Onset weighting of interaural time differences in neonatally deafened CI rats is comparable to normal hearing rats

Cochlear implants (CIs) are undeniably useful, but the auditory perception gained from current devices still fails for complex, real life, environments. One particular shortcoming is the inability to provide CI users with adequate temporal spatial cues, or interaural time differences (ITDs). We have previously shown that neonatally deafened rats are, perhaps surprisingly, capable of discriminating ITDs as small as 50 μs, apparently no worse than their normally hearing litter mates. Here we assessed whether both neonatally deaf (ND) CI and normally hearing (NH) rats show a similar strong emphasis on the onset in their ITD based location judgment of rapid sequences of pulses, as has been described in normal hearing humans.

Cochlear implants (CIs) are undeniably useful, but the auditory perception gained from current devices still fails for complex, real life, environments. One particular shortcoming is the inability to provide CI users with adequate temporal spatial cues, or interaural time differences (ITDs). We have previously shown that neonatally deafened rats are, perhaps surprisingly, capable of discriminating ITDs as small as 50 μs, apparently no worse than their normally hearing litter mates (Rosskothen-Kuhl et al,2019: BIORXIV/2018/498105). Here, we assessed whether both neonatally deaf (ND) CI and normally hearing (NH) rats show a similar strong emphasis on the onset in their ITD based location judgment of rapid sequences of pulses, as has been described in normal hearing humans (Brown and Stecker (2010); https://doi.org/10.1121/1.3436540).

Neonatal rats were deafened using an established kanamycin protocol and, as young adults, bilaterally implanted with CIs. Normally hearing rats served as controls. A 2-alternative forced choice task was used to determine the temporal weighting functions (TWFs) of ITD cues. The animals lateralised pulse trains consisting of eight binaural pulses (either biphasic electric stimuli delivered via CI for the ND rats or acoustic clicks delivered over tube phones for the NH animals). Correct responses were reinforced with drinking water. ITD values of individual pulses were jittered, pseudorandomely, within the physiological range. Presented trials were divided into honesty trials, where all pulses were either left or right leading, and probe trials with ITDs shifting between left and right for each of the eight pulses. TWFs were subsequently calculated using multiple regression analysis to determine the perceptual weight of each pulse in the train of the probe trials.

Both NH and ND rats had temporal weighting functions comparable with human subjects from previous studies in that they tended to put more weight on the first pulse in the train than on subsequent pulses. However, while the onset cues dominate ITD perception with CI stimulation, this up-weighting does not seem to be as pronounced as in normal hearing listeners. These findings are relevant to the design of future CI processing strategies aiming to provide better sound source localization and spatial unmasking with CIs.

Both NH and ND rats had temporal weighting functions comparable with human subjects from previous studies in that they tended to put more weight on the first pulse in the train than on subsequent pulses.

These findings suggest that, while the onset cues dominate ITD perception with CI stimulation, this up-weighting does not seem to be as pronounced as in normal hearing listeners. These findings are relevant to the design of future CI processing strategies aiming to provide better sound source localization and spatial unmasking with CIs.