Perception of amplitude modulation with single or multiple channels in cochlear implant users
PhD ceremony: | Mr J.J. (John) Galvin |
When: | October 24, 2016 |
Start: | 16:15 |
Supervisors: | prof. dr. ir. D. (Deniz) Baskent, prof. dr. F. Qian-jie |
Where: | Academy building RUG / Student Information & Administration |
Faculty: | Medical Sciences / UMCG |

Cochlear implants (CIs) are able to restore hearing to many deaf individuals. Contemporary CIs work by converting acoustic signals into pulse trains of electric current delivered to the implanted electrodes. These pulse trains are amplitude-modulated by the changes in energy over time associated with the input signal. CI users’ perception of amplitude modulation (AM) relates strongly to speech perception. However, perception of AM might be quite different when delivered to a single electrode (as in most previous studies) or to multiple electrodes (as in this thesis).
We studied tow forms of AM perception: 1) AM detection (sensitivity to changes in amplitude over time relative to steady pulse trains), and 2) AM discrimination (sensitivity to changes In AM frequency for modulated pulse trains). In clinical fitting of CIs, current levels on single electrodes must often be reduced to accommodate multi-electrode stimulation (“multi-channel loudness summation”).
We found that these current level reductions may worsen multi-electrode AM detection, relative to AM detection on an equally loud single electrode. However, these current level reductions did not affect multi-electrode AM discrimination, where performance was similar between equally loud single- and multi-electrode stimuli. When different AM information was delivered to multiple electrodes, AM discrimination depended on the spacing of electrodes. When electrodes were closely spaced, CI listeners were very sensitive to differences in AM information delivered to the multiple channels.
The results suggest that loudness summation and spacing of electrodes may greatly affect AM perception with multiple electrodes.