this reluctance. For
example, in the United States where binaural usage is
greatest, 3rd party payers are involved to some extent
in less than 40% of hearing aids purchased, as compared to the United Kingdom where 74% of hearing aid
purchases receive public funding yet the binaural rate
there is only 47% 2. So the issue is more complex than
just money.
Undoubtedly, cosmetics play a role in some rejection
of binaural usage. Users have often stated the belief
that wearing two hearing aids levels the perception of a
more severe impairment than wearing only one device.
This perception however, has never been confirmed
in the literature, and it would seem logical that overt
errors in communication that result from not having
proper hearing in both ears may create the impression
of a more significant impairment than would be made
on the basis of having aids in both ears 4. Moreover, the
recent move toward less obtrusive hearing aids (such
as open fit, mini BTEs) should lessen cosmetic based
apprehension.
A concern that has been expressed by some hearing
aid users is the sensation of a larger occlusion effect
when wearing two aids. While this may indeed have
been a justifiable reason for not using binaural amplification in the past, the increasing popularity and availability of open fit hearing aids should lessen this worry
in a large percentage of cases.
Other reasons for preferring only one aid include fears
of dependence that could arise from wearing two aids
(never substantiated), and apprehension that two hearing aids will be louder than one hearing aid. This would
not be the case for properly fitted devices that incorporate the concept of binaural summation (discussed
below).
atrophy and/or demyelination of corpus callosal fibers, resulting in delay or other loss of the efficiency of
inter-hemispheric transfer of auditory information.” It
is important to recognize that there is a difference between “binaural interference” and lack of demonstrable
binaural integration in a laboratory environment. Better
testing materials and procedures are clearly needed to
identify this group of listeners.
A concern that has likely biased some professionals is
the belief that there is a lack of outcome data supporting the superiority of binaural amplification. Some early
publications 6,7 supported this perception. However, it is
now clear that the sensitivity of test materials used was
clearly inadequate. For example, many of the studies
unable to demonstrate binaural superiority compared
monosyllabic word recognition scores in quiet under
headphones or in free field environments that did not
utilize a diffuse field. Thus, neither head diffraction nor
binaural squelch effects were shown. Furthermore,
when comparisons were made outside of the lab setting, the hearing aids available in the 1960s, 1970s, and
even 1980s contained peak clippers, restricted frequency bandwidths, and even body borne devices that
were void of benefit from the pinna effect. In addition,
some published consumer surveys examined satisfaction with binaural hearing aids rather than comparing
satisfaction between monaural and binaural fittings 8.
One other factor that may account for reluctance in
some professionals to recommend binaural hearing
aids is the belief that there is extra time required for
the fitting of two devices. While it is true that there is a
small amount of additional time required, certainly the
fitting of two hearing aids does not double the required
time of fitting, and over the long run, communication problems minimized by binaural superiority may
indeed reduce the amount of post-fitting time spent by
the professional dealing with dissatisfied users.
Evidence for the advantages of binaural
hearing
There are clearly reasons why human beings have two
ears. The most important known binaural advantages
are:
Elimination of the head shadow effect
A legitimate concern that may impact both users and
professionals is that some individuals actually do function better with monaural amplification than with binaural amplification. “Binaural interference” 5 may occur
in as many as 20% of elderly hearing impaired listeners.
This may be attributable to “age-related progressive
www.widex.com
The head creates both a barrier to sound (sound
decreases by approximately 6.5 as it crosses from one
side of the head to the other) and a diffraction effect
(by providing a boost to sounds originating on that
side of the head). These effects in combination range
from about 3 dB for the low frequencies to as much
widexpress
2
as 16 dB in the high frequencies 9. Thus, wearing of
binaural hearing aids minimizes the possibility that the
physical location will create a significant disadvantage
for receiving important speech cues. For example, if
noise is present on the aided side and speech on the
unaided side, the speech originating from the unaided
side will not be perceived in either ear because of the
hearing loss on the unaided side and the masking from
the noise on the aided side. Moreover, one should consider the fact that even in a room surrounded by sound,
binaural usage would allow the user the option of turning off one hearing aid (if, for example it is on the side
of predominant noise and no speech).
Binaural summation
The very fact that sounds may be audible in each
ear provides an advantage of approximately 2-3 dB
by virtue of binaural redundancy or diotic summation. In addition, the loudness of a sound is greater
when presented to both ears than when presented to
one ear. Binaural loudness summation ranges from
approximately 3 dB at threshold to 6 dB at suprathreshold where presumably amplified speech occurs 10.
Therefore, each hearing aid used in the binaural mode
requires less to achieve the same effect. This would
likely have a number of beneficial effects including
lower distortion, less feedback and thus the possibility of larger vents. This in turn, may account, at least in
part, for the findings of several studies showing better
sound quality and speech understanding 11-13 for binaural amplification.
Enhanced localization
The ability to identify the direction and distance of a
moving or stationary sound source outside the head is
known as localization. There are some obvious important implications of this ability relative to both safety
and sound enjoyment. To accurately locate the source
of a sound, it is necessary to determine the horizontal
and vertical planes, as well as the front to back site of
origin.
Accurate identification of the horizontal location of a
sound is based on the comparison of interaural time
difference (ITD) of arrival and thus the phase of the
sound at each ear (because any delay in time results in
a phase shift), and the interaural intensity level difference (ILD) at the two ears. For example, a sound
originating directly in front of a listener (0 degree azimuth) would have an ITD of 0 at the two ears, whereas
a sound originating at a 90 degree azimuth would have
an ITD of about 0.7 msec. The duplex theory hypothesized by Lord Rayleigh in 1907 is still largely accepted
today (at least with regard to horizontal location)18. It
states that high frequency (>1600 Hz) location is determined by ILD, and low frequency (<800 Hz) location
is determined by ITD and phase differences. Thus, low
frequency sounds would be localized in the horizontal
plane toward the ear that receives the sound first and
high frequency sounds would be localized in the horizontal plane toward the ear which receives the higher
intensity level. However, interaural time differences also
are present in the envelope of sounds and thus can be
conveyed across the entire frequency range 19.
Binaural squelch
There is a significant binaural advantage that occurs in
the form of release from masking for the low frequencies as a result of the brain receiving dichotic signals
differing in time of arrival and phase. The magnitude of
this effect of extracting a speech signal from noise has
been shown to be as high as 13 dB 14.
0 msec
≈ 0.7 msec
≈ 0.7 msec
Reduction of central auditory sensory deprivation
While it is debatable whether disuse of an ear will
necessarily produce a further reduction in peripheral
loss (i.e. on pure tones), there is ample evidence
of neurological degeneration. Several studies have
demonstrated individuals who were fitted monaurally
suffered a greater loss of speech recognition ability in
the unaided ear than in the aided ear 15-17.
www.widex.com
0 msec
Fig 1: The Interaural Time difference: a sound originating at a 90
degree azimuth would have an ITD of about 0.7 msec, wheras a
sound originating directly in front of a listener will have an ITD
of 0.
widexpress
3
In addition to these long known facts, more recent
discoveries have demonstrated that certain neurons
encode both arrival time and details about the shape
of the acoustic input in order to enhance localization20.
For example, while ITD was believed to be perceived
with identical processing at the two ears based on the
rise time of the input signals, it has now been shown
that the rise time of ipsilateral signals is faster than
that of contralateral signals. In addition, neurons not
only encode coincidence in arrival time, but they also
detect details about the input’s shape. Neurons located
in the medial superior olive that respond to the rise
time of the summed input are differentiators while
neurons that respond to the net amplitude and width
of summed inputs are integrators. These neurons are
tuned to a particular frequency (CF) and are tuned to
a particular ITD by acting as interaural “coincidence
detectors” so different neurons have different “best”
ITD values equal to the ITD that compensates for differences in ipsi- and contra- transmission delays. The
ability to detect “group delay” is especially important
in reverberant environments 21.
Vertical localization is based on the high frequency
(above 4-5 kHz) reflections, echos, and resonances of
the pinna, head, and upper torso 9. Listeners can detect
vertical differences as slight as 3 degrees 22. The ability
to localize in the front-back dimension is also based on
high frequency cues provided by the pinna. Specifically, the pinna provides a 3-5 dB boost between 2-5
kHz for sounds originating in front of the listener and
further attenuates sounds originating from behind the
listener 23.
15
10
Optimization of auditory scene analysis
Perhaps of even greater importance to hearing impaired individuals is that the ability to localize allows
listeners to not only identify the location of a sound or
speaker of interest (particularly in multi-talker environments) but to also group together elements from
one direction and assign a separate identity to sounds
originating from different directions. Auditory scene
analysis enables one to recognize the environment and
identify objects through sound. It is defined as “the organization of sound scenes according to their inferred
sources” 24. Without this, simultaneously presented
sounds blend together to form a confusing background
rather than a collection of individual signals having
different perceived qualities (such as pitch or timbre)
which can be perceived or ignored. In addition, errors
in sequential grouping can lead to perceiving a single
word created out of syllables originating from two different voices.
Among the many complex features employed in this
process of tagging and grouping to allow for better
understanding of speech in noise are the use of fundamental frequency information, harmonicity, common
onsets/offsets, common amplitude modulations, common frequency modulations, and timbre.
Better management of bilateral tinnitus
Binaural hearing aids are commonly used in the management of patients with bilateral tinnitus 25 (and some
even believe for unilateral tinnitus), because of the wide
representation of tinnitus throughout the central auditory nervous system. With two hearing aids, tinnitus
awareness is often reduced, but if only one hearing
aid is worn, the tinnitus sufferer may find that there is
increased awareness of the tinnitus in the unaided ear.
5
Greater user satisfaction: Improved sound quality
-0
-5
-10
-15
500
1k
2k
4k
Frequency [Hz]
8k
Last, but certainly not least, among the advantages of
binaural amplification are the abundance of surveys
demonstrating greater user satisfaction and enhanced
sound quality, even in reverberant situations 8, 11, 13, 26, 27,
more convenience when one hearing aid breaks down,
and decreased listening effort for binaural hearing 28, 29.
These advantages are particularly important for older
listeners.
Average
Fig 2: The average Pinna Effect of 45 different heads (CIPIC
database)
www.widex.com
widexpress
4
Have hearing aids taken advantage of
binaural cues?
15
14
In addition, interaural phase and time differences (ITD)
can be obscured or altered by the combinations of
processing delays, compression characteristics, and
vent sizes. For instance, there may be differences in
the time constants for hearing aids worn in each ear,
the two aids may have different vent characteristics
thereby modifying transmission times, and there is
a processing delay in hearing aids of approximately
3-6 ms in speech frequencies, yet the brain can take
advantage of much shorter ITDs (recall that the maximum ITD is about 0.70 ms). Moreover, the processing
delay varies depending on the number of channels in
the hearing aid and can be nearly 10 msec in the low
frequencies for 16 channel devices (see Figure 3), and
even more if the hearing aid has other processing features such as noise reduction. Given the fact that most
digital hearing aids contain multiple channels and noise
reduction, this can be an important factor.
www.widex.com
13
12
11
10
Time [ms]
In order to capture the true binaural experience, it is
necessary to preserve the auditory cues discussed
above. Most of these binaural cues are captured simply
by virtue of making sounds audible in each ear, and by
having a distinct input to both ears. For example, the
binaural advantages of minimization of head shadow,
binaural summation, reduction of central auditory
sensory deprivation, better management of bilateral
tinnitus, and greater user satisfaction are maintained.
However, a number of binaural cues can be obscured
or distorted by wearable amplification. For example,
several studies have shown that localization may actually be better with no hearing aid than it is with hearing
aids 30, 32, 33. This happens as a consequence of both the
presence of a hearing impairment (and therefore the
loss of certain spectral cues) as well as processing and
physical characteristics of hearing aids. For example,
If the ILD is to be maintained, the relative volume has
to be unaltered, however, this can be changed in a
number of ways. Some types of compression could
reduce ILD. For example, if sound is originated from
one side, compression can reduce the gain for that
hearing aid while simultaneously increasing the gain on
the other hearing aid. This would change the relative
ILD. Also consider that if the hearing aid user changes
the volume in only one ear, the relative ILD is altered.
Similarly, the ILD can be modified if the hearing aid
user chooses to have different programs or two different microphone arrays in each hearing aid.
9
8
7
6
5
4
3
2
1
0,1k
0,2k
0,5k
1k
2k
5k
10k
Frequency [kHz]
4 Channels
16 Channels
16 Channels with DNR
Fig 3: The processing delay varies depending on the number of
channels in the hearing aid31.
Fortunately, while non-synchronized compression and
gain mismatch can alter ILD and ITD, some studies 32,35
suggest that horizontal localization is not appreciably
impacted as long as low frequencies are more than 10
dB above threshold. However, it should be noted that
not all of the potential factors that could affect low
frequency ITD were accounted for, so it is still feasible
that horizontal localization can be affected.
While this relative preservation of horizontal localization is welcome news, the fact that localization through
hearing aids has been shown to be less than ideal leads
to the conclusion that it must be errors in vertical and
front to back localization that impair the aided listener.
Indeed, it is these errors that appear to be most common for hearing aid users. This is because the loss of
the pinna effect and restricted bandwidth limit the important high frequency spectral cues and can negatively impact overall polar localization (encompassing both
vertical and front/back localization). Moreover, polar
localization is most adversely affected in BTEs because
of the loss of the pinna effect and concha resonance 3,
less affected in ITEs (despite the loss of the concha resonance) and least affected in CICs. Even so, diminished
localization cues with CICs and ITEs can still exceed 7
dB above 3000 Hz relative to the open ear.
New strategies designed to maintain
binaural advantages
Given the significant importance of localization to auditory scene analysis, and thus to the ability to understand speech in noise and reverberation, it is important
for hearing aids to do everything possible to preserve
timing and spectral cues, rather than distorting them.
widexpress
5
Recently, a number of new strategies designed to
protect these cues have been implemented in Widex
hearing aids.
For example, the Digital Pinna re-creates the natural attenuation of sounds coming from behind by setting the
frequency bands from 2 kHz and higher (bands 10-15)
in a fixed directional mode hypercardioid which picks
up sound at the front and eliminates most sound from
the sides and rear, while leaving the lower bands (1-9)
in omni-directional mode. The Digital Pinna with adaptive locator works in the way that all the channels will
still adapt to the best possible signal to noise ratio just
with the limitation that the upper channels can only
adapt from hypercardioid to bidirectional whereas the
lower channels can go all the way from omnidirectional
to bidirectional. In this way the directional system will
be in full Digtal Pinna setting in quiet surroundings and
in full adaptive settings in very noisy situations. The rationale for this approach is that while it is important to
maintain low frequency information for auditory scene
analysis, it is even more important to not impair intelligibility by allowing too much noise to filter through.
As discussed earlier, different program settings,
microphone arrays, volume control adjustments, and
disparate time constants and compression characteristics for each ear can distort interaural cues essential for
localization. Ultra-fast wireless communication between binaural hearing aids can significantly minimize,
if not prevent, these misrepresentations of cues from
occurring.
Also, given the importance of high frequency spectral
cues (above 4 kHz) for vertical and front to back localization, Widex hearing aids offer an extremely wide
bandwidth (as wide as 70 Hz to 11,200 Hz in ClearBand
models) with stable high frequency gain, made possible by a unique strategy of determining true feedback
via InterEar communication.
www.widex.com
widexpress
6
2011-05
9 502 2454 001 #01
References
1. Hougaard S., Ruf S. (2011). EuroTrak1: A consumer survey
about hearing aids in Germany, France and the UK. Hearing Review, 18(2);12-28.
2. Kochkin S. (2009). MarkeTrak VIII: 25 year trends in the
hearing health market. Hearing Review. 16(11): 12-31.
3. Kochkin S. Binaural hearing aids: the fitting of choice for
bilateral loss subjects http://www.betterhearing.org/pdfs/
Binaural_hearing_aid_complete_review.pdf.
4. Surr R., Hawkins D. (1988). New hearing aid users'
perception of the "hearing aid effect". Ear and Hear, 9(3):
113-118.
5. Chmiel R., Jerger J. (1996). Hearing aid use, central auditory disorder, and hearing handicap in elderly persons.
J AM Acad Audiol 7:190-202.
6. Jerger J., Dirks D. (1961). Binaural hearing aids, an
enigma. J Acoust Soc Am. 33(4):71-74.
7. Jerger J., Carhart R., Dirks D. (1961). Binaural hearing
aids and speech intelligibility. J Speech Hear Res. 4:137-148.
8. Briskey R.J., Cole P. (1983). Consumer evaluation of
binaural hearing aids. Hear J. 36(3): 29-31.
9. Shaw E. (1974). Acoustic response of external ear
replica at various angles of incidence. J Acoust Soc Amer.
55:432(A).
10. Haggard M., Hall J. (1982). Forms of binaural summation and the implications of individual variability for
binaural hearing aids. Binaural Effects in Normal and
Impaired Hearing. Scand. Audiol. Suppl (15): 47-63.
16. Gelfand S., Silman S., Ross L. (1987). Long term effects
of monaural, binaural and no amplification in subjects with
bilateral hearing loss. Scand. Audiol. 16:201-207.
17. Silman S., Silverman C.A., Emmer M., and Gelfand S.A.
(1992). Adult-onset auditory deprivation. J. Am. Acad.
Audiol. 3:390-396.
18. Moore B. (1989). An Introduction to the Psychology of
Hearing. London: Academic Press.
19. Henning G. (1974). Detectability of interaural delay
with high frequency complex waveforms. J Acoust Soc
Am. 55: 84-90.
20. Gulick L., Gescheider G., Frisina R. (1989). Hearing:
Physiological Acoustics, Neural Coding, and Psychoacoustics. Oxford University Press. New York. P. 188-215.
21. Shinn-Cunningham B., Kawakyu K. (2003). Neural
representation of source direction in reverberant space.
In: IEEE workshop on applications of signal processing to
audio and acoustics, pp 79–82. New Paltz, NY: IEEE.
22. Perrott D., Sabert K. (1990). Minimum audible angle
thresholds for sources varying in both elevation and azimuth. J Acoust Soc Am. 87(4): 1728-1731.
23. Musicant A., Butler R. (1985). The influence of pinnae
based spectral cues on sound localization. J Acoust Soc
Am, 77(1): 202-208.
24. Bregman A. (1990). Auditory scene analysis. MIT
Press: Cambridge, MA.
25. Sweetow R. (2011). Evidence supporting the use of Zen
as a tool for tinnitus and relaxation. WidexPress no.28,
11. Erdman S.A., Sedge R.K.(1981). Subjective comparisons
of binaural versus monaural amplification. Ear Hear. 2:225229.
26. Kochkin S., Kuk F. (1997). The binaural advantage:
Evidence from subjective benefit & consumer satisfaction.
Hearing Review. 4(4):29-32 , 34.
12. Balfour P.B., Hawkins D.B. (1992). A comparison of
sound quality judgments for monaural and binaural hearing aid processed stimuli. Ear Hear. 13(5):331-339.
27. Nabelek A.K., Robinson P.K. (1982). Monaural and
binaural speech perception in reverberation for listeners
of various ages. J. Acoust. Soc. Am. 71(5):1242-1248.
13. Erdman S.A., Sedge R.K. (1986). Preferences for
binaural amplification. Hear. J. 39(11):33-36.
28. Feuerstein J. (1992). Monaural versus binaural hearing:
ease of listening, word recognition, and attentional effort.
Ear Hear. 13(2):80-86.
14. Levitt H., Rabiner L. (1967). Binaural release from
masking for speech and gain in intelligibility. J. Acoust.
Soc. Am. 42(3):601-608.
29. Noble W. (2010). Assessing binaural hearing: results
using the speech, spatial and qualities of hearing scale. J
Am Acad Audiol 21(9):568-574.
15. Silman S., Gelfand S., Silverman C. (1984). Late onset
auditory deprivation: effect of monaural versus binaural
hearing aids. J. Acoust. Soc. Am. 76(5):1357-1362.
www.widex.com
widexpress
7
30. Van den Bogaert T., Klasen T., Moonen M., Van Deun
L., Wouters J. (2006). Localization with bilateral hearing
aids: without is better than with. J Acoust Soc Am.
119:515-526.
31. Herbig R., Chalupper J. (2010). Acceptable Processing
Delay in Digital Hearing Aids. Hearing Review. 17(1):28-31.
32. Keidser G., Rohrseitz K., Dillon H., Hamacher V.,
Carter L. et al. (2006). The effect of multi-channel wide
dynamic range compression, noise reduction and the
directional microphone on horizontal localization performance in hearing aid wearers. International Journal of
Audiology 45, 563-579.
33. Noble W., Byrne D., & Ter-Host K. (1997). Auditory
localization, detection of spatial separateness and speech
hearing in noise by hearing-impaired listeners. Journal of
the Acoustical Society of America 102, 2343-2352.
34. Keidser G., Convery E., Hamacher V. (2011). The effect
of gain mismatch on horizontal localization performance.
Hear Journal 64(2) 26-33.
35. Westermann S., Topholm J. (1985). Comparing BTEs
and ITEs for localizing speech. Hear Instr. 36(2):20-24,36.
WWW.WIDEX.COM
9 502 2454 001 #01
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

advertisement