Cochlear implants are highly successful neural prostheses that enhance or restore hearing to the severely hearing impaired. However, performance varies considerably among cochlear implant listeners, particularly in noisy environments and for spectrally complex stimuli like music. Pathology and imaging studies suggest that two sources of performance variability are the distribution of surviving spiral ganglion neurons and their distance from individual CI electrodes. The objective of the proposed study is to develop clinically useful procedures to determine non-stimulable regions of the cochlea. The resulting functional maps could guide the production of patient-specific sound processing strategies, leading to improvements in the auditory ability of cochlear implant listeners. Three types of psychophysical and electrophysiological experiments will be conducted to identify the location of low functioning cochlear implant channels. All experiments will use a novel electrode configuration, partial tripolar, to activate restricted populations of neurons, analogous to the use of narrowband stimuli for assessing low functioning acoustic dead regions in hearing impaired listeners. The partial tripolar configuration is a hybrid between the broad monopolar and narrow tripolar configurations, in which the fraction of current that flows to two flanking electrodes can be adjusted to change the spread of current in the cochlea. In the first experiment, detection thresholds will be measured across the implant array as the partial tripolar fractional current varies. Low functioning channels will be inferred from channels with relatively high thresholds when the current field is narrow and the change in threshold with fractional current will provide information about the spatial extent of the functional deficit. In the second experiment, psychophysical tuning curves will be obtained using a forward masking procedure to provide a more detailed assessment of cochlear activation. Similar to the application of tuning curves to diagnose acoustic dead regions, the widths and tip shifts of the tuning curves will be measured as a way to identify low functioning cochlear implant channels. In the third experiment, evoked potentials will be recorded. Evoked potential thresholds and amplitude growth functions will be directly compared to psychophysical thresholds and tuning curves obtained with different partial tripolar current fractions. A significant correlation will suggest that the evoked potential procedure, which is a clinically practical technique, can detect low functioning channels. Ultimately, the findings of this study may lead to the development of a patient-specific mapping procedure that can be used clinically. Over 60,000 people with severe hearing loss have been fitted with cochlear implants to restore some auditory capabilities. Identifying low functioning cochlear implant channels in the proposed study could lead to the enhancement of speech and music perception in these patients.
