Reliability and validity of high-frequency (8-20 kHz) thresholds obtained on a computer-based audiometer as compared to a documented laboratory system.

A Macintosh computer-based audiometer (Virtual 320) was evaluated for reliability and validity of high-frequency (8-20 kHz) thresholds by comparison with a well documented laboratory high-frequency evaluation system (PARVA-HF). High-frequency earphones originally provided with the V320 for high-frequency testing required modification to improve reliability in calibration and in subject threshold testing. Twenty normal-hearing adults were evaluated in the 8-20 kHz frequency range on both testing systems. Results of intrasubject multiple-session testing were evaluated to determine the reliability of high-frequency thresholds obtained. The V320 produced reliable results comparable to the PARVA-HF. Validity of high-frequency thresholds was inferred by comparing V320 responses to those obtained with the PARVA-HF. Comparable findings between systems imply validity of thresholds obtained with the V320. Conventional frequency (0.25-8 kHz) threshold evaluation with the V320 and a Grason-Stadler 1701 audiometer also yielded comparable results.