The relationship between auditory brainstem response intensity functions and loudness

ABR wave I and wave V intensity-latency and intensity-amplitude functions and loudness estimation functions were obtained from eight normally hearing and eight hearing impaired listeners in a sound field situation. Data were collected from the hearing impaired listeners under three conditions of amplification: none, amplification with compression circuitry, and linear amplification. ABR and loudness functions were examined to determine if a relationship existed between the electrophysiological and psychophysical data. It was also determined if the use of compression amplification caused intensity functions generated by hearing impaired listeners to normalize.

When comparisons of F values were made, none of the ABR data produced the same statistical results across amplification conditions as the magnitude estimation data and no clear relationship was inferred. Although data from compression aided hearing impaired listeners tended to approach mean values of the normally hearing group, analysis of variance revealed that statistically significant differences existed between groups for three of the dependent variables. In the case of the other two dependent variables, no statistically significant differences existed, but statistical non-significance also occurred when the hearing impaired listeners were unaided. Correlational analysis revealed high correlations between magnitude estimation and wave V amplitude (r=.73) for normally hearing listeners and between magnitude estimation and wave I amplitude (r=.73) for compression aided hearing impaired listeners. The strength of these correlations, however, would not permit the use of either ABR intensity-amplitude function to predict loudness growth.

Although this study did not rule out the use of the ABR test procedure to fit hearing aids, assumptions about aided loudness perception should not be made based on information from the intensity functions. In a few cases, the ABR intensity functions appeared to be related to audiometric contour rather than to loudness perception.