Modeling Binaural Hearing Percepts of Bilateral Cochlear Implant Users
* Presenting author
Cochlear implants (CIs) restore hearing for profoundly deaf people by stimulating the auditory nerve fibers (ANFs) directly with electrical pulses. Several stimulus- and ANF-related aspects affect the responsiveness of the ANF to a particular stimulation. Here, we present a functional model for the ANF response to pulse-train sequences and show how it can be applied, in conjunction with functional models for binaural-cue decoding (Takanen et al., Hear. Res. 2014), to predict bilateral CI users’ sensitivity to localization cues. The ANF model builds on the biphasic leaky integrate-and-fire model by Horne et al. (Front. Comput. Neurosci., 2016) which we have extended to include elements that simulate refractoriness and facilitation/accommodation by affecting the model’s threshold value momentarily after supra- and subthreshold stimulation. We show that our revised model can reproduce neurophysiological data from single-neuron recordings considering refractoriness, facilitation, accommodation and spike-rate adaptation phenomena that affect the responsiveness of the ANF to ongoing pulsatile stimulation. Outputs from the binaural CI model are shown to demonstrate pulse-rate dependency of just-noticeable differences for localization cues in electrical hearing. Consequently, the model offers a versatile instrumental tool for optimization of CI coding strategies, being able to predict both peripheral and perceptual responses that different stimulations evoke.