Comparison among four psychophysical procedures used to assess sound localization

Yost et al. (1974; Percept. Psychophys. 15, 483–487) measured the effect of psychophysical procedure to evaluate movement-related cues for auditory lateralization. A similar approach was used in the current work to assess sound-source localization in a 10-foot x 13-foot semi-anechoic sound field. Performance was measured in four listening conditions, a localization task utilizing a pointer response and three location-discrimination tasks (single-interval, same-different, or 2AFC). Four signals were used: a 750-Hz tone, broadband noise, or two versions of broadband noise with narrowband levels roved + /−20 dB in 1/3-octave bands between intervals, or frozen across intervals with the same rove. Stimuli were generated as phantom sources using tangent-law panning between + /−30° azimuth loudspeaker positions. For each signal type, d′ was lower in the pointer condition than in the three discrimination tasks. Across tasks, performance was best for broadband noise and noise frozen across intervals, relative to the pure-tone and fully roved signals. As suggested by Yost etal., task effects may relate to source movement cues present in only the two-interval discrimination conditions. Alternatively, slow fluctuations in bias may negatively affect only the pointer localization task, with results from single-interval discrimination offering some support for this interpretation.