The relation between ECAP measurements and the effect of rate on behavioral thresholds in cochlear implant users

It has been shown that electrically evoked compound action potential (ECAP) thresholds are not sufficiently predictive of behavioral thresholds to allow their use as a totally objective method to program a cochlear implant. Previous animal studies have shown that two other ECAP parameters (the ECAP amplitude growth slope and the effect on ECAPs of changing the phase duration (PD) or interphase gap (IPG)), and the way that behavioral thresholds change with increasing rate of stimulation, are associated with cochlear health. This experiment tested the hypotheses that a) the degree to which behavioral thresholds change with rate of stimulation is associated with either or both of those two ECAP parameters, and that b) the accuracy of ECAP thresholds for predicting behavioral thresholds at clinically relevant rates can be increased by including those additional ECAP parameters. Both these hypotheses were confirmed by the data. The ECAP slope was associated with within-subject variation across electrode positions of both behavioral thresholds and the change of thresholds with increasing rate. The effect of changes in IPG or PD on ECAPs was moderately associated with between-subjects variations in both average absolute behavioral thresholds and the average effect of rate on thresholds. The inclusion of the IPG/PD effect to predict average absolute behavioral thresholds for each subject and inclusion of the ECAP growth slope to predict variation in relative thresholds across electrode positions in the same subject led to a significant increase in accuracy of the predicted behavioral thresholds.