ELECTROPHYSIOLOGIC TECHNIQUES IN AUDIOLOGY AND OTOLOGY: PDF OnlyHigh-Pass Digital Filtration of the 40 Hz Response and its Relationship to the Spectral Content of the Middle Latency and 40 Hz ResponsesKavanagh, Kevin T.; Domico, William D.Author Information The University of Tennessee, Memphis; Department of Otolaryngology and Maxillofacial Surgery, and Memphis State University Department of Audiology and Speech Pathology, Memphis, Tennessee Address reprint requests to Kevin T. Kavanagh, M.D., The University of Tennessee, Department of Otolaryngology, Rm. 2824 Coleman Building, 956 Court Ave., Memphis, TN 38163. Received June 18, 1985; September 18, 1985. Ear and Hearing: April 1986 - Volume 7 - Issue 2 - p 93-99 Buy Abstract This study analyzed the effects of high-pass filtration on the 40 Hz response using zero and standard phase shift digital filtration. Filtration effects were compared to the spectral content of the middle latency response (MLR) and 40 Hz response. The spectral composition of the MLR was found to contain major response energy at 10 and 40 Hz. The major energy of the 40 Hz response was at 40 Hz with minor peaks every 40 Hz above the fundamental frequency. The elimination of the 10 Hz response energy in the 40 Hz response was due to cancellation caused by phase differences which occurred in the production of a steady state potential. The reduction was confirmed by the minor amplitude effects created by zero-phase-shift high-phase filtration (cutoff frequency 30 Hz with 36 dB/oct slopes) on the 40 Hz response as compared to the MLR. The only beneficial effect of high-pass filtration was the elimination of the mild undulation of the unfiltered waveform. The major effect of phase-shift high-pass filtration on the 40 Hz response was the displacement of the high frequency ABR from the peak to the descending slope and finally to the trough of the MLR. Shifting of the ABR from the peak of the MLR resulted in amplitude loss which may affect response identification at threshold. The authors recommend minimal highpass filtration (i.e., less than 12 dB/oct slope and a cutoff frequency less than or equal to 15 Hz) when recording the 40 Hz response. The spectral analysis of the evoked response may be able to predict the optimal rate for eliciting a steady state response in divergent subject populations. © Williams & Wilkins 1986. All Rights Reserved.