A PROBE-SIGNAL STUDY OF AUDITORY-DISCRIMINATION OF COMPLEX TONES

The present experiments used an analogue of the probe-signal method of Greenberg and Larkin [J. Acoust. Sec. Am. 44, 1513-1523 (1968)] to in vestigate the extent to which listeners direct attention to a particul ar spectral region when discriminating complex tones. The listeners' t ask was to discriminate between two seven-component complex tones on t he basis of an increment in the level of a single component. On two-th irds of trials the increment was achieved by adding a fixed primary si gnal to one component of the complex. The primary-signal trials were r elatively easy and were intended to cue listeners to attend to the com ponent to which the primary was added. On the remaining trials a small er probe signal was added either to the cued component, or to one of t hree other components. The results of the first experiment, in which t he complex tones had a flat spectrum, showed significantly better perf ormance for probe signals applied to the cued component compared to th e other three components. To control for the possibility that the obse rved pattern of results was due to the use of timbral cues, a second e xperiment was conducted in which the spectral profile of the tones was randomized between trials. The results for the second experiment were similar to those of the first experiment and are consistent with the idea that listeners were focusing attention on the spectral region def ined by the primary signal. (C) 1997 Acoustical Society of America. [S 0001-4966(97)03909-X] PACS numbers: 43.66.Fe, 43.66.Lj [RHD].