Contribution
A computational model of human auditory evoked potentials at the level of the periphery and brainstem
* Presenting author
Abstract:
Auditory evoked potentials (AEP) are electrophysiological measures used as objective diagnostic tools. However, the relationship between AEP morphology and cochlear status is complex, making interpretation challenging. Here, an AEP modeling framework was developed using a state-of-the-art computational auditory nerve (AN) model and a transformation corresponding to a given electrode configuration (e.g., for peripheral and brainstem potentials). AEPs to periodic stimuli (e.g. pure tones and amplitude-modulated tones, representing frequency- and envelope-following responses, respectively) were simulated across a range of levels. Additionally, compound action potentials and auditory brainstem responses to transients were simulated and compared to recordings in 20 young, normal-hearing participants at a range of levels. The model successfully replicated full AEP waveforms for transient and periodic stimuli across stimuli levels. Additionally, it was used to explore AEP generation mechanisms, with the interaction of AN fibers explaining several phenomena observed in experimental AEPs. However, discrepancies in the latency-level functions and the amplitude growth functions suggested that the spread of excitation may not be accurately modeled in the current human AN model. This modeling work has the potential to test hypotheses and interpret experimental findings related to the effects of cochlear damage, such as neural versus hair cell loss, on AEP patterns.