Objective versus visual detection of the auditory brain stem response.

The use of objective (mathematical) procedures in auditory brain stem response (ABR) testing has been suggested in order to eliminate the subjectivity of visual judgments and to increase sensitivity of detection. This investigation compared with effectiveness of visual and objective detection of the ABR. Auditory brain stem responses were obtained in normal adults at several click levels. An equal number of waveforms obtained without stimulation were used to measure false-positive responses. Visual judgments of ABR presence were made by four experienced raters. Objective analysis was accomplished using three methods: correlation, variance ratio, and multiple pre-post z tests. The mean sensitivity score was highest for visual detection, followed in order by the correlation, variance ratio, and multiple z test methods. While visual scoring was statistically the most sensitive, the practical difference between measures was small. In addition, there was considerable disagreement between raters in their judgments. Therefore, the most sensitive of the objective measures--correlation, is considered to be a useful detection procedure to eliminate observer bias and to achieve consistency of scoring.