Authors:

GOCKEL, H.E., CARLYON, R., Farooq, R., Muhammed, L., & Plack, C.

Reference:

Year of publication:

2012

CBU number:

7390

Abstract:

Studies of the pitch of complex tones often use noise to mask distortion products such as the cubic difference tone (CDT), which, for 2 components F1 and F2, has a frequency of 2F1-F2. This is true both for psychophysical studies and for measures of the frequency following response (FFR) - a scalp-recorded measure of phase-locked brainstem activity that is often assumed to reflect residue pitch. However, whereas audible CDTs probably arise from cochlear processes and are propagated along the basilar membrane, a component at the CDT frequency may also arise in the FFR from nonlinearities in the responses of neurons that are driven by more than one harmonic. To assess this we compared the spectral magnitude of the FFR at the CDT frequency in response to a complex tone, to that in response to a pure tone of identical frequency presented at various levels. The FFR was measured for six subjects, using 100-ms 75-dB-SPL complex tones containing components 2-4 of a 300-Hz fundamental frequency with all harmonics shifted together up or down from their nominal (harmonic) frequencies by 56 Hz. This frequency-shifted complex allowed independent assessment of the levels of the CDT and the quadratic difference tone at F2-F1 (the envelope rate). Stimuli were presented in alternating polarity at a rate of 3.57 Hz. We used a “vertical” montage (+ Fz, - C7, ground=Fpz) for which the FFR is assumed to reflect phase-locked neural activity from rostral generators in the brainstem. FFR waveforms for each polarity were averaged and then either added, to enhance responses related to stimulus envelope, or subtracted, to enhance temporal fine structure responses. The effective CDT level in the subtraction waveform of the FFR was substantially higher than that commonly estimated in psychophysical experiments, indicating (possibly neural) contributions to the CDT in the FFR beyond that originating from the audible propagated component. Supported by Wellcome Trust Grant 088263.