Discrimination of complex tones with unresolved components using temporal fine structure information

Abstract

The information used to discriminate complex tones with (largely) unresolved components was assessed. In experiment 1, subjects discriminated a harmonic complex tone, H, with fundamental frequency F0, from an inharmonic tone, I, in which all components were shifted upwards by the same amount in hertz. Tones H and I had the same envelope repetition rate but different temporal fine structure (TFS). The tones were passed through a fixed bandpass filter centered on harmonic N, to reduce excitation pattern cues. For all F0s (35-400 Hz), performance decreased as N was increased from 11 to 15, but, except for F0=35 Hz, remained above chance for N=15, where all harmonics should be unresolved. This suggests that discrimination can be based on TFS rather than on partially resolved components. In experiment 2, subjects discriminated the F0 of complex tones filtered as in experiment 1. Here, both envelope rate and TFS cues were available. Except for F0=35 Hz, discrimination thresholds, expressed as the Weber fraction for a change in time interval, were similar to those measured in experiment 1, suggesting that performance in experiment 2 was dominated by the use of TFS rather than envelope cues.