A comparison of power-based and admittance measures of the acoustic reflex threshold

Abstract

Contralateral acoustic reflex thresholds obtained using a standard clinical admittance system with a 220-Hz probe tone were compared to those obtained using a frequency-domain-based system for calculating wideband middle ear acoustic impedance and energy reflectance (ZR) [D. H. Keefe et al., J. Acoust. Soc. Am. 91, 470–485 (1992)]. A white noise activator presented through an ER-3A insert phone was used to elicit the contralateral reflex in ten normal-hearing young subjects. The probe signal, a 40-ms wideband chirp, was presented through an ER-10C microphone system for the ZR measures. The microphone signal was then digitized and stored for data analysis. Measurements of reflex-induced changes in the power absorbed by the middle ear and in wideband admittance were then derived to estimate reflex thresholds. An analysis of variance revealed that the average reflex threshold was about 16 dB lower for the power-based and wideband admittance measures than for the clinical measure (p<0.001). However, the two ZR measures were not significantly different from each other. These results have implications for reducing the level of sound exposure during clinical acoustic reflex testing, and for obtaining reflex thresholds that may be beyond system limits for standard clinical systems. [Work supported by OURC.]