Assessing the Firing Properties of the Electrically Stimulated Auditory Nerve Using a Convolution Model

Area of Science:

• Neuroscience
• Biomedical Engineering
• Audiology

Background:

• The electrically evoked compound action potential (eCAP) is a standard measure for assessing the auditory nerve in cochlear implant recipients.
• A convolution model can extract additional neural firing properties from eCAPs, potentially indicating auditory nerve health.

Purpose of the Study:

• To investigate the relationship between neural firing properties and nerve histology using a convolution model in guinea pigs with varying degrees of nerve degeneration.
• To assess the clinical applicability of this model in human eCAPs.

Main Methods:

• Applied a convolution model to eCAPs from guinea pigs with induced nerve degeneration and from human cochlear implant users.
• Analyzed neural firing probability, latency, and component scaling factors.
• Correlated model parameters with nerve histology in guinea pigs.

Main Results:

• Auditory nerve firing probability was predominantly bimodal, characterized by two Gaussian distributions with a 0.4 ms latency difference.
• The ratio of late to early firing components increased with neural degeneration in guinea pigs and decreased with stimulation intensity in humans.
• Early component latency decreased with neural degeneration in guinea pigs, with similar trends observed in humans.

Conclusions:

• Deconvolution of eCAPs reveals two distinct firing components in the auditory nerve, offering valuable insights beyond traditional analyses.
• The model demonstrates potential for assessing auditory nerve health and clinical applicability in cochlear implant users.
• Observed species-specific differences in firing component characteristics highlight the complexity of auditory nerve responses.