A behavioral measure of neural adaptation in normal-hearing and cochlear-implant listeners.

Two experiments tested the hypothesis that adaptation of the neural response to pulsatile stimuli is greatest at high pulse rates. In experiment 1, normal-hearing listeners loudness-balanced bandpass-filtered harmonic complexes having different envelope repetition rates. The complexes had durations of 40, 120, or 400 ms and resembled filtered pulse trains. All had a fundamental frequency of 50 Hz and were generated using different component phase relationships to yield repetition rates of 50, 100, 200, or 450 pulses per second (pps). In experiment 2, cochlear-implant listeners loudness-balanced electrical pulse trains presented to a single electrode, also at durations of 40, 120, and 400 ms. The pulse trains had rates of 100 and 2000 pps. Both experiments showed that lower-rate stimuli needed a higher level to match the loudness of higher-rate stimuli, with this effect decreasing with increasing duration. In experiment 2, this duration effect was only observed in trials where the 2000-pps train was presented before the 100-pps train, consistent with the idea that participants compared the end of the first stimulus in each trial to the start of the second. We conclude that the results of these two experiments are consistent with rate-dependent adaptation, and we discuss theoretical and clinical implications.