Institut für Schallforschung, Österreichische Akademie der Wissenschaften; A-1010 Wien, Reichsratsstrasse 17

Tel: +43 1/4277-29514 Fax: +43 1/4277-9295 email: Bernhard.Laback@oeaw.ac.at

Envelope vs. Fine Structure in Bilateral Hearing

Bernhard Laback1, Piotr Majdak1, and Wolf-Dieter Baumgartner2

1 Acoustics Research Institute, Austrian Academy of Sciences2 ENT-department, Vienna University Hospital

6th Meeting on Bilateral CI and Binaural Sigal Processing, BernMarch 30, 2007

http://www.kfs.oeaw.ac.atBernhard.Laback@oeaw.ac.at

Sound Source

ITD Interaural Time Differences (ITDs) occur in

• Gating portions (ITDON/ITDOFF)

• Temporal fine structure (ITDFS)

• Ongoing envelope (ITDENV)

ITDONITDOFF

Interaural Time Difference (ITD)

Motivation for study

Fine structure ITD is important for • Lateralizing sound sources (Wightman and Kistler, 1992;

Smith et al., 2002) and for • Speech understanding in noise (Nie et. al., 2005; Zeng et

al., 2005)

CI listeners are often sensitive to ITD (e.g. van Hoesel and Tyler, 2003)

Open Questions: • Are bilateral CI listeners sensitive to ITD in the fine

structure? • What is the contribution of gating ITD and ongoing

envelope ITD?

Lateralization Discrimination of ITD in Fine Structure, Onset, and Offset: Four-Pulse Sequences

Laback, Majdak, and Baumgartner (2007) JASA 121 (April)

Study I

Methods I

Electric stimuli− Biphasic pulse trains (4 pulses) presented

at interaurally pitch-matched and loudness-balanced electrode pair (selected in pretests)

via Research Interface (RIB) with interaural accurracy of 2.5 µs

ITD conditions (see right side)

No Delay(comparison condition)

Waveform(Wave)

Ongoing

Onset

Gating

TimeAmpl

Offset

ITD

Left

Right

Left

Right

Left

Right

Left

Right

Left

Right

Left

Right

(≈ Fine Structure)

JNDs for Left/Right Discrimination

Subjects

− Four CI listeners (C40+, MED-EL), postlingually deafened, selected based upon minimum ITD sensitivity criterion

− Five NH listeners (listening to CI simulation)

Independent Variables− ITD condition

− Pulse Rate

Methods II

Expectations

Pulse Rate (pulses per second, pps)

Just

Not

icea

ble

Diff

eren

ce (µ

s)

Waveform ITD

Expectations

Pulse Rate (pulses per second, pps)

Just

Not

icea

ble

Diff

eren

ce (µ

s)

Waveform ITD

Onset ITD

Expectations

Ongoing ITD

Pulse Rate (pulses per second, pps)

Just

Not

icea

ble

Diff

eren

ce (µ

s)

Waveform ITD

Onset ITD

Results for CI listeners: CI1

100 200 300 400 500 600 700 800Pulse Rate (pulses per second, pps)

0

200

400

600

800

1000

JND

(µs)

CI1NDND Wave

OngoingGatingOnsetOffset

• Wave ITD sensitivity decreases with rate (p < 0.003)

• Ongoing ITD contributes at 100 pps only

• Onset ITD contribution increases with rate (p < 0.0001)

Results for CI listeners: CI12

100 200 300 400 500 600 700 800Pulse Rate (pulses per second, pps)

0

200

400

600

800

1000

JND

(µs)

CI12NDND Wave

OngoingGatingOnsetOffset

• Wave ITD sensitivity independent of rate

• Ongoing ITD contributes up to 400 pps

• Onset ITD contribution increases with rate (p < 0.034)

Results for CI listeners: CI8

100 200 300 400 500 600 700 800Pulse Rate (pulses per second, pps)

0

200

400

600

800

1000

JND

(µs)

CI8NDND Wave

OngoingGatingOnsetOffset

• Wave ITD sensitivity up to 400 pps only

• Ongoing ITD contributes up to 400 pps

• Onset ITD does not contribute

Results for CI listeners: CI3

WaveOngoingGatingOnsetOffset

100 200 300 400 500 600 700 800Pulse Rate (pulses per second, pps)

0

200

400

600

800

1000

JND

(µs)

CI3NDND

• Wave ITD sensitivity independent of rate

• Ongoing ITD contributes at all rates tested (up to 800 pps)

• Onset ITD contribution increases with rate (p < 0.04)

CI listeners are sensitive to ongoing ITD in four-pulse sequences, thus to “pure” fine structure ITD

Highest rate showing fine structure ITD sensitivity varies from 100 to 800 pps

Contribution of onset ITD increases with pulse rate (in three out of four subjects)

Conclusions of Study I

Lateralization Discrimination of ITD in Fine Structure and Ongoing Envelope: Modulated pulse trains

Majdak, Laback, and Baumgartner (2006) JASA 120, 2190-2201

Study I

Methods

Stimuli− Amplitude modulated pulse trains (300 ms)

(Fmod = 13 Hz)

Independent variables

− ITDFS: 0…IPI (inter-pulse interval)

− ITDENV: 0…800µs

− Pulse rate: 100 … 1600 pps

Subjects − 4 CI listeners (postlingually deafened)− 4 NH listeners (listening to CI simulation)

Expectations

½ IPI IPI0%

Late

raliz

atio

n di

scrim

inat

ion

bette

rbe

tter

ITDENV= 0 µs

¼ IPI

100%

-100%

¾ IPIITD FS

„correct“ side

„wrong“ side

½ IPI IPI0%

Late

raliz

atio

n di

scrim

inat

ion

bette

rbe

tter

ITDENV= 0 µs

ITDENV > 0

¼ IPI

100%

-100%

¾ IPIITD FS

„correct“ side

„wrong“ side

Expectations

Results for Lower Pulse Rates

Results for Lower Pulse Rates

0 2 0 0 0 4 0 0 0 6 0 0 0 8 0 0 0 1 0 0 0 0- 1 0 0

- 8 0

- 6 0

- 4 0

- 2 0

0

2 0

4 0

6 0

8 0

1 0 0

I T D F S i n µ s

Late

ralis

atio

n di

scrim

inat

ion

in %

C I 2 - 1 0 0 p p s

E N V = 0 µ s

E N V = 6 2 5 µ s

E N V = 8 0 0 µ s

m e a nm e a n

m e a n

P R = 4 0 0 p p s

Results for Higher Pulse Rates

Sensitivity to Fine Structure ITD

Pulse rate CI1 CI2 CI3 CI8100 < 0.001 - -150 < 0.001 - -200 < 0.001 0.01 - -400 0.75 0.21 < 0.001 < 0.001600 - - - -800 - - <0.001 <0.001938 - - - 0.451600 0.46 - 0.11 -

NHs - - -

< 0.001< 0.001< 0.139< 0.712

-

High sensitivity to ITDFS (in 2/4 subjects up to 800 pps)

Low sensitivity to ITDENV (low modulation rate)

High inter-subject variability of performance

Conclusions of Study II

CI listeners are likely to benefit from encoding fine structure ITD at lower rates in CI systems

The highest rate providing fine structure ITD cues is lower than in acoustic hearing with sinusoids (up to 1500 Hz)

Overall Conclusions