Auditory guided search: Free-field and synthesized

Abstract

Target discrimination tasks effectively test the ability to locate and interpret a given target. Adding sound, spatially located with the target, was hypothesized to decrease time necessary to localize and discriminate that target. This spatially correlated sound (SCS) was compared to spatially uncorrelated sound (SUS) at 260 points surrounding the subjects. Average reaction times of a two-alternative forced-choice discrimination task were better by over 150 ms in the frontal field (FE, +/−40 deg azimuth × +/−40 deg elevation) for SCS compared to SUS. Performance declined as targets moved into the periphery for both conditions, at an accelerated rate for SUS. When in the rear hemifield (RHF), SCS performance fell to almost two times what it was in the FF (780 vs 1325 ms), while it was almost three and one half times for SUS (934 vs 3310 ms). Spatially correlated synthesized sound (SSS) presentation over headphones proved to be about as accurate as SCS. Performance decline of FF vs RHF was about 2.5 to 1. These values show, especially in the RHF, ALCS’s provide almost as much information about target location as SCS cues and provide a subject with considerably more resources to succeed in a target acquisition task compared to SUS. [Work supported by NSF and AFOSR.]