Analysis and modeling of click and chirp evoked auditory steady state responses

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)

First Committee Member

Ozcan Ozdamar - Committee Chair


This study aims to investigate the possible advantages of using cochlear chirp stimulus instead of click for recording auditory steady-state responses (ASSRs), and to test the validity of the superposition hypothesis for the generation of the ASSRs through simulations, and modeling. Simulation of the ASSRs at different stimulation rates was achieved by shifting and adding of the transient auditory evoked potentials (AEP) with and without the adaptation effects incorporated into the model.ASSRs to 100 musec clicks and 4 msec cochlear chirps (60 dBnHL) were recorded from 9 normal hearing adult subjects (11 ears) at repetition rates of 20 to 100 Hz in 10 Hz increments. Auditory brainstem responses (ABRs) and middle latency responses (MLRs) to both stimuli were recorded at 10 Hz in the same session as well. A special synthetic AEP waveform was prepared for modeling purposes using gammatones.Chirp evoked ABRs had significantly larger Wave V amplitudes as reported in the literature. Similarly, chirp evoked MLRs had significantly larger P a amplitudes than the corresponding click evoked responses. The amplitude-rate plots of both click and chirp evoked ASSRs were similar in terms of general shape, but chirp stimuli evoked significantly larger amplitudes. The signal-to-noise ratios (SNR) for the chirp evoked ASSRs were also greater than those of the click evoked ASSRs. The simulations without adaptation predicted the real 40 Hz responses well, but had larger amplitudes at most other rates. The inclusion of the adaptation effects of increasing stimulus rate improved the simulation results and resulted in better predictions of the real ASSR measurements.Simulations using the synthetic AEP waveform resulted in ASSR amplitude-rate plots similar to those obtained in real recordings. The synthetic 40 Hz ASSR diminished when the MLR component was eliminated from the AEP equation, and the 80 Hz ASSR diminished when the slow ABR component was eliminated. These modeling and simulation results support the theory that the ASSRs are predominantly generated by overlapping transient responses, and the slow ABR is the main contributor to 80 Hz responses, whereas overlapping MLRs mostly generate the 40 Hz responses.


Engineering, Biomedical

Link to Full Text


Link to Full Text