Charles S. Watson
Ward R. Drennan
Dept. of Speech and Hear. Sci., Indiana Univ., Bloomington, IN 47405
``Profile'' stimuli consisting of multiple simultaneous fixed-frequency sinusoidal components are more representative of naturally occurring sounds than the spectrally simpler waveforms more often used in psychoacoustic experiments. However, most naturally occurring sounds are characterized by dynamic rather than static spectra, and by harmonically spaced rather than the log-spaced components used in most profile experiments. In preparing to study a variety of dynamic profiles, the discriminability of static and frequency-glide profiles was determined, using both log- and harmonically spaced components. Discriminations were based on the detection of an intensity increment added to the mid-frequency component of 11-component, 400-ms profiles. Each profile had a starting frequency range of 200 to 2000 Hz. Dynamic profiles increased in frequency continuously over their 400-ms durations. Each subject was run under four stimulus conditions (static-log, static-harmonic, dynamic-log, dynamic-harmonic) generally for a minimum of 2000 trials per condition, in an adaptive-tracking procedure. Mean differences between asymptotic thresholds for the stimulus conditions were small compared to differences among the subjects. Harmonic yielded somewhat lower thresholds than the log-spaced components, while very modest differences, if any, were found between static and dynamic profiles. [Work supported by grants from NIH/NIDCD and AFOSR.]