Discrimination of envelope frequency

Abstract

Discrimination of envelope frequency was measured [in humans] as the just noticeable increase in envelope frequency of monotic, 60-dB 2-tone complexes at geometric center frequencies (CF) of 500, 1000, 2000 and 4000 Hz. (The envelope frequency is equal to the frequency separation, .DELTA.F, between the 2 components.) At a given CF the jnd [just noticeable differences] were approximately constant up to a critical envelope frequency between 10% and 20% of the CF, roughly equivalent to the critical band, beyond which they increased to a maximum at about 40% of the CF. Below the critical envelope frequency, the jnd increased with CF. Measurements of 40 dB SPL [sound pressure] with and without masking of the aural distortion products showed that that influence of aural distortion is minimal. Additional measurements showed that at narrow frequency separations, discrimination was better for monotic 2-tone complexes than for pure tones, 3-tone complexes, and dichotic 2-tone complexes. Envelope frequency may serve as a cue for discrimination up to a frequency between 30% and 40% of the CF, although its effectiveness at high envelope frequencies is severely diminished by the auditory filter. A comparison with other data on envelope frequency discrimination indicates that the jnd dependence on the envelope frequency may depend on how the slope of the temporal envelope varies with envelope frequency.