| Summary: | The auditory brainstem response (ABR) is an evoked potential that has achieved widespread acceptance as a technique for evaluating the status and function of the auditory nervous system. For many diagnostic applications, the latency of an obtained ABR peak is compared to clinical norms. One who uses this approach makes some basic assumptions regarding between-subject and within-subject variability of latency. Although a great deal is known about between-subject variability of ABR latency, virtually nothing is known about such variability within a single subject. The purpose of this investigation was to describe the nature of within-subject variability of ABR latency. Nine male subjects participated in the study. Each met the following criteria: 10-12 years of age; normal speech and language development; normal academic progress; normal hearing; and, normal middle ear pressure. A repeated measures design was employed. Four sessions were scheduled for each subject and five ABRs were obtained at each session for each of three stimulus conditions: monaural left, monaural right, and binaural. Stimuli were 100 μs condensation clicks presented at 80 dB nHL. For each ABR peak, the within-subject distribution of latencies was analyzed with regard to symmetry, kurtosis, range, and standard deviation using the SPSSx "Descriptives" procedure. For every subject, variability of latency was observed. Most often, the latencies were normally distributed and the magnitude of variability was small. The variability of latency, as indexed by the standard deviation, was less within any single subject than is commonly reported for groups of subjects. It was concluded that: (a) standard parametric techniques would be appropriate for subsequent analysis of such data; and, (b) by establishing a baseline, the sensitivity of the ABR might be increased for certain within-subject monitoring applications.
|
|---|
