Norms for a headphone simulation of the Hearing in Noise Test: Comparison of physical and simulated spatial separation of sound sources

Abstract

Norms for a headphone-based procedure for administering the Hearing in Noise Test has been developed [Nilsson etal., J. Acoust. Soc. Am. 95, 1085–1099 (1994)]. Sentence speech reception thresholds (SSRTs) were measured adaptively in the presence of spectrally matched noise for 23 young, normal hearing male and female adults using two presentation systems: a sound field presentation and a headphone simulation of the sound field. Speech was presented at 0° azimuth in all conditions, and noise (when present) was presented at either 0°, 90°, or 270° azimuth at 65 dB(A). The headphone simulation used head-related transfer functions (HRTFs) measured off a KEMAR mannequin to introduce the appropriate interaural cues. Correction factors between the current headphone and sound field measurements were calculated to correct previous sound field norms. SSRTs were based upon 3 repeated measurements in each condition, and differences between the sound field and headphone data varied between approximately 0.5 and 2.5 db, depending upon the condition. SSRTs in quiet are elevated in the headphone system, most likely because of a higher noise floor caused by the instrumentation. SSRTs in noise are all lower in the headphone system, attributable to the elimination of room and speaker effects. Improvements in SSRTs with spatial separation of the signal and masker were 6.38 dB in the sound field and 6.65 dB under headphones. [Work supported by Hoover Foundation.]