Contributions of Peripheral Spatial Resolution to Speech Perception

Contributions of Peripheral Spatial Resolution to Speech Perception in Cochlear Implant Users Statistical Consulting February 11, 2013 8: 30 – 9: 20 am 30 Schaeffer Hall

Background • Personal – Clinical Audiologist – Ph. D Student in Hearing Science working on Dissertation • Research Project – Cochlear Implant Users

Two Prevailing Questions in Cochlear Implant Research • Why is there so much variability in performance among individuals who use cochlear implants? • When children are implanted, how can we form realistic expectations about their potential performance? How can the clinician ensure that the device is programmed optimally?

Hearing Loss Cochlear Implant Auditory Nerve Cortex Perception ECAP ACC Speech Simple Complex

Goal 1. Auditory Nerve and Cortex • ECAP Channel Interaction Functions – Thirteen channel interaction functions total – Quantified as an area of separation (channel separation index) between two functions

Goal 1. Auditory Nerve and Cortex • Simple ACC – Thirteen electrode pairs to assess discrimination

Can ECAP channel separation indices predict simple ACC amplitude? Sub-question: Are there 2 unique curves or just 1? F 19 R E 60 Assume a monotonically increasing, saturating function: y=a(1 -e-x/b)

Goal 2. Simple versus Complex Stimulation • Complex ACC – Three experimental speech processor programs, each uses a different subset of 7 electrodes from the 13 used for ECAP and simple ACC measures 1 2 3 4 5 6 7 8 P 1 P 2 P 3 X X X 9 10 11 12 13 14 15 X X X X 16 X X 17 18 19 20 21 X X X 22

Goal 2. Simple versus Complex Stimulation • Three sets cortical responses to complex stimuli.

Can ECAP channel separation indices and/or cortical responses to electrode changes predict a cortical response to a complex change?

Goal 3. Speech Perception • Three vowel perception scores (% correct), one for each experimental program. • Three “signal-to-noise ratios” for 50% correct, one for each experimental program.

Can ECAP channel separation indices, cortical responses to electrode changes and/or cortical responses to complex changes predict speech perception? *Repeat analysis using a speech-in-noise test as the speech perception measure. *

Progress to Date and Plans • Data Collection: – 2 pilot subjects have completed testing – Plan: 10 subjects total • Decisions I still have to make: – ECAP: use raw or normalized amplitudes – ECAP: use an average or maximum separation index – Simple ACC: use raw or normalized peak-to-peak amplitudes or raw or normalized rms amplitudes – Complex ACC: ripple depth threshold or rms amplitude at a set ripple depth

Project Summary Goal 1: Auditory Nerve versus Cortex for Simple Stimulation – 13 ECAP Channel Separation Indices (6 Basal, 6 Apical, 1 Zero Point) – 13 Spatial ACC (6 Basal, 6 Apical, 1 Zero Point) Goal 2: Simple versus Complex Stimulation – 3 ECAP Channel Separation Indices (average of the adjacent electrodes in the experimental program) – 3 Calculated Simple ACC Amplitudes—from the coefficients determined in Goal 1 – 3 Ripple Depth Thresholds from Complex ACC Amplitudes Goal 3: Speech Perception – 3 ECAP Channel Separation Indices – 3 Calculated Simple ACC Amplitudes—from the coefficients determined in Goal 1 – 3 Ripple Depth Thresholds from Complex ACC Amplitudes – 3 Vowel Perception Scores – 3 Speech-in-noise Scores