Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Intensity discrimination in single and multi-electrode patterns in cochlear implants
(USC Thesis Other)
Intensity discrimination in single and multi-electrode patterns in cochlear implants
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INTENSITY DISCRIMINATION IN SINGLE AND MULTI-ELECTRODE
PATTERNS IN COCHLEAR IMPLANTS
by
Ashmita Gaur
A Thesis Presented to the
FACULTY OF THE VITERBI SCHOOL OF ENGINEERING
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOMEDICAL ENGINEERING)
May 2007
Copyright 2007 Ashmita Gaur
ii
ACKNOWLEDGEMENTS
I would like to express my gratitude to my advisor, Dr. Robert Shannon, for his
guidance, motivation, expertise and patience which have added considerably to my
graduate experience. I thank the other members of my committee, Dr. Ellis Meng and
Dr. Ernest Greene for their assistance. I also thank John Galvin for providing me
direction and support at each step of my research. I am grateful to Mark Robert and
Qian-Jie Fu for providing the experimental software. Finally, I would like to thank the
CI subjects for their time. This work was supported in part by NIDCD.
iii
TABLE OF CONTENTS
Acknowledgements……………………………………………………………………...ii
List of Tables…………………………………………………………………………….v
List of Figures…………………………………………………………………………..vi
Abstract………………………………………………………………………………..viii
1. Introduction………………………………………………………..........................1
1.1 Normal Hearing…………………………………………………………................1
1.2 Deafness……………………………………………………………………….…..1
1.3 Cochlear Implants………………………………………………….........................2
1.4 Psychophysics………………………………………………………………….…..5
1.5 Channel Interactions in Cochlear Implants……………………………….……….8
2. Experiment I: Effect of Space Profile on Intensity Discrimination using a Long
Probe (200ms) ……………………………..……………………………………...9
2.1 Objective………………………………………….………………………………..9
2.2 Methods……………………………………………….………………………….10
2.3 Results………………………………………………………….………………...12
2.4 Discussion……………………………………………………………….………..14
3. Experiment I: Effect of Space Profile on Intensity Discrimination using a Short
Probe (20ms) ……………………………………………………………….……14
3.1 Objective……………………………………………………………………….…14
3.2 Methods……………………………………….………………………………….15
3.3 Results……………………………………………….…………………...............17
3.4 Discussion………………………………………………….……………………..21
4. Experiment I: Effect of Inter-channel Temporal Offset………………………….21
4.1 Objective……………………………………………………………….…....……21
4.2 Methods…………………………………………….…………………………….22
4.3 Results…………………………………………………….…………………...…25
4.4 Discussion………………………..……………………………………………….29
iv
5. Conclusion………………………………………………………………..............29
5.1 Effect of Space Profile…………………….…………………………………...…29
5.2 Effect of Temporal Profile…………………...........………………………...........30
Bibliography…………………………………………………………………................32
v
LIST OF TABLES
Table 1: CI Subject Information for Experiment I
10
Table 2: CI Subject Information for Experiment II
15
Table 3: CI Subject Information for Experiment III
23
vi
LIST OF FIGURES
Figure 1: Components of a Cochlear Implant
2
Figure 2: Schematic representation for a 2-Alternative Forced Choice (2-
AFC) trial
7
Figure 3: Electrode Configuration for Experiment I depicting presentation of
probe electrode (11,13) in between two pairs of surrounding
masker electrodes in three different configurations: Far (masker
6.75mm from probe), Medium (masker 4.5mm from probe) and
Close (masker 2.25 mm from probe)
11
Figure 4: Intensity DLs (%) for Subjects S1-S4 as a function of probe
percent DR
13
Figure 5: Short Probe (20ms) interleaved in time between Two Maskers
(200ms each) such that the probe pulse is centrally located
between the two masker pulses
16
Figure 6: Probe centered between two maskers; Intensity DLs were
measured at loudness balanced reference levels at combinations of
masker and probe levels for 10%, 30% and 50% of the electrode’s
DR
17
Figure 7: (a) Intensity DL (%) vs. probe level for Subject S5 for three
different configurations of masker and probe parameters
(b) Intensity DL (%) vs. masker level for Subject S5 for three
different configurations of masker and probe parameters
(c) Normalized Intensity DL (%) vs. probe level for Subject S5 for
three different configurations of masker and probe parameters
18
19
20
vii
Figure 8:
(a) Interleaving of masker and probe pulses: Probe onset at 500 µs
(b) Interleaving of masker and probe pulses: Probe onset at 3500
µs
24
Figure 9: (a) Intensity DL versus inter-channel delay for Subject S1 at probe
levels 30 % and 50 % of the DR
(b) Intensity DL versus inter-channel delay for Subject S2 at probe
levels 30 % and 50 % of the DR.
(c) Intensity DL versus inter-channel delay for Subject S3 at probe
levels 30 % and 50 % of the DR
26
27
28
viii
ABSTRACT
Single-channel psychophysics may not reflect performance with dynamic, multi-
channel stimuli (e.g., speech stimuli) in cochlear implant (CI) listeners due to
interactions between electrodes at the periphery or due to central processing interactions
or both. The first part of the study measured single and multi-channel intensity
discrimination in 5 Nucleus CI users, as functions of the relative level, electrode
location and stimulation rate of the masker and probe electrodes. Spatial distance and
relative stimulation levels between interleaved pairs of electrodes were found to
significantly influence the degree of channel interaction. The second part of the study
investigated how temporal offset between interleaved pulses may affect channel
interaction at suprathreshold levels in 3 Nucleus CI users. Results showed that
interactions were elevated for short temporal offsets and were reduced as the temporal
offset was increased to half the stimulation period.
1
1. INTRODUCTION
1.1 NORMAL HEARING
Sound undergoes a series of transformations as it travels through the outer ear,
middle ear, inner ear, auditory nerve and into the brain. The outer ear picks up acoustic
pressure waves that are converted to mechanical vibrations by a series of small bones in
the middle ear. In the inner ear, the cochlea, a snail-shaped cavity filled with fluid, the
mechanical vibrations are transformed to vibrations in fluid which lead to displacements
of a flexible membrane, called the basilar membrane. These displacements contain
information about the frequency of the acoustic signal. Attached to the basilar
membrane are hair cells that bend when the basilar membrane gets displaced. When
these hair cells bend, they cause neurons to fire. The neurons communicate with the
central nervous system and transmit information to the brain via the auditory nerve.
1.2 DEAFNESS
If the hair cells are damaged, there is no way of communicating auditory
information to the brain. The hair cells can be damaged by certain diseases such as
meningitis, Meniere’s disease, congenital disorders, or certain drug treatments.
Damaged hair cells can lead to degeneration of adjacent auditory neurons. Research has
shown that the most common cause of hearing loss is the loss of hair cells rather than
loss of auditory neurons. Cochlear Implants bypass the normal hearing mechanisms of
2
the outer, middle and inner ear by direct electrical stimulation of the remaining
auditory neurons.
1.3 COCHLEAR IMPLANTS
The cochlear prosthesis is one of the most successful neural prosthesis that uses
electric stimulation to enhance or restore human neural, sensory and motor function. All
the implant devices have the following features in common: a microphone that picks up
the sound, a signal processor that converts the sound into electrical signal, a
transmission system that conveys the electrical signals to the implanted electrodes, and
an electrode or an electrode array (consisting of multiple electrodes) that is inserted into
the cochlea by a surgeon.
Shown below is a picture of a typical cochlear implant:
Figure 1: Components of a Cochlear Implant
3
Sound is picked up by the microphone (a), where it is sent down a cord (b) to the
speech processor (c). The speech processor filters the incoming acoustic signal into
separate frequency bands, and then converts this information into digital form. The
digitized signal is sent back up the cord to the transmitting coil (d), which is held in
place by a magnet. The coil transmits the digitized information across the skin to the
internal receiver/stimulator (e). The internal receiver/stimulator decodes the incoming
signal and sends information in each frequency band to a different electrode within the
cochlea (f). High-frequency information is sent to electrodes in the basal end of the
cochlea (f), and low-frequency information is sent to electrodes in the apical end of the
cochlea (g). The auditory nerve (h) picks up the electrical signals from the electrodes
and relays that information to the auditory cortex in the brain where it is interpreted as
sound.
In single-channel implants only one electrode is used. In multi-channel cochlear
implants, an electrode array is inserted in the cochlea so that different auditory nerve
fibers can be stimulated at different places in the cochlea, thereby exploiting the place
mechanism for coding frequencies. Different electrodes are stimulated depending on the
frequency of the signal. Electrodes near the base of the cochlea are stimulated with high
frequency signals, while electrodes near the apex are stimulated with low frequency
signals. The signal processor is responsible for breaking the input signal into different
frequency bands or channels and delivering the filtered signals to the appropriate
electrodes. The main function of the signal processor is to decompose the input signal
4
into its frequency components, much like how a healthy cochlea analyzes the input
signal into its frequency components.
The perceived loudness of the sound may depend on the number of nerve fibers
activated and their rates of firing. If a large number of nerve fibers are activated, then
the sound is perceived as being loud. Likewise, if a small number of nerve fibers are
activated, then the sound is perceived as being soft. The number of fibers activated is a
function of the amplitude of the stimulus current. The loudness of the sound can
therefore be controlled by varying the amplitude of the stimulus current.
The pitch, on the other hand, is related to the site in the cochlea that is being
stimulated. Low pitch sensations are elicited when electrodes near the apex are
stimulated, while high pitch sensations are elicited when electrodes near the base are
stimulated.
In summary, the implant can effectively transmit information to the brain
containing the loudness of the sound, which is a function of the amplitude of the
stimulus current, and the pitch, which is a function of the place in the cochlea being
stimulated.
5
1.4 PSYCHOPHYSICS
Psychophysics is a discipline of psychology dealing with the relationship
between physical stimuli and their subjective correlates, or percepts. Psychophysicists
usually employ experimental stimuli that can be objectively measured, such as pure
tones varying in intensity, or lights varying in luminance. Experiments seek to
determine whether the subject can detect a stimulus, identify it, differentiate between it
and another stimulus, and describe the magnitude or nature of this difference.
Difference Limen (Dl):
The difference limen (DL) is the smallest difference in a specified modality of
sensory input that is detectable by a human being or other animal. It is the difference in
stimuli that the subject notices some proportion p of the time (50% is usually used for
p). It is also known as the just noticeable difference (JND) or the differential threshold.
Intensity Discrimination:
In auditory psychophysics, the DL is an increasing function of the base level of
input, and the ratio of the two is roughly constant. Measured in physical units, we have
6
Where is the original intensity of stimulation, is the increase in intensity
for the difference to be perceived (the DL), and k is a constant. It is true, at least to a
good approximation, of many sensory dimensions, for example the brightness of lights,
and the intensity and the pitch of sounds.
The DL is a statistical, rather than an exact quantity: from trial to trial, the
difference that a given person notices will vary somewhat, and it is therefore necessary
to conduct many trials in order to determine the threshold. The DL usually reported is
the difference that a person notices on 50% of trials.
Masking:
Auditory masking is when the perception of one sound is affected by the
presence of another sound. Masking can be simultaneous or non simultaneous. In
simultaneous masking, a signal, the sound that is desired to be heard (called the
“probe”), is made inaudible by a masker, noise or unwanted sound that is present
throughout the signal. The basic masking test involves the unmasked thresholds being
measured on a subject (called the probe). Then the masking noise is introduced at a
fixed sound level and the signal is presented at the same time. The level of the signal on
the probe is varied until the new threshold is measured. This is the masked threshold.
The phenomenon of masking is often used to investigate the auditory system’s ability to
separate the components of a complex sound.
7
Forced Choice Methods:
In forced choice, the subject is presented with a number of spatial or temporal
alternatives in each trial in which the stimulus is presented. The subject is forced to
choose the location or interval in which the stimulus occurred.
In a 2-alternative forced choice (2AFC) procedure: two temporal alternatives are
presented to the subject for each trial, and the subject selects the interval in which the
sound occurred. A schematic representation of a trial is presented below:
Figure 2: Schematic representation for a 2-Alternative Forced Choice (2 AFC)
trial
Each of the two intervals is preceded by a signal, a beep, which indicates the
intervals. The sound to be distinguished is presented randomly in either the first or
second interval. The subject's task is to respond with which interval the sound was
presented in.
interval 2 interval 1
“beep” “beep”
8
1.5 CHANNEL INTERACTIONS IN COCHLEAR IMPLANTS
Multi-channel cochlear implants provide better speech recognition than single-
channel cochlear implants due to more number of channels to sample information and
deliver information to the frequency selective cochlea. However, users show little or no
improvement in channel interaction when the number of activated channels is increased
beyond about 4 to 7 in quiet or beyond 7 to 10 in noise. On the other hand, normal-
hearing listeners continue to show improvement in speech recognition in quiet as well
as noise, as the number of channels is increased to 20 in an acoustic vocoder simulation
of a cochlear implant speech processor. This can be attributed to multi-channel cochlear
implants exhibiting channel interactions due to overlapping neuronal populations.
Channel interactions limit the number of information channels that can be transmitted to
the brain.
One cause of overlapping activation of neural populations is the summation of
electrical fields presented simultaneously through nearby electrodes. The summation
can be minimized if electrical pulses are interleaved in time, that is, pulses are never
presented simultaneously. Such strategies of nonsimultaneous pulsatile simulation are
common in clinical use, such as Continuous Interleaved Sampling (CIS), Spectral Peak
(SPEAK) and Advanced Combined Encoding (ACE).
We conducted a series of experiments using the Cochlear Corporation Nucleus®
device, to quantify channel interactions. In the first two experiments, we measured
effect of space profile on intensity discrimination for long and short probe conditions.
9
We found that the spatial distance and relative stimulation levels between
interleaved pairs of electrodes significantly influenced the degree of channel interaction.
When the spatial distance is small, or when the spread of excitation is large (e.g., broad
stimulation modes, high current levels), stimulation of two interleaved electrodes may
target the same neural population, giving rise to temporally-based channel interactions.
In the third experiment, we tried to examine how the temporal offset between
interleaved pulses may affect channel interaction at suprathreshold levels. Hence the
first two experiments demonstrated the space profile of a cochlear implant, whereas the
last two experiments demonstrated the temporal profile.
2. EXPERIMENT I: EFFECT OF SPACE PROFILE ON INTENSITY
DISCRIMINATION USING A LONG PROBE (200ms)
2.1 OBJECTIVE
Research has shown poor correlations between psychophysical capabilities and
speech recognition in normal hearing, hearing impaired and cochlear implants. One
possible reason for this lack of correlation is that the psychophysical tests are typically
performed on single electrodes whereas speech presents a dynamically changing
stimulus pattern across the entire electrode array. It is possible that psychophysical
performance is quite different for single electrodes compared with multi-electrode
activation, due to interactions between electrodes at the periphery or due to central
processing interactions or both. The present study measured intensity discrimination in
multi-electrode stimuli in 5 users of the Nucleus Cochlear Implant (CI) device to see if
10
single-channel psychophysical performance could be extrapolated to multi-electrode
performance.
2.2 METHODS
2.2.1 SUBJECTS
Subject Information S1-S4 for 4 post-lingually deaf Cochlear Implant (CI) users is
shown in Table 1.
Subject Age Gender
Implant
Type
Etiology Strategy
Vowel
Recognition
(%)
Consonant
Recognition
(%)
Duration
of
implant
use
(years)
S1 60 F
Nucleus-
24
Unknown ACE 60 63 6
S2 76 M
Nucleus-
22
Noise
Induced
SPEAK 86 67 10
S3 49 M
Nucleus-
22
Noise
Induced
SPEAK 92 81 14
S4 56 M
Nucleus-
22
Unknown SPEAK 73 72 16
Table 1: CI Subject Information for Experiment I
2.2.2 PROCEDURE AND STIMULI
Intensity Discrimination was demonstrated using intensity difference limens
(DLs) for a probe electrode, presented alone (unmasked) or in multi-electrode contexts.
Probe intensity DLs were measured at various levels that spanned the probe’s dynamic
range (DR).
11
For each probe reference level, masked thresholds were measured as a function
of the cochlear distance between masker and probe electrodes and as a function of
masker level. The stimulation mode, stimulation rate, and pulse train duration were
fixed for masker and probe electrodes. Stimuli were presented at 250
pulses/sec/electrode on single electrodes and on three electrode clusters, stimulated with
interleaved biphasic pulses.
Figure 3: Electrode configuration for Experiment I depicting presentation of probe
electrode (11,13) in between two pairs of surrounding masker electrodes in three
different configurations: Far (masker 6.75mm from probe), Medium (masker 4.5mm
from probe) and Close (masker 2.25 mm from probe)
6.75
mm
2.25
mm
4.5
mm
Probe location
Masker locations:
Close (8,10), (14,16)
Medium (5,7), (17,19)
Far (2,4), (20,22)
(11,13)
12
Intensity discrimination on a single target electrode (probe) was measured as a
function of level and as a function of the amplitude of the flanking electrodes (maskers)
in the multi-electrode complex.
Masker/Probe Parameters:
Burst Length: 300ms; Phase Duration: 200 µs; Inter-Phase Delay: 45 µs
Stimulation Rate: 250 Hz
Stimulation Mode: BP+1 (Bipolar plus 1)
Maskers presented at 20, 50 and 80 % DR. Intensity DLs measured for probe at 10, 30,
50, 70 and 90 % DR.
Task: 2-AFC adaptive procedure, adaptively changing pulse phase duration.
2.3 RESULTS
Plotted below are the Intensity DLs calculated for 4 CI users S1, S2, S3 and S4.
Intensity DLs for each subject are plotted as a function of probe % DR. For each
subject, the probe % DR varies from 10% to 90 % of its Dynamic Range. The three
panels for each subject show the three different configurations of masker-probe
separation i.e. masker (or flanker) close, medium and far apart from the probe. For each
configuration of the masker-probe configuration, 4 separate levels of masker are
chosen: that is 0%, 20%, 50% and 80% of its Dynamic Range.
Hence the different combinations of masker-probe combinations depict the
space profile of a multi-electrode configuration.
13
Figure 4: Intensity DLs (%) for Subjects S1-S4 as a function of probe percent DR
F la n k e r - p r o b e d is ta n c e
6 .7 5 m m
P r o b e P e r c e n t D R
1 0 3 0 5 0 7 0 9 0
0
2 0
4 0
6 0
8 0
DI / I (%)
0
2 0
4 0
6 0
8 0
F la n k e r L e v e l ( % D R )
0
2 0
4 0
6 0
8 0
1 0 0
0
2 0
5 0
8 0
F la n k e r - p r o b e d is ta n c e
4 .5 0 m m
F la n k e r - p r o b e d is ta n c e
2 .2 5 m m
S 1
F la n k e r - p r o b e d is ta n c e
6 .7 5 m m
P r o b e P e r c e n t D R
1 0 3 0 5 0 7 0 9 0
0
2 0
4 0
6 0
8 0
DI / I (%)
0
2 0
4 0
6 0
8 0
F la n k e r L e v e l ( % D R )
0
2 0
4 0
6 0
8 0
1 0 0
0
2 0
5 0
8 0
F la n k e r - p r o b e d is ta n c e
4 .5 0 m m
F la n k e r - p r o b e d is ta n c e
2 .2 5 m m
S 2
F la n k e r - p r o b e d is ta n c e
6 .7 5 m m
P r o b e P e r c e n t D R
1 0 3 0 5 0 7 0 9 0
0
2 0
4 0
6 0
8 0
DI / I (%)
0
2 0
4 0
6 0
8 0
F la n k e r L e v e l ( % D R )
0
2 0
4 0
6 0
8 0
1 0 0
0
2 0
5 0
8 0
F la n k e r - p r o b e d is ta n c e
4 .5 0 m m
F la n k e r - p r o b e d is ta n c e
2 .2 5 m m
S 3
F la n k e r - p r o b e d is ta n c e
6 .7 5 m m
P r o b e P e r c e n t D R
1 0 3 0 5 0 7 0 9 0
0
2 0
4 0
6 0
8 0
DI / I (%)
0
2 0
4 0
6 0
8 0
F la n k e r L e v e l ( % D R )
0
2 0
4 0
6 0
8 0
1 0 0
0
2 0
5 0
8 0
F la n k e r - p r o b e d is ta n c e
4 .5 0 m m
F la n k e r - p r o b e d is ta n c e
2 .2 5 m m
S 4
14
2.4 DISCUSSION
In general, the subjects show the following pattern in varying degrees:
1. Unmasked intensity DLs were relatively constant at 1-10% and increased
slightly at very soft presentation levels.
2. Masked threshold intensity DLs were elevated for standard probe levels ≤ 30 %
DR. As the masker intensity was increased, intensity DLs increased to as high as
70 %.
3. For standard probe levels ≥ 50 % DR, maskers had little effect, even when
located near the probe at 50 % DR.
4. In general, spatially remote maskers produced less interference than nearby
maskers. Louder maskers produced more interference than softer maskers.
3. EXPERIMENT II: EFFECT OF SPACE PROFILE ON
INTENSITY DISCRIMINATION USING A SHORT PROBE (20ms)
3.1 OBJECTIVE
Experiments for intensity discrimination were repeated using a short probe (20ms) and
longer masker duration (200 ms) to create a continuous stimulation field for the probe.
An optimum masker distance 4 electrodes away from the probe was chosen to zoom
into the region where maximum effect of multi-electrode stimulation is likely to be
15
observed. Stimuli were presented at 250 pulses/sec/electrode and 2000
pulses/sec/electrode using interleaved monophasic pulses.
3.2 METHODS
3.2.1 SUBJECTS
Cochlear Implant Subject Information S5 is shown in Table 2.
Subject Age Gender
Implant
Type
Etiology Strategy
Vowel
Recognition
(%)
Consonant
Recognition
(%)
Duration
of
implant
use
(years)
S5 63 F
Nucleus
24
Genetic ACE 83 85 3
Table 2: CI Subject Information for Experiment II
3.2.2 PROCEDURE AND STIMULI
Intensity DLs were measured for a short probe (20 ms) interleaved with a longer
masker (200 ms); the probe was temporally centered within the masker. Similar to Exp.
1, masked and unmasked intensity DLs were measured as functions of masker and
probe level; masker and probe levels were loudness balanced at different levels (i.e.,
very soft, soft and medium loudness). Masker electrodes were located 3 mm toward the
base and the apex, relative to the probe electrode. Masked and unmasked thresholds
were also measured for both low (250 pulses per second, or pps) and high (2000 pps)
stimulation rates applied to the masker and probe electrodes. The stimulation mode was
fixed for masker and probe electrodes.
16
Fig 5: Short probe (20ms) interleaved in time between two maskers (200ms each) such
that the probe pulse is centrally located between the two masker pulses
Masker/Probe Parameters:
Masker: 200 ms; 50 µs/phase; 50 µs inter-phase gap; Electrodes 6, 14, 6+14, MP1+2
Probe: 20 ms, 50 µs/phase; 50 µs inter-phase gap; Electrode 10, MP1+2
Relative masker/probe stimulation rates: 250/250 pps, 2000/250 pps, 2000/2000 pps
Masker/probe loudness-balanced reference levels: 10, 30, 50 % of reference electrode’s
DR
Task: (2-AFC) adaptive procedure, adaptively changing pulse phase duration.
Masker 1
(6)
Probe (10)
Masker 2
(14)
t=0ms
t=0.32 ms
t=90.16 ms
17
Figure 6: Probe centered between two maskers; Intensity DLs were measured at
loudness balanced reference levels at combinations of masker and probe levels for 10%,
30% and 50% of the electrode’s DR
3.3 RESULTS
Shown below are three views for the data from Subject S5. Intensity DL is shown as a
function of probe level, masker level and normalized intensity DLs in Figures 5 a, b and
c respectively.
Loudness-balanced
reference level
50 %
30 %
10 %
18
Fig 7 (a): Intensity DL (%) vs. probe level for Subject S5 for three different configurations of masker and probe
parameters
0
10
20
30
40
No masker
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Intensity DLs (%DR)
0
10
20
30
Probe level
(loudness-balanced reference level - % DR)
10 30 50
0
10
20
30
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 250 Hz
Masker: 200ms, 250 Hz
0
10
20
30
40
50
No masker
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Intensity DLs (%)
0
10
20
30
40
Probe level
(loudness-balanced reference level - % DR)
10 30 50
0
10
20
30
40
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 250 Hz
Masker: 200 ms, 2000 Hz
0
10
20
30
40
50
None
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Intensity DLs (%)
0
10
20
30
40
Probe level
(loudness-balanced reference level - % DR)
10 30 50
0
10
20
30
40
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 2000 Hz
Masker: 200ms, 2000 Hz
19
Fig 7 (b): Intensity DL (%) vs. masker level for Subject S5 for three different configurations of masker and probe
parameters
Masker
0
10
20
30
Intensity DLs(%)
0
10
20
30
0
10
20
30
40
10% DR
30% DR
50% DR
Probe reference level Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
None (6,30) (14,30) (6,30) +
(14,30)
Probe: (10,30), 20 ms, 250 Hz
Masker: 200 ms, 250 Hz
Masker
0
5
10
15
20
Intensity DLs (%)
0
5
10
15
20
0
5
10
15
20
25
10% DR
30% DR
50% DR
Probe reference level Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
None (6,30) (14,30) (6,30) +
(14,30)
Probe: (10,30), 20 ms, 250 Hz
Masker: 200 ms, 2000 Hz
Masker
0
10
20
30
40
Intensity DLs (%)
0
10
20
30
40
0
10
20
30
40
50
10% DR
30% DR
50% DR
Probe reference level Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
None (6,30) (14,30) (6,30) +
(14,30)
Probe: (10,30), 20 ms, 2000 Hz
Masker: 200 ms, 2000 Hz
20
Fig 7 (c): Normalized Intensity DL (%) vs. probe level for Subject S5 for three different configurations of
masker and probe parameters
-9
-6
-3
0
3
6
9
12
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Masked threshold shift (us/ph)
-9
-6
-3
0
3
6
9
12
Probe level
(loudness-balanced reference level - % DR)
10 30 50
-3
0
3
6
9
12
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 250 Hz
Masker: 200ms, 250 Hz
-9
-6
-3
0
3
6
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Masked threshold shift (us/ph)
-9
-6
-3
0
3
6
9
Probe level
(loudness-balanced reference level - % DR)
10 30 50
-10
0
10
20
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 250 Hz
Masker: 200 ms, 2000 Hz
-9
-6
-3
0
3
6
(6,30)
(14,30)
(6,30) + (14,30)
Simultaneous masker
Masked threshold shift (us/ph)
-9
-6
-3
0
3
6
Probe level
(loudness-balanced reference level - % DR)
10 30 50
0
10
20
Masker reference level:
10 % DR
Masker reference level:
30 % DR
Masker reference level:
50 % DR
Probe: (10,30), 20 ms, 2000 Hz
Masker: 200 ms, 2000 Hz
21
3.4 DISCUSSION
1. No significant difference was observed in unmasked intensity DLs for low and
high stimulation rates.
2. In general, there was no large difference between low- and high-rate maskers on
probe intensity DLs, especially for relatively low masker levels. Some
differences seen between masker stimulation rate for higher masker levels.
When masker and probe were equally loud, the 2000 pps masker produced more
interference than the 250 pps masker.
3. Low masker levels (i.e., 10 % DR) sometimes reduced intensity DLs to be lower
than unmasked intensity DLs.
4. Apically-situated maskers produced more interference than basally-situated
maskers. Combined apical and basal maskers sometimes produced more
masking than either masker alone, but often produced less masking than the
apical masker alone.
4. EXPERIMENT III: EFFECT OF INTER-CHANNEL TEMPORAL
OFFSET ON INTENSITY DISCRIMINATION
4.1 OBJECTIVE
Multi-electrode stimulation requires pulse trains to be interleaved in time.
Psychophysical results with implant users demonstrate appreciable interaction among
implant channels stimulated with interleaved pulse trains. A pulse on one electrode was
22
shown to influence the threshold on another electrode even when the onsets of the
pulses were separated by as much as 640 µs during a study involving the measurement
of thresholds for activation of the auditory cortex using single 80 µs/phase pulses. At
high stimulation rates (e.g.. 2000 pulses per second per channel or pps/channel), the
delay between interleaved pulses can be as small as 500 µs. High stimulation rates have
been shown to produce greater channel interactions at threshold stimulation in the
guinea pig auditory cortex at threshold levels thresholds measured in the auditory cortex
are largely determined by the integration of cochlear stimuli within the first ~1ms of
stimulation. In a previous pilot study, we found that the spatial distance and relative
stimulation levels between interleaved pairs of electrodes significantly influenced the
degree of channel interaction. When the spatial distance is small, or when the spread of
excitation is large (e.g., broad stimulation modes, high current levels), stimulation of
two interleaved electrodes may target the same neural population, giving rise to
temporally-based channel interactions. It is unclear how the temporal offset between
interleaved pulses may affect channel interaction at suprathreshold levels. This
motivates the effort to study in detail, the impact of inter-pulse timing, or temporal
offset, on channel interactions in cochlear implants.
4.2 METHODS
4.2.1 SUBJECTS
3 post-lingually deafened Cochlear Implant (CI) listeners participated in this
study:
23
Subject Age Gender
Implant
Type
Etiology Strategy
Vowel
Recognition
(%)
Consonant
Recognition
(%)
Duration
of
implant
use
(years)
S6 49 M
Nucleus-
22
Noise
Induced
SPEAK 92 81 14
S7 68 F
Nucleus-
24
Genetic ACE 60 63 4
S8 76 M
Nucleus-
22
Noise
Induced
SPEAK 86 67 10
Table 3: CI Subject Information for Experiment III
4.2.2 PROCEDURE AND STIMULI
In the present study, intensity difference limens (DLs) were measured for two
simultaneously interleaved channels each stimulated at 250 pps/channel at 200 µs/phase
and 50 µs inter-phase delay, using BP+1 stimulation mode. The masker-probe pair was
spatially separated by 2.35-3 mm. The temporal offset between the masker and the
probe was varied from 500µs to 3.5ms. Intensity DLs were measured for a short probe
(20 ms) interleaved with a longer masker (200 ms); the probe was temporally centered
within the masker. Masked and unmasked intensity DLs were measured as functions of
masker and probe level; masker and probe levels were loudness balanced at different
levels (i.e., very soft, soft and medium loudness). For each probe reference level,
masked thresholds were measured for masker levels less than or equal to the probe
levels as these were the conditions where a greater effect of temporal offset was
observed.
24
Fig 8 (a): Interleaving of masker and probe pulses: Probe onset at 500 µs
Fig 8 (b): Interleaving of masker and probe pulses: Probe onset at 3500 µs
Probe onset
t=500 µs
MASKER 250 Hz
PROBE 250 Hz
Masker onset
t=0 µs
Next masker pulse
t=4000 µs
Next probe pulse
t=4500 µs
TIME (µs)
Probe onset
t=3500 µs
MASKER 250 Hz
PROBE 250 Hz
Masker onset
t=0 µs
Next masker pulse
t=4000 µs
TIME (µs)
25
Masker/Probe Parameters:
Masker: 220 ms; 50 µs/phase; 50 µs inter-phase gap; BP+1
Probe: 20 ms, 50 µs/phase; 50 µs inter-phase gap; Electrode 10, BP+1
Stimulation rates: 250 pps
Masker/probe loudness-balanced reference levels: 10, 30, 50 % of reference
electrode’s Dynamic Range (DR)
Task: (2-AFC) adaptive procedure, adaptively changing pulse phase duration. The task
was carried out for stimulation levels where the masker % DR ≤ probe % DR.
4.2.3 RESULTS
Shown below are Intensity DLs for 3 subjects for inter-channel delays between
masker and probe ranging from 500 µs to 3500 µs. For each probe level, masker levels
less than or equal to the probe level were chosen.
Results show that DLs were elevated for short temporal offsets (or short inter-
channel delay) and were reduced as the temporal offset was increased to half the
stimulation period. At smaller temporal offset, the masker and probe are closer in time
and this produces greater channel interaction than at larger temporal offset. At temporal
offset greater than half the stimulation period, the masker pulse starts moving closer to
probe pulse from the next time period and so the Intensity DLs start increasing again.
26
Intensity DLs (%)
6
7
8
9
10
11
12
Unmasked
10% DR
30% DR
Probe: (10,12), 20ms, 250 Hz
Masker: (6,8), 220ms, 250 Hz
Masker reference level
Probe reference level
30% DR
Inter-channel delay (μ μ μ μs)
0 500 1000 1500 2000 2500 3000 3500 4000
6
7
8
9
10
11
Unmasked
10% DR
30% DR
50% DR
Masker reference level
Probe reference level
50% DR
Masker reference level
S6
Figure 9 (a) Intensity DL versus inter-channel delay for Subject S1 at probe levels 30 %
and 50 % of the DR.
27
Inter-channel delay (μ μ μ μs)
0 500 1000 1500 2000 2500 3000 3500 4000
Intensity DLs (%)
10
15
20
25
30
35
Unmasked
10% DR
30% DR
50% DR
Probe: (18,20), 20ms, 250 Hz
Masker: (16,18), 220ms, 250 Hz
Masker reference level
Probe reference level
50% DR
15
20
25
30
35
Unmasked
10% DR
30% DR
Masker reference level Probe reference level
30% DR
S7
Figure 9 (b) Intensity DL versus inter-channel delay for Subject S2 at probe levels 30 %
and 50 % of the DR.
28
Inter-channel delay (μ μ μ μs)
0 500 1000 1500 2000 2500 3000 3500 4000
Intensity DLs (%)
8
10
12
14
16
18
Unmasked
10% DR
30% DR
50% DR
Probe: (10, 12), 20ms, 250 Hz
Masker: (6, 8), 220ms, 250 Hz
Masker reference level
Probe reference level
50% DR
Masker reference level Probe reference level
30% DR
S8
8
10
12
14
16
18
20
Unmasked
10% DR
30% DR
Masker reference level Probe reference level
30% DR
Figure 9 (c) Intensity DL versus inter-channel delay for Subject S3 at probe levels 30 %
and 50 % of the DR.
29
4.2.4 DISCUSSION
The two intra-cochlear electrodes appeared to produce a considerable overlap of
electrical fields and that the temporal effects of interest are on a sub-millisecond scale.
We speculate that the effect of elevated Intensity DLs is due to the induction of a partial
refractory state in the overlapping neuronal population by the leading masker pulse
which makes it harder to discriminate the following probe pulse.
Intensity DLs were highest when masker and probe levels were equivalent and
the inter-channel delay between pulses was minimum. This produced greatest channel
interaction due to proximity of the masker and probe levels in both amplitude and time,
making it difficult for either pulse to be distinct.
5. CONCLUSION
Intensity DLs indicate the ability to detect a change in intensity on one of the
many stimulated electrodes. These DLs may be comprised by channel interactions.
Channel interactions may be spatial or temporal, and may result from three phenomena:
a. summation of electric fields from multiple electrodes
b. stimulation of common nerve bundles by multiple electrodes
c. perceptual summation or interference effects
5.1 EFFECT OF SPACE PROFILE
Multi-electrode stimulation can reduce the ability to detect amplitude changes
on individual electrodes, which might reduce the saliency of a formant peak or a change
in formant peak levels, for example. Thus, the ability to discriminate intensity
30
differences within a complex stimulus pattern may be considerably poorer than the
levels implied from single electrode data.
Intensity DLs were significantly affected by location and (sometimes)
stimulation rate of the masker. When two masker electrodes were interleaved, the
amount of masking was often less than that produced by a single apically-located
masker, suggesting that multiple channels may interact in ways that are difficult to
predict.
Intensity DLs were significantly affected by both masker and probe level.
Depending on relative levels of masker and probe, multi-channel stimulation sometimes
produced lower intensity DLs than single-channel stimulation.
Differences in intensity discrimination for single and multi-electrode complexes
presumably reflect a combination of peripheral interactions between electrodes and
central integration of information across electrodes. It is not clear in the present data
whether the source of the interference is peripheral or central.
5.2 EFFECT OF TEMPORAL PROFILE (INTER-CHANNEL TEMPORAL
OFFSET)
Temporally-based interactions among interleaved electrodes may limit the
number of spectral channels transmitted to the brain. This may be particularly
prominent at high stimulation rates which have more number of pulses per second and
so the pulses are closer together in time, and this is likely to cause more channel
interactions. This is a condition in which the impact of temporal offset may be critical;
31
the temporal offset between adjacent channels should be optimized to minimize these
channel interactions.
High stimulation rates may need to utilize very short phase durations to achieve
the optimal time constant for interleaving multiple channels. The trade-off is that higher
current (and potentially greater spread of excitation) will be required for short phase
durations to achieve comfortable listening levels, possibly resulting in increased (place-
based) channel interaction.
Threshold measured in the auditory cortex is determined largely by the
integration of cochlear stimuli within the first ~1ms of stimulation. Inter-pulse intervals
especially at high pulse rates fall in this time scale and are significant.
32
BIBLIOGRAPHY
1. Chatterjee, M. (1998) “Temporal mechanisms underlying recovery from forward
masking in multielectrode-implant listeners”, J. Acoust. Soc. Am. 105(3), March
1999, 1853-1863
2. Drennan, W. (2006) “Current-Level Discrimination in the Context of
Interleaved, Multichannel Stimulation in Cochlear Implants: Effects of Number
of Stimulated Electrodes, Pulse Rate, and Electrode Separation”, JARO 7, 308-
316
3. Ehrenstein, W. and Ehrenstein, A. “Psychophysical Methods”, Chapter 43.
4. Lim, H. (1989) “Forward masking paaterns produced by intracochlear electrical
stimulation of one and two electrode pairs in the human cochlea”, J. Acoust.
Soc. Am. 86(3), September 1989, 971-980
5. Loizou, P. (1998). “Mimicking the human ear”, IEEE Signal Processing
Magazine,15(5), 101-130
6. Middlebrooks, J. (2004). “Effects of cochlear-implant pulse rate and inter-
channel timing on channel interactions and thresholds”, J. Acoust. Soc. Am.,
116(1), 452-468
7. Rubinstein, J. (2004) “How cochlear implants encode speech”, Hearing Science,
444-448
8. Zeng, F. (2005) “Temporal Masking in Electric Hearing”, JARO (2005)
Abstract (if available)
Abstract
Single-channel psychophysics may not reflect performance with dynamic, multi-channel stimuli (e.g., speech stimuli) in cochlear implant (CI) listeners due to interactions between electrodes at the periphery or due to central processing interactions or both. The first part of the study measured single and multi-channel intensity discrimination in 5 Nucleus CI users, as functions of the relative level, electrode location and stimulation rate of the masker and probe electrodes. Spatial distance and relative stimulation levels between interleaved pairs of electrodes were found to significantly influence the degree of channel interaction. The second part of the study investigated how temporal offset between interleaved pulses may affect channel interaction at suprathreshold levels in 3 Nucleus CI users. Results showed that interactions were elevated for short temporal offsets and were reduced as the temporal offset was increased to half the stimulation period.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Multi-region recordings from the hippocampus with conformal multi-electrode arrays
PDF
Speech enhancement and intelligibility modeling in cochlear implants
PDF
Improving frequency resolution in cochlear implants with implications for cochlear implant front- and back-end processing
PDF
Perceptual adaptation to spectrally shifted vowels: experiment studies and an acoustic analysis framework
PDF
Understanding music perception with cochlear implants with a little help from my friends, speech and hearing aids
PDF
Synaptic second wind: phosphene fading of the Argus II and glutamate secretion in the inner retina
PDF
Nonlinear dynamical modeling of single neurons and its application to analysis of long-term potentiation (LTP)
PDF
A million-plus neuron model of the hippocampal dentate gyrus: role of topography, inhibitory interneurons, and excitatory associational circuitry in determining spatio-temporal dynamics of granul...
PDF
Functional connectivity analysis and network identification in the human brain
PDF
Towards health-conscious spaces: building for human well-being and performance
Asset Metadata
Creator
Gaur, Ashmita
(author)
Core Title
Intensity discrimination in single and multi-electrode patterns in cochlear implants
School
Viterbi School of Engineering
Degree
Master of Science
Degree Program
Biomedical Engineering
Publication Date
04/05/2009
Defense Date
03/15/2007
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cochlear implants,OAI-PMH Harvest,psychophysics
Language
English
Advisor
Shannon, Robert (
committee chair
), Greene, Ernest (
committee member
), Meng, Ellis F. (
committee member
)
Creator Email
ashmitag@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m350
Unique identifier
UC1279688
Identifier
etd-Gaur-20070405 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-321894 (legacy record id),usctheses-m350 (legacy record id)
Legacy Identifier
etd-Gaur-20070405.pdf
Dmrecord
321894
Document Type
Thesis
Rights
Gaur, Ashmita
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
cochlear implants
psychophysics