The impact of asymmetric rates on interaural time difference lateralization and auditory object formation in bilateral cochlear implant and normal hearing listeners

Tanvi Thakkar, Alan Kan, and Ruth Y. Litovsky1University of Wisconsin-Madison, Madison, WI

e-mail: tanvi.thakkar@wisc.edu

WAISMAN CENTER

INTRODUCTION

METHODS

Binaural Hearing and Speech Laboratory

Acoustics ‘17 BostonBoston, MA

2017

Hypothesis: If stimulation rate is a major component of lateralization and auditory object formation…

1. Asymmetric across-ear rates will weaken object formation and yield

poor lateralization.

2. Symmetric across-ear rates will strengthen object formation and yield

good lateralization.

RESULTS I: How many sounds were reported as a single auditory object?Components that may influence a listener's ability to successfully

attend to a signal in noise include:

As expected, BiCI and NH listeners have a high degree of AOF when information across the ears is the same. However, when information is different across the ears, range of AOF differs between BiCI and NH listeners suggesting that detection of rate differences with simultaneous presentation across the ear is more difficult with electrical stimulation (Fig. 3) .

As expected, when AOF is poor, ITD lateralization is also poor. However, a high degree of AOF did not guarantee good ITD lateralization in both NH and BiCI listeners (Fig. 3, 4 & 6).

Overall, our results suggest that having AOF even when information is different across the ears is insufficient for good lateralization for both populations of listeners.

1. Steiger H. and Bregman, A. (1982b). Competition among auditory streaming, dichotic fusion, and diotic fusion. Perception & Psychophysics. 1982, 32(2), 153-162.

2. Kan A., and Litovsky RY (2015). Binaural hearing with electrical stimulation. Hear Res. 2015 Apr;322:127-37.3. M. A. Stellmack and H. Dye, “The combination of interaural information across frequencies : The effects of

number and spacing of components , onset and harmonicity,” vol. 93, no. May 1993, pp. 2933–2947, 1993.4. Carlyon RP., Long CJ., and Deeks JM (2008). Pulse-rate discrimination by cochlear-implant and normal-hearing

listeners with and without binaural cues. J Acoust Soc Am. 2008 Apr;123(4):2276-86.

Figure 3: Proportion of ‘One’ sound responses (“One-Left”, “One-Center”, or“One-Right”) averaged across listeners. Error bars represent standard error.

BiCI listeners (N=8): Presented with biphasic electrical pulse trains using synchronized research processors (Cochlear RF Generator) at a single pitch-matched pair of electrodes (Fig. 1).

NH listeners (N=12): Presented with a Gaussian enveloped tone (GET) acoustic pulse train centered at 4 KHz.

Experiment: single-interval task with 6 response options (Fig 2).

“One-Left”“One-

Center”“One-Right”

“Left-

dominant”“Center”“Right-

dominant”

Figure 2: Response options

Bilateral cochlear implant (BiCI) listeners

RESULTS II: How many responses were correctly lateralized (NH Listeners)?

NH listeners show AOF for a narrow range of asymmetric rates across the ears.

NH listeners never report hearing one sounds when rates differ more than -15% and +25%.

BiCI listeners show AOF for a broader range of asymmetric rates across the ears.

NH listeners appear to show some, yet weaker lateralization when rates are symmetric even in the presence of AOF.

However, NH listeners poorly lateralize when rates are completely asymmetric and AOF is lacking.

Normal hearing (NH) listeners

DISCUSSION

Response patterns across listeners appear to differ: 1. IBK, IBF, and IBK no longer lateralize the ITD of the perceived single fused image

with the introduction of dichotic pulse rates2. ICD, ICJ, ICP, IBQ, and IAU, lateralize by making more “left” and “right” decisions

in the dichotic pulse-rate conditions. BiCI listeners with poorer ITD sensitivity appear to show poor lateralization and

greater AOF regardless of rate symmetry.

Figure 4: Proportion of correctly lateralized responses per target ITD presented, averaged acrossthe group of NH listeners. Predicted results shown in inset. Error bars represent standard error.

??? ???Accurate lateralization Inaccurate lateralization

Binaural Sensitivity

Having good sensitivity to binaural cues can help with locating the direction of a coherent object

Have good binaural sensitivity that leads to good lateralization of interaural time differences (ITDs) 1.

When provided with symmetric or identical across-ear information, listeners perceive a coherent auditory object3.

Lack of auditory object formation (AOF) may lead to poor lateralization of ITDs.

Binaural sensitivity is limited in clinical processors from poor representation of the ITD in the signal. However, listeners demonstrate ITD sensitivity when listening with research processors2.

Sound processors operate independently and may encode the same signal differently between the two ears4.

It is unknown whether poor ITD sensitivity with sound processors is due to poor AOF.

Figure 1: Stimulus examples of BiCI (a,b) and NH (c) pulse-rate conditions.

c)

RESULTS III: How many responses were correctly lateralized (BiCI Listeners)?

Figure 6: Proportion of lateralized responses reported as “one” sound per targetITD for each individual BiCI listener; ITDs tested shown in top right of each panel;each BiCI listener’s sensitivity to ITDs of symmetric rates is shown in the top left.

Res

po

nse

s si

mil

ar

to p

red

icti

on

Perceived Location

“One-Left”

“One-Center”

“One-Right”

Rate in right ear (pps)

Pro

po

rtio

n r

epo

rted

as

“on

e”

sou

nd

in t

he

corr

ect

dir

ecti

on

0

0.5

1|ITD|=±500IBF |ITD|=±500IBK |ITD|=±1000IBZ

0

0.5

1|ITD|=±500ICD |ITD|=±500ICJ |ITD|=±500ICP

0

0.5

1|ITD|=±500IBQ |ITD|=±800IAU

~52 µs ~62 µs

~58 µs ~203 µs

~264 µs ~638 µs

~203 µs

~671 µs

Res

po

nse

s d

evia

tin

g fr

om

pre

dic

tio

n

ITDs (in µs): randomized trial-by-trial. BiCI: 0 (all), ±500 (5 listeners), ±800 (2

listeners), ±1000 (1 listener). NH: 0, ±700.

Rate conditions (pulses per second, pps): Left: always 100 pps. Right: 75, 85, 99, 100, 101, 125, 150, 175,

200, 250, 300 pps, interleaved. 20 repetitions for each ITD and rate

combination.

BiCI listeners NH listeners

Auditory Object Formation

0 0.5 1 1.5 2 2.5 3 3.5

x 104

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

Weak object formation Strong object formation

“One” sound

0 0.5 1 1.5 2 2.5 3 3.5

x 104

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

“Two” sounds

Auditory object formation occurs when pieces of information within and across ears are bound together to perceive one coherent object

ACKNOWLEDGEMENTSWe would like to thank all our participants and Cochlear Ltd for providingequipment. A special thank you to for her comments on the poster. Thiswork is funded by NIH-NIDCD (R03DC015321 to AK and R01DC003083 toRYL), and NIH-NICHD (U54HD09256 to Waisman Center).

Perceived Location

“One-Left”

“One-Center”

“One-Right”

Pro

po

rtio

n r

epo

rted

as

“on

e” s

ou

nd

in

th

e co

rrec

t d

irec

tio

n

Rate in right ear (pps)

Pro

po

rtio

n r

epo

rted

as

“o

ne”

so

un

d in

th

e co

rrec

t d

irec

tio

n

BiCI listeners show inconsistent lateralization and AOF when rates are interaurally asymmetric.

Figure 5: Proportion of correctly lateralized responses per target ITD presented,averaged across the group of BiCI listeners. Error bars represent standard error.

Left:100 pps & Right:100 pps Left:100 pps & Right: 150 pps (example)

Asymmetric electric pulse train (BiCI)

ITD (µs)

Symmetric electric pulse train (BiCI) a) b)

Asymmetric GET pulse train (NH)

Left:100 pps & Right: 150 pps (example)

Pro

po

rtio

n r

epo

rted

as

“on

e”

sou

nd

fro

m t

ota

l tri

als

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5 3 3.5

x 104

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

Rate in right ear (pps)

Perceived Location

0

1

0.5

0

0.2

0.4

0.6

0.8

1

Pro

po

rtio

n r

epo

rted

as

“tw

o”

sou

nd

s in

th

e co

rrec

t d

irec

tio

n

0

0.2

0.4

0.6

0.8

1

0

0.2

0.4

0.6

0.8

1

“Two-Left”

“Two-Center”

“Two-Right”

Pro

po

rtio

n r

epo

rted

as

“tw

o” s

ou

nd

s in

th

e co

rrec

t d

irec

tio

n

“One-Left” “One-Center” “One-Right”

Prediction

“Two-Left” “Two-Center” “Two-Right”

REFERENCES

Poster #6

0

0.2

0.4

0.6

0.8

1

Rate in right ear (pps)