Characterizing inner-hair-cell specific dysfunction from spike-train-derived transduction functions using a phenomenological auditory-nerve model

The impact of outer-hair-cell damage on sensorineural hearing loss (SNHL) has been extensively studied compared with inner-hair-cell (IHC) damage (e.g., stereocilial). Pre-clinical SNHL animal models provide unique data to directly address IHC-specific deficits (e.g., carboplatin-exposed chinchillas). Spike-train data (period histograms) to low sound level tones can be used to derive IHC-transduction functions by mapping instantaneous spike rates to corresponding sinusoidal pressure values (Horst et al., 2018). However, this approach depends on sensation level and spontaneous rate, which are both affected by IHC dysfunction. To better understand the effect of these dependencies, we used a phenomenological auditory-nerve (AN) model (Bruce et al., 2018) that provides parametric control over IHC and OHC dysfunction allowing exploration of optimal experimental design. Here, we explore the utility of spike-train-derived IHC-transduction functions in capturing parametric changes in IHC dysfunction as currently implemented in the AN model. We used unsupervised methods and information-theoretic approaches to quantify the dependency of transduction functions on IHC dysfunction. We also compared model findings with AN-fiber data recorded from carboplatin-exposed chinchillas. Preliminary findings suggest that transduction functions obtained from the AN model can capture IHC-specific dysfunction within specific regimes of controllable model parameters, thus showing promise for characterizing IHC dysfunction.