Note: Descriptions are shown in the official language in which they were submitted.
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
APPARATUS AND METHODS FOR MITIGATING IMPAIRMENTS DUE TO
CENTRAL AUDITORY NERVOUS SYSTEM BINAURAL PHASE-TIME
ASYNCHRONY
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION:
The present invention relates to apparatus and methods for diagnosing,
quantifying,
and correcting for human central auditory nervous system (CANS) impairment,
and in
particular binaural phase time delay asynchrony.
DESCRIPTION OF THE PRIOR ART:
Preliminary studies indicate an important connection between the binaural
synchronization of the central auditory nervous system (CANS) and gross, fine
and oral
motor function. A binaural phase time delay (BPTD) is defined herein as a
synchronization
disruption (delay) in phase and time of the auditory input signals to the two
ears. Two types
of BPTDs have been defined by the investigators: pathological BPTDs which are
"built-in"
to a person's CANS, as is the case with a person with neurological injury or
disease
process, and clinical BPTDs which are induced in a person's CANS using an
external
device, to compensate for a pathological phase time delay.
A BPTD is a combination of a phase shift and a time delay. For pure tones, a
specific phase shift results in a specific time delay. For example, at 1000 Hz
a 180 phase
shift results in a 0.5 ms time delay. However for speech and other multi-
frequency sounds,
one specific time delay would result in several different frequency-dependent
phase shifts.
Note that a time delay can be much larger than the maximum phase shift for a
given
frequency.
Operationally, binaural interaction of the CANS requires the two ears to
integrate
dichotic signals separated in time, frequency, and/or intensity. The brain
stem is crucial for
binaural interaction of acoustic stimuli. Stillman (1980) has emphasized that
precise timing
of excitatory and inhibitory inputs to each cell along the auditory pathway is
critical if each
cell is to respond in an appropriate manner. Oertel (1997) has also studied
the effects of
timing in the cochlear nuclei. The superior olivary complex is an important
relay station of
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
the ascending tract of the CANS and is critical for binaural listening
capabilities. It is this
cross correlation behavior of the two ears that afford the selective listening
capability in
noisy environments, and the ability to spatially localize sound sources.
However, it has been
shown that signals from the two ears must have synchronized arrival times for
binaural cells
to be activated in the superior olivary complex. A delayed signal from one ear
negates a
binaural response. There is evidence that the synchronization of auditory
stimuli is
important above the superior olivary complex, at the levels of the brainstem
and cerebral
cortex.
In individuals (adults and children) with an impaired CANS, a pathological
BPTD
1 0 has been observed between the two ears which is in some cases is quite
large (15-20 msec).
The pathological BPTD not only decreases speech intelligibility in complex
listening
environments, but also (somewhat surprisingly) degrades motor (gross, fine,
oral) and visual
performance. Furthermore, a clinically-induced BPTD, designed to compensate
for the
pathological BPTD in the subject, significantly improves the speech
intelligibility, gross and
1 5 oral motor function the subject.
It is well known that head injury frequently results in generalized trauma to
the
brainstem and to higher cortical mechanisms which include the central auditory
nervous
system, resulting in central auditory processing function abnormalities. Other
conditions
such as sensory integration problems, speech and language delays, hearing
impairment,
2 0 learning disabilities, multiple sclerosis, Parkinson's Disease, autism,
stuttering,
developmental delays, central auditory processing disorders, psychological
disorders, and
neurological disorders have been associated with CANS dysfunction. Individuals
with
central auditory processing problems often demonstrate difficulty
comprehending and
remembering auditory information. In addition, these individuals have
particular difficulty
2 5 attending to auditory information in the presence of auditory
distractions.
In some traumatic brain injuries and other debilitative neurological brain
disorders,
the processing of information by the central auditory nervous system is
impaired and affects
comprehension and recall of auditory information. Operationally, the central
auditory
nervous system typically receives auditory information from both ears and
integrates the
3 0 input received, even though the acoustic signals received by the ears may
be somewhat
separated in time, frequency, and/or intensity. Such binaural integration by
the central
auditory nervous system may be substantially provided in the brain stem.
Further, it has
been observed that the precise timing of excitory and inhibitory inputs to
cells of the central
auditory nervous system can affect these cells' behavior to respond
appropriately. In
3 5 particular, it has been shown that auditory signals from both ears must
have a relatively
2
CA 02386653 2002-04-05
WO 01/26420 PCT/USOU/27460
synchronized arrival time for certain binaural celis to be activated in the
superior olivary
complex. Thus, a delayed (e.g. millisecond) rtsponse from one ear can impair
the
integration of binaural response. This is not reflected in the function of the
inner ear.
However, an individual with a peripheral hearing loss may also have CANS
dysfunction or a
mechanical effect that creates a disruption of the synchrony between the two
ears.
Behavioral and physiological (auditory brainstem response, middle latency
response,
cortical evoked potentials and mismatched negativity) methods have been
employed to
measure time parameters of the central auditory nervous system. Previous
studies, however,
have only analyzed the relationship of timing differences with respect to
various pathologies
(e.g. a latency in response has occurred). In particular, the development of
tests quantifying
the changes in auditory input between a subject's ears has been solely used as
a diagnostic
procedure for identifying a central auditory processing dysfunction. Since the
anatomy of
the brain stem indicates links between binaural signal processing and
integration and motor
control, it is not surprising that disorders of the central auditory nervous
system often affect
1 5 other functions such as sensory perception, integration, fine and gross
motor, oral motor and
visual processing. Accordingly, it would be useful to provide procedures and a
diagnostic
device that more accurately identify and quantify binaural processing
disorders and the
relationship of such disorders to other neurologically-based abnormalities.
Further, it would
also be useful to have a device that subjects manifesting binaural dysfunction-
derived
disorders can utilize to enhance day-to-day activities so that there may be
enhanced speech
understanding and recognition, concentration, gross motor movements (e.g.
walking), fine
motor movements (e.g. writing), oral-motor movement (e.g. speaking) or visual
function.
The following references are relevant :o the present invention:
U.S. Patent 5,434,924 to Jampolsky; "Two New Methods for Assessment of
Central Auditory Function in Cases of Brain Disease," Matzker, Annals Of
Otology,
Rhinology, & Laryngology 68, 1185-1196, 1959; "Auditory and Vestibular
Aberrations in
Multiple Sclerosis," Noffsinger et al, Acta Otolaryngologica, 303 (Suppl.), 1-
63, 1972;
"Assessing Central Auditory Behavior in Children," Willeford, Central Auditory
Disfunction, 43-72, 1977; and Westone style # 47 Soft PVC Custom Molded Ear
Plug with
Quiet Tech Int. Filter.
A need remains in the art for apparatus and methods for diagnosing,
quantifying,
and correcting for binaural phase-time delay asynchrony.
SUMMARY OF THE INVENTION
It is an object of an aspect of the present invention to provide apparatus and
methods for
3
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
diagnosing, quantifying, and correcting for binaural phase-time delay
asynchrony.
The ideal overall relative time delay portion of the BPTD for the subject is
measured
by separating the high and low frequencies of a variety of words, and time
shifting one of
the two components relative to the other. The subject's comprehension of the
words will be
highest or best at that subject's ideal relative time delay. Time delays can
also be measured
by inducing a preselected time delay of monosyllabic or bisyllabic words in
one ear relative
to the other ear in the presence of multitalker babble.
A phase analysis test (PAT) measures the appropriate binaural phase shift at a
variety of frequencies by assessing the subject's ability to discriminate pure
tones from
1 0 narrow band noise centered in frequency around the target tone. The BPTD
device can be
used as a diagnostic tool in this situation. The BPTD device is capable of
interfacing with
standard audiometers to generate two types of stereo signals. One is the
target signal that is
comprised of a pure tone presented to both ears with a relative phase
difference of q degrees
between the target tone channels. The narrowband noise signal is also
processed by the
1 5 BPTD device on a stereo basis to generate a relative phase difference of f
degrees between
the two narrowband noise channels. The BPTD device then mixes these two types
of
signals together by performing a weighted summation operation and the output
can then be
presented to the subject. The weighting value on the summation processes is
used to vary
the signal-to-noise power ratio between the target signal and the noise. Since
the resulting
2 0 output signal is a combination of these two signal types we call it the
SqNf output signal.
The BPTD device can implement a variety of combinations of q and f parameters
for
research, diagnostic, and accommodative purposes. The PAT test is produced by
generating
SqNq (q = f) output signals over a range of tone frequency and phase (q)
values. For each
frequency and phase combination the amplitude of the target tone is varied in
a procedure to
2 5 establish the minimum target strength level needed to hear the target
signal in the presence
of narrowband noise. For diagnostic purposes, threshold results from a
specific SqNq at
various frequencies represent the baseline condition. Normative threshold
measures for
each target frequency and phase value tested will be used to determine
atypical phase results
for various frequencies.
3 0 The optimal phase value for a given operating frequency for accommodative
purposes is the one in which the tone is heard at the lowest hearing threshold
value. An
operator interface allows the BPTD device to be used to systematically collect
the optimal
phase values over the range of test frequencies.
The PAT test also includes the synthesis of all of the phase information to
form a
4
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
phase correction filter as illustrated in Figure 9. The BPTD device also has
the capability
of implementing the phase correction filter in real-time. With the correction
filter in place, a
speech stimulus can be used to repeat the phase analysis paradigm described
above.
However, since speech is a frequency rich stimulus, the narrowband noise is
replaced with
noise that has a broader frequency profile, such as broad-band or white noise.
As before, a
procedure for determining the minimum hearing threshold for when the target
speech signal
is heard above the broad-band noise signal is implemented. Comparisons can
then be made
to the situation where the correction filter is not in place and performance
improvements can
be verified. It should also be noted that the phase correction filter can be
compared or
1 0 combined with optimal time delay parameters (such as those obtained from
the Delayed
Binaural Fusion Test). Furthermore, the BPTD device is capable of implementing
this
hearing threshold approach to investigate and diagnose time delay parameters
such as those
considered for the Delayed Binaural Fusion Test.
An electronic device is used for diagnosing and measuring the phase and time
1 5 portion of BPTD, and verifying the best overall relative time delay for
the subject. An
operator controls the relative time delay and phase delay applied to the
subject's ears via an
operator interface. The test set up considers one frequency (tone) at a time
and applies the
selected phase shift (and/or time delay) to whatever frequency is applied. The
operator
interface may include a keypad to enter control signals, and a display to show
which control
2 0 signal is being applied. Control signals set the amount of phase shift to
be applied. The
relative time delay shifter applies the overall relative time delay, and the
phase shifter applies
the phase shift.
A real time, active, digital signal processing electronic device is used for
correcting
BPTD, once it has been measured in the testing phase. Equivalent analog
devices could also
2 5 be used, but digital devices are more practical. In general, only one of
the devices will be
used, since sound is typically delayed to the same ear at every frequency.
Sound enters a microphone, which turns it into an analog electrical signal
representing the sound. The signal is amplified by a preamp, and is digitized
in an analog to
digital converter (ADC). A digital signal processor (DSP) operates much as the
test device
3 0 operated, applying an overall relative time delay and a phase shift
profile. A digital to analog
converter (DAC) converts the processed signal back to an analog signal, an
amplifier filters
and amplifies the signal, and a microphone turns the signal into an audio
signal to be
delivered to the ear of the subject.
The BPTD applied by the DSP is programrned according to the overall relative
time
3 5 shift and the phase shift versus frequency profile obtained in the testing
phase. The BPTD
5
CA 02386653 2009-04-08
profile is unique for each subject. The DSP could be reprogrammable, via a
control signal, so
it could be optimized for the wearer in actual use. Note that other hearing
aid processing
(compression or the like) may also be incorporated into the DSP if desired.
Amplitude
changes may also be implemented. In addition, the BPTD profile used may change
with the
kind of background noise detected by the device, or the type of activity the
subject is
performing.
A physical filter (a passive earplug) may alternatively be used for correcting
BPTD.
A physical device in the ear can delay the sound in the ear, and can delay
different
frequencies differently, as an electronic device does. The passive earplug
induces a BPTD to
sound entering the ear by altering the propagation time of the acoustic waves.
The primary
method of delaying an acoustic signal in this manner is through the use of
ducting, through
which the signal propagates. The velocity of propagation of sound in air is
approximately 331
meters per second, and the length of the ducting in the ear canal is about 10
cm (ducting along
an eyeglass frame can be longer). Thus the time delay applied by a passive
device in the ear
canal is on the order of 30 us, corresponding to a phase shift of about p13 at
5000Hz. This
time delay may be increased by about a factor of two by using a fluid rather
than air in the
ducting. In addition to the overall delay created by ducting, the frequency
response of the
earplug may also be tuned by using acoustical filter elements. Standard
elements include
chambers, Helmholtz resonators, and dampers. In addition, other acoustic
elements such as
horns, collectors, domes, trumpets, and resonators may be used.
In accordance with another aspect of the present invention, there is provided
an
apparatus for developing a binaural phase time delay profile (BPTD) of an
individual,
comprising:
means for separating spoken words into high frequency and low frequency bands;
means for delaying one of the bands relative to the other band;
means for presenting the delayed band to one ear and the undelayed band to the
other
ear;
means for measuring the individual's comprehension of the words at a variety
of
delays;
means for determining the overall phase delay at which comprehension is
optimal;
6
CA 02386653 2009-04-08
means for providing a pure tone;
means for dividing the pure tone into two channels;
means for applying a phase shift to one of the channels;
means for presenting the phase shifted channel to one ear, and the non-phase
shifted
channel to the other ear;
means for varying the amplitude and the frequency of the tone;
means for finding a threshold (lowest amplitude at which the tone is heard)
for
various frequency tones at various phase shifts; and
means for assigning an optimal phase shift at the various frequency tones
according
to the lowest amplitude at which each tone is heard.
In accordance with another aspect of the present invention, there is provided
a method
for developing a binaural phase time delay profile (BPTD) of an individual
comprising the
steps of:
separating spoken words into high frequency and low frequency bands;
delaying one of the bands relative to the other band;
presenting the delayed band to one ear and the undelayed band to the other
ear;
measuring the individual's comprehension of the words at a variety of delays;
determining the overall phase delay at which comprehension is optimal;
providing a pure tone;
dividing the pure tone into two channels;
applying a phase shift to one of the channels;
presenting the phase shifted channel to one ear, and the non-phase shifted
channel to
the other ear;
varying the amplitude and the frequency of the tone;
finding a threshold (lowest amplitude at which the tone is heard) for various
frequency tones at various phase shifts; and
assigning an optimal phase shift at the various frequency tones according to
the
lowest amplitude at which each tone is heard.
In accordance with another aspect of the present invention, there is provided
an
apparatus to develop a binaural phase time delay profile of an individual
comprising:
6a
CA 02386653 2009-04-08
a high frequency and low frequency bands separation element;
a frequency band delay element;
a presentation element adapted to present a delayed band to one ear and an
undelayed
band to another ear;
a comprehension measurement element responsive to a variety of delays;
an optimal overall phase delay determination element;
a pure tone element;
a two channel divider responsive to said pure tone element;
a phase shift element responsive to one of the channels of said two channel
devider;
a presentation element adapted to present a phase shifted channel to one ear;
a non-phase shifted channel to another ear;
a tone amplitude and frequency variation element;
a threshold determination element responsive to said tone amplitude and
frequency
variation element; and
an optimal phase shift assignment element responsive to the lowest amplitude
at
which each tone is heard.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a flow diagram showing a set of diagnostic procedures for
diagnosing
and quantifying pathological binaural phase time delay (BPTD) in subjects.
Figure 2 is a flow diagram showing the conventional tests performed in Figure
1, in
more detail.
Figure 3 is a flow diagram showing the phase analysis tests performed in
Figure 1,
in more detail.
Figure 4 is a flow diagram showing the delayed binaural fusion tests performed
in
Figure 1, in more detail.
Figure 5 is a block diagram of an electronic device for diagnosing and
measuring
BPTD.
Figure 6 is a block diagram of an electronic device for correcting BPTD.
6b
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
Figure 7 is a more detailed block diagram of the digital signal processor
(DSP) of
Figure 6.
Figure 8 is a flow diagram showing the process accomplished by the DSP of
Figure
6.
Figure 9 is a diagram showing an example of the correction accomplished in the
BP
block of the DSP of Figure 7.
Figure 10 is a cutaway side view of a physical filter for correcting BPTD.
Figures 11-19 are charts illustrating test results for an embodiment of the
present
invention.
1 0 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Figure 1 is a flow diagram 100 showing a set of diagnostic procedures for
diagnosing and quantifying pathological binaural phase time delay (BPTD) in
subjects. The
tests may be performed in any order, but the order shown is the most logical,
for reasons
described below.
1 5 The test routine begins with step 102. In general it will be desirable to
perform a
series of conventional hearing tests 104 on the subject first, in order to
determine whether
other hearing problems or central auditory processing problems exist. These
tests are
shown in more detail in figure 2. Next, a series of pure tone phase analysis
tests 108 are
performed to determine the optimal clinical phase shift at a variety of sound
frequencies.
2 0 The subject's ability to identify a tone out of noise centered around the
tone and the
resultant threshold is assessed at a variety of relative phase shifts between
ears, and at a
variety of frequencies, and a profile of the subject's phase shift frequency
profile is
generated. The frequency profile will be used to complement a phase correction
filter. With
this filter in place, speech stimuli can be used as a target in a similar
fashion with broad band
2 5 noise. These tests are shown in more detail in Figure 3.
Finally, a series of delayed binaural fusion tests are performed. These tests
assess
comprehension of words at a variety of relative time delays between ears, and
at a variety of
frequencies. Again the results of this test are used to develop the subject's
time delay
versus frequency profile. This test is matched up to the pure phase test to
complete the
3 0 BPTD profile. This test also allows for testing of relative shifts greater
than one wavelength,
which cannot be done with tones. Figure 5 shows an embodiment of an electronic
device
that assists in performing the tests of Figures 3 and 4.
7
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
In step 112, the results for the subject are compiled in a database. If the
pathological
BPTD for the subject is significant, this database is used to design an
electronic filter (see
Figure 6) or a physical filter (see Figure 7) to apply compensating clinical
BPTD to the
subjects ears.
Figure 2 is a flow diagram showing conventional tests 104 in more detail. The
conventional tests start at step 202. In general these tests include a pure
tone evaluation 204
to evaluate hearing loss at various frequencies. A central auditory processing
evaluation 206
is performed, and various electro-physiological assessments 208 are performed.
Other
assessments 210 may be added. The conventional tests end at step 212.
1 0 Central auditory processing evaluation 206 may include (but is not limited
to) such
tests as: Willeford central auditory test battery; Dichotic digits test;
Ipsilateral/ contralateral
competing messages; Synthetic sentence identification with contralateral
competing
messages; Masking level differences; Auditory duration patterns; Speech-in-
noise; Pediatric
speech intelligibility test; segment altered CVCs; pitch patterns; dichotic
chords; compressed
1 5 speech, with and without reverberation.
Electro-physiological assessments include such tests as: ABR; Middle
latencies;
Late latencies; P300; Mismatched Negativity. Binaural interaction components
will also be
calculated. Since electrophysiological measurements use various latency
classifications or
markers, this information may yield added information to the diagnosis and
quantification of
2 0 auditory asynchronies.
Figure 3 is a flow diagram showing the pure tone phase analysis test 108,
according
to the present invention. A device such as that shown in Figure 5 may be used
in this test.
This test assesses the subject's ability to discriminate pure tones from
narrow band noise
centered around the tone. The pure tone and the noise are presented to each
ear at a
2 5 different phase (giving a relative phase shift or clinical BPTD).
Thresholds are obtained for
each tone at varying phase shifts. In the preferred embodiment, a relative
phase shift is
selected, and the amplitude of the tone is increased until the subject can
pick it out of the
noise. The optimal phase shift is the phase shift that produces the smallest
amplitude
hearing threshold.
3 0 The test starts at step 302. Narrow band noise and pure tones are applied
to both
ears in step 304. In outer loop 310, frequencies are stepped through, for
example from 500
Hz to 12000 Hz. In inner loop 311 phase shifts of the pure tone between the
two ears are
stepped through for each frequency, for example 30, 60, 90, 120, and 180
degrees. These
two loops can be exchanged if desired.
8
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
Step 312 tests the subject's threshold for the pure tone at that frequency and
phase
shift, and stores the result, for example in a table. Step 314 determines the
optimal relative
phase shift between ears at each tested frequency, by determining at which
phase shift the
tone was heard best (at the lowest amplitude) over the background noise, for
each frequency,
compared to normal phase shift function. Thus a clinical phase shift versus
frequency
profile has been developed to compensate for the phase portion of the
subject's pathological
BPTD. The phase analysis test ends at step 320. This test procedure is also
used with
speech stimuli as the target signal using a phase correction filter and broad
band noise.
Figure 4 is a flow diagram showing delayed binaural fusion test 112. This test
1 0 measures the subject's ability to comprehend words (preferably bisyllabic)
at various time
delays between the two ears. Thus, it provides information regarding timing
differences
between the two ears and adds information beyond the phase analysis test.
Examples of the
tests done are given below.
The test starts at step 402. High Low Frequency Lags Test 404 tests
1 5 comprehension of a series of bisyllabic words separated into two frequency
components
(e.g. a high frequency component from 1900-2100 Hz and a low frequency
component
from 500-770 Hz) presented at a variety of relative time delays to the ears.
The purpose of
step 404 is to determine whether a significant impairment due to CANS binaural
phase-time
asynchrony exists for the subject. For a person without this type of
impairment, the change
2 0 in relative time delays does not significantly effect comprehension - the
CANS can account
for the changes. In addition, the best comprehension occurs for the case of no
relative time
delay, as would be expected. For a person with significant impairment due to
CANS
binaural phase-time asynchrony, however, different relative time delays result
in very
different levels of comprehension, and the best comprehension occurs at a
relative time delay
2 5 other than zero. If the results of step 404 indicate that a CANS-BPTD
impairment exists,
step 408 determines the optimal time delay for the subject.
Each word is presented to both ears, the high frequency portion of the word
going to
one ear and the low frequency portion of the word going to the other ear. The
relative time
delay between the ears is changed for each word, and a variety of words are
used at each
3 0 relative time delay. The words are generally bisyllabic, familiar to most
people, and the
emphasis is placed on both syllables equally (e.g. woodwork, bedroom,
inkwell). For
example, a series of 120 words may be used, divided among the selected
relative time delays.
Other speech stimuli can be used along with other novel ways to split or
partition out speech
segments. A computer program for sequentially selecting the words and setting
the relative
35 time delay for each word makes this process much easier. The program may
also provide a
9
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
score sheet for entering whether each word was correctly identified, and
computing the
correct averages at each phase shift.
Steps 404 and 408 zero in on the ideal clinical relative time delay, because
it is
difficult for a subject with CANS-BPTD impairment to understand a word if the
high and
low frequency components are not correctly time shifted relative to each
other, or when they
are lagged while being embedded in noise or speech-babble. In other words,
auditory
discrimination improves with an induced time delay in one ear for individuals
with a CANS
disfunction. Other speech modifications using a lag paradigm may be used for
identifying
and quantifying asynchronies.
1 0 The described tests are scored by computing the percentage of correct
responses
given by the subject at each relative time delay, and each step refines the
results of the
previous step. A software program for sequentially selecting the words and
setting the
relative phase shift for each word makes this process much easier. The program
also
provides a score sheet display for entering whether each word was correctly
identified, and
1 5 computes the correct averages at each phase shift when the test is
completed. The phase
shifts tested may be selected in view of the ideal overall clinical phase
shift that was
computed at the end of the pure tone or speech phase test of Figure 3, in
order to make this
test more efficient.
High-low frequency lags test 404 tests comprehension of a series of words at
(for
2 0 example) relative time delays of 5, 10, 15, and 20 msec to the left and
right ears. The best
comprehension level might be achieved at, for example, 5 msec time delay to
the right ear.
Incremental DBFT test 408 then tests comprehension of a series of words at
relative time
delays of 2.5, 5, and 7.5 msec (assuming a 5 msec delay gave the best results
in step 404).
The best comprehension level might be achieved at, for example, 7.5 msec time
delay.
25 Those skilled in the art will appreciate that further fine tuning can be
accomplished with
smaller relative time delays using the BPTD diagnostic device, if desired.
Zero delay word lists test 410 then verifies the results from steps 404 and
408 by
testing comprehension at the selected relative time delay, using a device such
as that shown
in Figure 5.
3 0 Step 412 stores the optimal time delay selected by the previous steps. The
test ends
at 414. A correction device such as that shown in Figure 6 may now be
designed, by
combining the results of the Phase Analysis Test shown in Figure 3 and the
DBFT test
shown in Figure 4.
Figure 5 is a block diagram of an electronic device 500 for diagnosing and
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
measuring the phase and time portions of BPTD, and for verifying the best
overall relative
time delay for the subject (see Figure 4, step 408). An operator controls the
relative time
delay and phase delay applied to the subject's ears via an operator interface
502. The test
setup shown in this figure tests one frequency (tone) at a time and applies
the selected phase
shift to whatever frequency is applied. The steps of the phase shift test are
shown in Figure
3.
Operator interface 502 may include, e.g. a keypad to enter control signals,
and a
display to show which control signal is being applied. Control signals 504 set
the amount
of relative time delay and phase shift to be applied by digital signal
processor 510. Sound
1 0 506 is digitized via channels 507 and 508 (or only one microphone may be
used). Relative
time delay shifter 512 applies the overall relative time delay per control
signals 504, and
phase shifter 514 applies the phase shift. The output of block 510 is
delivered to the left ear
522 of the subject 520 via signal 516, to the right ear 524 of the subject via
signal 518.
Figure 6 is a block diagram of real time, active, digital signal processing
electronic
1 5 device 600 for correcting BPTD, once it has been measured. Equivalent
analog devices
could also be used, but digital devices are more practical.
Sound enters microphone 602, which turns it into an analog electrical signal
representing the sound. The signal is amplified in preamp 604, and is
digitized in analog to
digital converter (ADC) 606. Digital signal processor (DSP) 608 operates much
as test
2 0 device 500 in Figure 5 operated, applying a time delay and a phase shift
profile (see Figure
7). Digital to analog converter (DAC) 610 converts the processed signal back
to an analog
signal, amplifier 612 filters and amplifies the signal, and microphone 614
turns the signal
into an audio signal to be delivered to the ear of the subject.
The BPTD applied by DSP 608 is programmed according to the BPTD versus
2 5 frequency profile obtained in the testing phase. It is unique for every
subject. As an option,
the DSP could be reprogrammable, via control signa1616, so it could be
optimized for the
wearer in actual use. Note that other hearing aid processing (compression or
the like) may
also be incorporated into the DSP if desired.
Furthermore, we have observed that BPTD's produce a very noticeable auditory
3 0 effect in the presence of noise regardless of whether the subject has CAP
difficulties or not.
The fact that the implementation of BPTD parameter changes produce differences
in target
signal (speech stimuli) perceived loudness in the presence of masking noise
indicates that
BPTD's have the potential for enhancing hearing aid performance
In addition, the BPTD profile used may change with the kind of background
noise
11
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
detected by the device. A different BPTD profile may be used when the wearer
is in a noisy
environment, for example, or for different actions, as when the wearer is
walking rather than
sitting and writing.
Figure 7 is a more detailed block diagram of DSP 608 of Figure 6. BPTD control
block 702 (via control signa1616, if used) controls the overall time delay and
the phase shift
profile applied to the sound signal. TD correction block 704 applies the
overall time delay.
BP correction block 706 applies a phase shift profile to the sound signal. See
Figure 9 for
an example of a phase shift profile.
Figure 8 is a flow diagram showing the process accomplished by the DSP of
Figure
1 0 6. The audio input signal is applied in step 802. In step 804, the overall
time delay is
applied. In step 806, the phase profile is applied. In step 808 other
processing is
accomplished, if desired (compression or the like). The corrected output
signal is output in
step 810.
Figure 9 is a diagram showing an example of the phase shift profile correction
1 5 accomplished in BP block 706 of DSP 608 of Figure 7. Dotted line 902
indicates linear
phase, and solid line 904 indicates the phase profile after the phase
corrections. ql, q2, and
q3 indicate phase shifts at specific frequencies f 1, f2, and f3 respectively.
Note that a
different phase shift is applied at each frequency, and that positive and
negative phase shifts
may be applied. Of course, the overall time delay applied by block 704 means
that, overall,
2 0 the time delay plus the phase shift will be positive.
Figure 10 is a cutaway side view of a physical filter (a passive earplug) for
correcting BPTD. A physical device in the ear can delay the sound in the ear,
and can delay
different frequencies differently, as electronic device 600 (in Figure 6)
does. However, a
physical device is capable of much smaller time shifts, and the control at
different
25 frequencies is far less precise. Since the physical device is much cheaper
and does not
require batteries, it is the preferred device is some cases, for example when
a small phase
shift or delay is required, and the phase shift required doesn't vary much at
various
frequencies. The physical filter is also smaller and more convenient.
Passive earplug 1000 induces a BPTD to sound entering the ear by altering the
3 0 propagation time of the acoustic waves. The primary method of delaying an
acoustic signal
in this manner is through the use of ducting 1002, through which the signal
propagates.
The velocity of propagation of sound in air is approximately 331 meters per
second, and the
length of the ducting in the ear canal is about 10 cm (ducting along an
eyeglass frame can be
longer). Thus the time delay applied by a passive device in the ear canal is
on the order of
12
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
30 us, corresponding to a phase shift of about p/3 at 5000Hz. This time delay
may be
increased by about a factor of two by using a fluid rather than air in ducting
1002. For
example, the velocity of sound in iodine is around 108m/s.
In addition to the overall delay created by ducting 1002, the frequency
response of
earplug 1000 may also be tuned to some degree by using acoustical filter
elements 1004
(limited by space available). Standard elements include chambers, Helmholtz
resonators,
and dampers. In addition, other acoustic elements such as horns, collectors,
domes,
trumpets, and resonator may be used.
A direct analog may be made to an electrical BPTD system such as that shown in
1 0 Figure 6, with Helmholtz resonators and expansion chambers used to create
filter
characteristics. The number of cavities relates to the order of filter that
can be designed.
In general, the phase time delay provided by passive element 1000 is dependent
on
the length of the auditory ducting 1002 within the plug, the diameters and
locations of the
cavities (side branch chambers 1008 or expansion chambers 1006) and the
working fluid in
1 5 the ducting.
While the exemplary preferred embodiments of the present invention are
described
herein with particularity, those skilled in the art will appreciate various
changes, additions,
and applications other than those specifically mentioned, which are within the
spirit of this
invention.
2 0 What is claimed is:
13
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
APPENDIX A
EXAMPLES OF TESTING PROCEDURES
Table 1
Delayed Binaural Fusion Test Table of Words
List 1 List 2 List 3 List 4 List 5 List 6
woodwork redbird doorbell sunset highway dishcloth
chalkboard desktop racetrack bookmark doormat handbell
shortcake playground treetop meatball grandson padlock
baseball icecream wildcat headlight toothbrush northwest
scarecrow lighthouse downtown shipwreck hardware hardhat
hatrack armchair whitewash bullfrog flagpole mushroom
housework corncob windmill handshake hairbrush eardrop
railroad highchair tshirt pancake airplane football
duckpond schoolbell bedspread bedroom inkwell drugstore
nightfall workshop bluejay iceberg keyhole toolbox
stairway drawbridge daybreak birdnest hottub teabag
beanbag goldfish driftwood dirtbike mousetrap lightbulb
horseshoe lefthand sidewalk farewell birdhouse sandbox
farmhouse shirttail billboard bathtub pinwheel cowboy
starfish snowball greyhound toothpick earthworm yardstick
doghouse eyebrow oatmeal daylight iceskate eardrum
cookbook birthday snowman schoolboy bustop rainbow
offshore blackboard stringbean shoelace thumbtack treehouse
swingset shortcut forehead dollhouse hothouse carwash
hotdog timeout cardboard lifeboat jailhouse footstool
14
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
Table 2
Delayed Binaural Fusion Test Time Delay
Sequencing Table
Condi- List 1 List 2 List 3 List 4 List 5 List 6 Lag-
tion ging
Chan-
nel
1 0 5 1 0 15 2 0 2 5 Low
Freq.
2 25 0 5 10 15 20 Low
Freq.
3 2 0 2 5 0 5 1 0 1 5 Low
Freq.
4 15 20 25 0 5 10 Low
Freq.
1 0 1 5 2 0 25 0 5 Low
Freq.
6 5 10 15 20 25 0 Low
Freq.
7 0 5 1 0 1 5 2 0 2 5 Low
Freq.
8 2 5 0 5 10 15 2 0 Low
Freq.
9 2 0 2 5 0 5 1 0 1 5 Low
Freq.
1 0 1 5 2 0 25 0 5 1 0 Low
Freq.
1 1 1 0 1 5 2 0 25 0 5 Low
Freq.
1 2 5 10 15 2 0 2 5 0 Low
Freq.
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
APPENDIX B
CLINICAL RESULTS
Clinical results indicative of the efficacy of the present invention are
provided
hereinafter. In particular, representative results will be provided for the
diagnostic
effectiveness of the delayed binaural fusion test (DBFT) and for the efficacy
of
binaural phase-time delay (BPTD) compensating devices as clinically
demonstrated
by subjects having BPTD impairments, wherein the compensating device
substantially alleviated the debilitating effects of such BPTD generated
impairments.
Delayed Binaural Fusion Tests (DBFT)
1 0 The results of the preliminary studies presented below demonstrate the
understanding of the significance and feasibility of the patent. In brief, the
inventor's
research documents that subjects with normal CANS function demonstrate optimal
BPTDs at 0 msec. In other words, their auditory systems function optimally
when
acoustic stimuli have a matched-timed onset to the two ears. In contrast,
these
1 5 preliminary studies clearly demonstrated that subjects with CANS
dysfunction show
that matched-timed onset of acoustic signals (i.e., a 0 BPTD) do not result in
optimal
auditory function. In fact, optimal auditory function can only be obtained
when a
BPTD is induced between the two ears. The degree of the BPTD is quantified by
DBFT results that are specific to each individual. The following descriptions
of these
2 0 preliminary studies will clarify what is meant by a BPTD and how a BPTD is
induced.
The first study assessed the effects of BPTDs on auditory performance for
individuals with both normal and atypical CANS function using the DBFT in a
high-pass and low-pass frequency filtered format (404). This investigation
2 5 concentrated on ascertaining the percentage change in auditory
discrimination ability
of bisyllabic words presented in a binaural interaction format at various msec
lag
16
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
times. A total of 115 subjects from 12 to 58 years of age were included in
this study.
Forty subjects were included in the normal CANS function group and 75 in the
a ical CANS function group. The mean age for the normal CANS group was 25.74
with a range of 12 to 56. The mean age for the atypical CANS group was 20.3
years
with an age range of 12 to 58 years. Selection for each group was determined
by the
results obtained on the central auditory processing (CAP) test battery. This
battery
included the Competing Sentences, Filtered Speech, and Binaural Fusion Tests
of the
Willeford Central Auditory Test battery (Willeford, 1977), the
Ipsilateral/Contralateral Competing Sentence Test (IC/CST) (Willeford 1985a,
1985b,
1 0 Willeford et al., 1985; Willeford et al., 1994), Synthetic Sentence
Identification-Ipsilateral Competing Messages (SSI-ICM) (Jerger et al., 1974,
1975;
Speaks et al., 1965), Dichotic Digits (Musiek, 1983; Musiek et al., 1979;
Musiek et
al., 1979) and Masking Level Differences (MLD) (Noffsinger et al., 1972; Olsen
et al.,
1976). Using one-way ANOVAs, test performance showed significant differences
1 5 between the two groups as defined by performance on this test battery. A
former
study by Burleigh (1996) clearly showed significant differences between normal
CANS function and atypical CANS function groups when using this test battery
and
agreed with these findings
The low-pass and high-pass frequency filtered format, one version of the
2 0 DBFT (404), and shown in Table 1, was used to quantify inherent BPTDs
between
ears. Statistically significant differences in speech recognition performance
between
normals and atypicals in the DBFT study of 115 subjects were observed.
Significant
differences in percent performance were also evidenced in all conditions for
both ears
except for a left ear lag at a 15msec delay. The reason for this decrease in
function for
2 5 both normal and atypical groups at a 15msec lag deserves further study.
One can
speculate that interaural timing for the left ear may reflect transfer of
information at
the level of the corpus callosum.
Further analysis showed that 78.67% of the atypical group showed a 20% or
17
CA 02386653 2002-04-05
WO 01/26420 PCTIUSOO/27460
better increase in speech discrimination ability in noise with induced or
compensating
BPTDs as compared to the Omsec-lag condition. For those with normal CANS
function, only 35% showed 20% or better improvement with the implementation of
various lag times as compared to Omsec lag. Figure 11 shows the percent of
individuals who improved 20% or better for speech discrimination ability for
both
groups as compared to each subject's Omsec score.
Statistically significant differences in auditory discrimination performance
between normals and atypicals at 5msec lag intervals are shown in Figure 12.
As
demonstrated in this figure, significant differences in subjects who improved
at least
1 0 20% from 0 msec were observed between groups for auditory conditions of 5
msec
through 25 msec lags as compared to a baseline of 0 msec (absence of a time
lag).
Standard Passive Earplugs
Initial clinical exploration of BPTDs involved the use of commercially
available
noise reduction filters that were originally designed for occupational safety.
The
1 5 specific earplug that has been used for this study is designed to
attenuate damaging
sounds while maintaining the amplitude of conversational frequencies in order
to
perform routine tasks while wearing the earplug. It is primarily used for
industrial
purposes. The earplug that was used for this portion of the clinical study is
made of
polyvinylchloride in a half shell earmold. This standard filtered earplug was
modified
2 0 to get the following clinical data. Due to the non-specific design of this
earplug, only a
single time delay was possible that was altered slightly by modifications to
the length
of the ear mold portion of the plug. The current art regarding diagnostics, as
described
earlier in this document, requires more flexibility with design and materials
to
accomplish closer control of and variations of the BPTDs. In particular it is
evident
25 that greater control of BPTDs at targeted frequencies is needed based on
the diagnostic
results. In particular, to utilize the commercially available earplug, we had
to shorten
the canal length in order to reduce the attenuation of the earplug while still
providing a
18
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
significant time delay from the filter elements. By shortening the length of
the ear
canal we could use the hard surface of the external auditory meatus itself for
frequency
filtering. Optimally, to achieve better performance with frequency filtering,
a harder
material should be used in the canal portion of the earplug. Further, to
accommodate
variations in phase differences between individuals with CANS dysfunction,
different
frequencies will be altered using acoustic filter elements within the canal
and by
altering the acoustic impedance of the materials surrounding the filter
elements.
Commercially available filters are designed for wearing binaurally to reduce
damaging noise while maintaining speech. It is not an intentional effect of
the filter to
1 0 produce a time delay in the acoustic signal. However, the inventors
recognized that
monaural use of such a filter was a passive filter approach to inducing a
BPTD. The
modifications to the filter to change the length of the filter were done on
custom
trial-and-error basis. The filter, with the resultant notched frequency
configuration at
2000-3000 Hz, when worn monaurally, has the effect of a BPTD. The proposed
1 5 passive device would be designed specifically for inducing a BPTD, and
could be
designed to induce the a predetermined BPTD obtained from the testing
procedure
described above. In addition, the BPTD could be limited to a particular
frequency
range (depending on PAT results) for specific amplitudes within the
limitations of the
capabilities of passive filtering elements. This would provide the ability to
control the
2 0 induced BPTD for individual fitting based on the diagnostic results.
However, as
mentioned earlier, the current art does not offer flexibility with size of
BPTDs at
specific frequencies for optimal auditory and human performance enhancement.
In order to understand the time delay in a plug, it is necessary to use a
model
that includes the complex response characteristics of the ducts. This is
similar to the
2 5 familiar insertion loss calculations that are performed in normal design;
however,
instead of only considering the magnitude of the response; the phase angle is
also
included. The model design however can include significant simplifications due
to the
long wavelength of the sound relative to the diameter of the ear plug duct.
19
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
Note that unilateral muffing does not result in a BPTD due to the gradual
attenuation across all frequencies (i.e., no selective notch). With the muffs,
from 250
Hz to 6000 Hz there is a 5-10 dB attenuation per octave.
Speech recognition scores were obtained in the sound field with the subject
seated 3 feet from a front facing speaker. Monosyllabic words were presented
at 40
dB SL re: sound field speech reception threshold in the presence of broad band
noise
(s/n= +5) presented from a back speaker.
Figure 13 shows speech recognition performance in noise for 22 individuals
with
CANS dysfunction. Five different conditions are shown in the figure.
1 0 1."no plug": nothing in either ear
2."RE muffed": hearing protection muff (noise reduction rating 25 dB - ANSI
S 12.6) worn over right ear
3."LE muffed": hearing protection muff (noise reduction rating 25 dB - ANSI
S 12.6) worn over left ear
1 5 4."unilat plug": standard filtered earplug worn in one ear only.
5."bilat plug": standard filtered earplugs worn in both ears.
Of these five conditions, only condition 4 introduces a BPTD. Conditions 2 and
3
result in unilateral noise reduction and condition 5 results in bilateral high
frequency
noise reduction. All differences are statistically significant (p < 0.05),
except for the
2 0 three conditions, which included the left ear muffed, and no plugs,
bilateral plugging
and no plugging, and left ear muffed and bilateral plugging. Of particular
interest is the
significant improvement in speech recognition with an induced BPTD ("unilat
plug"),
P < 0.0001. Noise reduction without a BPTD (unilateral muff or bilateral
earplugs)
does not result in enhanced speech recognition.
2 5 Figure 14 shows speech recognition ability in noise results for 12
individuals
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
without CANS dysfunction. Five different conditions are shown in the figure.
1."no plug": nothing in either ear
2."RE muffed": hearing protection muff (noise reduction rating 25 dB - ANSI
S 12.6) worn over right ear
3."LE muffed": hearing protection muff (noise reduction rating 25 dB - ANSI
S 12.6) worn over left ear
4."RE plug": standard earplug worn in right ear only.
5."LE plug": standard earplug worn in left ear only.
The "no plug" results are significantly (p < 0.01) greater than any of the
other
1 0 conditions. These results demonstrate the importance of synchronous
binaural
processing of auditory input for enhanced speech discrimination in noise for
individuals with a normal CANS. Furthermore, these results indicate that
unilateral
noise reduction (conditions 2 and 3) or introducing a BPTD (conditions 4 and
5) do
not enhance speech discrimination in noise for individuals with normal CANS
1 5 function.
Note that speech discrimination results for the atypical group under condition
1("no
plug") are significantly lower than those for the normal group under condition
1("no
plug"). Note also that under condition 4 ("unilat plug"), the atypical group
performs
at approximately the same level as the normal group does under condition 1("no
2 0 plug").
Preliminary Data of BPTDs on Human Performance Using Electronic
Device
Ten normal CANS subjects and 10 atypical CANS subjects were selected for
various human performance testing using the prototype BPTD electronic device.
The
2 5 mean age of the 10 normal subjects was 31 years with an age range of 21-43
years. In
21
CA 02386653 2002-04-05
WO 01/26420 PCT/USOO/27460
this group, five individuals were male and five were female. In the atypical
CANS
group, the mean age was 29.1 years with an age range of 15 to 47 years. Seven
females and 3 males were included in this group. All subjects in the two
groups
passed the initial subject selection criteria. The ten subjects that were
included in the
atypical CANS group failed at least one test in either ear in the CAP test
battery.
Further, to provide for a balancing of BPTDs for the normal group, subjects
for the
atypical group were selected when they showed maximum improvement for the
DBFT with a 2.5-7.5 msec BPTD to either the right or left ear.
Another version of the DBFT (406) was developed to include time-lagged
1 0 bisyllabic stimuli that were presented in 2.5 msec increments (the
previous DBFT
used 5 msec increments). This version included thirty bisyllabic words per
list (4
lists) that were lagged in time between ears of 0 msec, 2.5 msec, 5 msec, and
7.5 msec
and recorded in a CD format with an 8-talker babble (s/n ratio of +2 dB)
embedded in
the background and presented binaurally. These words were presented under
1 5 earphones at 40 dB sensation level relative to pure tone averages for both
ears. The
lag ear was determined by results from the DBFT 5 msec version (404). The
highest
speech recognition percent score for bisyllabic words was determined to be the
"optimal" msec setting for the BPTD device for atypical CANS subjects. The
optimal condition for normal CANS subjects was randomized at 2.5 msec, 5 msec,
and
2 0 7.5 msec. Statistical analysis showed that these word lists were
equivalent.
The 5 auditory conditions used in the preliminary studies were as follows: (1)
natural condition which was unoccluded with 8-talker babble presented at 40 dB
HL,
(2) quiet which consisted of an unoccluded condition with 8-talker babble
presented at
2 5 25 dB HL, (3) optimal condition using the BPTD electronic device in the
presence of
40 dB HL 8-talker babble; (4) 0 msec condition using the BPTD electronic
device
with 40 dB HL 8-talker babble, and (5) opposite BPTD (same setting but in
opposite
ear) from the optimal setting (e.g., if "optimal" was a 5 msec lag to the
right ear, then
"opposite" is a 5 msec lag to the left ear). Not all tests examined condition
5.
22
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
Auditory Discrimination
Percent improvement in speech recognition ability using the BPTD device was
assessed using four thirty-bisyllabic word lists that were recorded on a CD at
0 msec
(408). Auditory stimuli were presented in the sound field from a front facing
speaker
two meters from the individual. Test stimuli were presented at 45 dB sensation
level
(re: sound field speech reception threshold) in the presence of an 8-talker
babble
recorded at a +2 s/n ratio. Both stimuli were presented from the front speaker
in a
double-walled sound proof room.
1 0 Results (see Figure 15) under these conditions showed a type effect across
conditions (p=0.0039). There was also a condition effect (p=0.0474). The
atypical
group showed the greatest percent improvement for speech recognition ability
with
the BPTD device set at their "optimal" lag time.
Motor
1 5 Gait studies were performed in a controlled uniform sound environment -- a
semi-anechoic chamber. Reverberation times were long enough (or amplitudes
sufficiently low) that the room was taken as representative of the sound field
in an
open or large room with high damping. The ambient sound level in the chamber
was
45 dB SPL (Metrosonics dosimeter, Model dB307, Class Type 2A, Rochester, NY).
2 0 Sound sources were then introduced into the chamber under controlled
amplitude and
directionality. This is a highly controlled sound field relative to all
previous gait
studies in the literature.
Speakers were placed in the anechoic chamber at 0( (far left), 45(, 90(
(center),
135(, and 180( (far right) locations in a clockwise direction relative to the
direction of
2 5 travel. The five sound sources were randomly presented to create the
general localized
sound condition (LS). Two additional cases were run in the chamber without
speaker
input: walking with and without earmuffs (Peltor, Model H6A/V) created the
general
23
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
reduced sound level condition (RS). For the BPTD device study, the center
speaker
was used.
The speaker output (i.e., sound source) was a tape-recorded eight-person
multi-talker babble presented at a sound level that was within three dB of a
56 dB SPL
in the chamber calibrated gait area. To reduce the influence of visual
stimuli, all
materials used in the chamber were monochromatic (i.e., either gray or black)
and the
room lighting was reduced to approximately 0.9 footcandles (equivalent to a
moderately lighted parking lot).
A calibrated three-dimensional video gait analysis (Peak MotusTM, Englewood,
1 0 Colorado) was completed with three camera views. The three cameras
recorded each
subject walking straight ahead within the calibrated area at a comfortable
pace for two
strides. The subject repeated each condition until three to five gait cycles
were
recorded between consecutive heel strikes of the same foot. As part of the
motion
analysis system, retro-reflective markers were mounted on the skin using 3MTM
1 5 hypoallergenic double-sided tape. On the head, two markers were placed on
a
spandex swim cap pointing upward directly above the ears. Body markers were
placed on the vertebra prominens (C7 vertebra), shoulders, elbows, wrists,
greater
trochanters, knees, and ankles.
To normalize the gait data for subject stature, the kinematic data were
divided
2 0 by the subject's height in meters. The following gait parameters will be
presented:
walking speed (% height/sec), stride width (%height), and center of mass
position
(COM) (% height). The relative COM (COM Del) was the difference of the lateral
COM position from the origin between the first heel strike and the time of
measurement. The Root Mean Square Error (RMSE) method was used to calculate
2 5 the mean lateral deviation from a straight-line path of the COM.
A study was performed assessing gait while using the electronic device set at
24
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
an optimal BPTD (Burleigh et al., 1999). Optimal BPTD was determined by the
DBFT results, (i.e., if a subject had the highest DBFT score with a 5msec
delay to the
right ear, a right 5msec delay was the "optimal" setting). The BPTD device was
used
with 6 normal and 6 atypical subjects in three different conditions:
"optimal," "0
msec," and "opposite" as described above.
Figures 16, 17, and 18 show the walking speed, stride width and RMSE results
for the atypical and normal groups under the different BPTD device conditions.
The
data graphically presents the differences in these parameters between the
three BPTD
device conditions. For example, the first bar in Figure 16 shows the average
of the
1 0 differences in walking speed for each of the atypical subjects between the
"optimal"
and "Omsec" conditions.
It is evident that atypical group's gait is significantly improved (i.e., they
walk faster)
under both the "optimal" and "opposite" BPTDs, compared to the absence of a
BPTD (0 msec), while the BPTD settings have little effect on the normal
subjects'
1 5 gait. The differences between the atypical and normal groups in the
walking speed
figure are not quite statistically significant (p=0.074 for "optimal-0" and
p=0.064 for
"opposite-0'"). Given the relatively small number of subjects in this study,
and large
variation in gait parameters, this was not surprising. However, the trends are
quite
strong and with a larger number of well-matched (on gender and age) subjects,
2 0 significant differences are expected.
The results for stride width (Figure 17) and Root Mean Square Error (RMSE)
(Figure 18) are similar, showing an improvement in gait (decrease in stride
width and
RMSE) for the atypicals under both the "optimal" and "opposite" conditions,
and
little effect on the gait of the normals. The differences between the
atypicals and
2 5 normals for stride width were not statistically significant for the
"optimal-0" case
(p=0.1042), but were for the "opposite-0" case (p=0.0422). For RMSE (Figure
18)
the differences between the atypicals and normals were not significant
(p=0.1552 and
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
p=0.1012, respectively for "optimal-0" and "opposite-0").
The decrease in stride width and RMSE is seen as an improvement in gait as
the decreases tended to bring the values of these parameters for the atypicals
closer to
the values for the normals. While the gait of the normal subjects was impacted
very
little by the device, as compared to the atypical group, the graphs indicate a
trend that
the "optimal" and "opposite" BPTDs actually degrade the normal subjects' gait.
This
trend supports the notion that while an induced BPTD in a normal CANS system
is
disruptive, it can be accommodated.
It was surprising that the "optimal" and "opposite" settings both seem to
1 0 enhance gait in the atypical group. Perhaps this is evidence that while
BPTDs impact
gait, the effect is different from the impact of BPTDs on other aspects of
human
performance, such as speech discrimination.
Interestingly, most changes in individual atypical subjects' gait under the
different BPTD device conditions were significant. For example, most atypical
1 5 subjects walked significantly faster under the BPTD device "optimal"
setting when
compared to their walking speed under the "0 msec" setting, using multiple
trials
under each condition as the multiple measures.
Speech
Acoustic measures of diadochokinetic rate or maximum repetition rate for
2 0 non-speech material includes: duration in msec of 5 correct syllable
sequences out of 7
consecutive utterances. This measure was used as an assessment of articulatory
speed; however, because only correct syllable productions were counted, it
probably
more accurately reflected articulatory efficiency. Syllable and pause
durations were
also measured as were irregularities or variances among successive syllable
and pause
2 5 durations within each condition.
26
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
Results of performance of these tasks under auditory conditions in normal and
atypical subjects revealed statistically significant differences in
articulatory efficiency,
syllable duration and variances in syllable and pause duration when comparing
those
persons with normal and atypical CANS function (p <.05 for each). This
suggested
that while all subjects were considered normal speakers, differences in
abilities to
make rapid alternating movements differ in persons with atypical versus normal
CANS function.
There was also a statistically significant difference in articulatory
efficiency
and intersyllabic pause durations among experimental auditory conditions for
both
1 0 groups
(p < .05). Durations for completions of 5 accurate syllable sequences and
intersyllabic pauses were shorter under the optimal or accommodating auditory
condition compared to conditions of 0 msec delay or reduced noise. These
findings
suggest that not only do auditory conditions impact the performance of rapidly
1 5 alternating nonspeech movements, but that specific adjustments in binaural
timing
between ears may improve performance over conditions where no differences are
introduced.
In oral reading of an 85-word paragraph, perceptual measures were taken of
the number of dysfluencies and reading speed in words per second. Perceptual
2 0 dysfluency types included: part and whole word repetitions, phrase
repetitions/restarts, prolongations, phonatory disruptions, interjections,
blocks, and
pauses. This system of dysfluency classification was modified from Kent, 1994.
A review of number of dysfluencies in oral reading provided the most
convincing evidence of benefits of accommodating BPTDs. While there were
2 5 statistically significant differences in the total number of dysfluencies
between normal
and atypical experimental groups (p < .05), for this parameter there was also
a
statistically significant condition-type interaction (p < .05) suggesting that
the
27
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
differences in performance in favor of the normal subjects was a function of
the
auditory condition. Specifically, as can be seen in Figure 19, not only do the
persons
without CANS dysfunction show a flatter profile of performance across auditory
conditions when compared with those with CANS deficits, those with atypical
CANS function make the fewest errors, performing very close to normal at their
optimal or accommodated auditory condition. In fact, statistically significant
differences are apparent between normals and atypicals under natural and 0
msec
conditions but are not at the accommodating condition. Further, the
performance of
atypicals in pairwise condition comparisons, reveals statistically fewer
dysfluency
1 0 errors made under optimal or accommodating auditory condition when
compared
against each: 0 msec, natural and reduced noise conditions.
Reading speed as measured by words per second in an 85-word oral reading
sample revealed statistically significant differences when comparing those
persons
with normal and atypical CANS function (p < .05). It should also be noted that
1 5 although statistically significant differences were not observed in
pairwise
comparisons across auditory conditions for the subjects with atypical CANS
function, the fastest reading rate condition for this group was their optimal
or
accommodated condition.
THE COMMERCIALLY AVAILABLE STANDARD PASSIVE EARPLUGS
2 0 USED IN THIS STUDY IS NOT AN EMBODIMENT OF THE INVENTION (see
paragraph below) BUT ILLUSTRATES VARIOUS FEATURES OF THE
METHOD UTILIZED AND DESCRIBED HEREIN.
The foregoing discussion of the invention has been presented for purposes of
illustration and description. Further, the description is not intended to
limit the
2 5 invention to the form disclosed herein. Consequently, variation and
modification
commensurate with the above teachings, within the skill and knowledge of the
relevant
art, are within the scope of the present invention. The embodiment described
28
CA 02386653 2002-04-05
WO 01/26420 PCT/US00/27460
hereinabove is further intended to explain the best mode presently known of
practicing the invention and to enable others skilled in the art to utilize
the invention
as such, or in other embodiments, and with the various modifications required
by their
particular application or uses of the invention. It is intended that the
appended claims
be construed to include alternative embodiments to the extent permitted by the
prior
art.
29
