Auditory temporal processing: responses to sinusoidally amplitude-modulated tones in the inferior colliculus.

Abstract

Time-varying envelopes are a common feature of acoustic communication signals like human speech and induce a variety of percepts in human listeners. We studied the responses of 109 single neurons in the inferior colliculus (IC) of the anesthetized Mongolian gerbil to contralaterally presented sinusoidally amplitude-modulated (SAM) tones with a wide range of parameters. Modulation transfer functions (MTFs) based on average spike rate (rMTFs) showed regions of enhancement and suppression, where spike rates increased or decreased respectively as stimulus modulation depth increased. Specifically, almost all IC rMTFs could be described by some combination of a primary and a secondary region of enhancement and an intervening region of suppression, with these regions present to varying degrees in individual rMTFs. rMTF characteristics of most neurons were dependent on sound pressure level (SPL). rMTFs in most neurons with "onset" or "onset-sustained" peri-stimulus time histograms (PSTHs) in response to brief pure tones showed only a peaked primary region of enhancement. The region of suppression tended to occur in neurons with "sustained" or "pauser" PSTHs, and usually emerged at higher SPLs. The secondary region of enhancement was only found in eight neurons. The lowest modulation frequency at which the spike rate reached a clear peak ("best modulation frequency" or BMF) was measured. All but two mean BMFs lay between 0 and 100 Hz. Fifty percent of the 49 neurons tested over at least a 20-dB range of SPLs showed a BMF variation larger than 66% of their mean BMF. MTFs based on vector strength (tMTFs) showed a variety of patterns; although mostly similar to those reported from the cochlear nucleus, tMTFs of IC neurons showed higher maximum values, smaller dynamic range with depth, and a lower high-frequency limit for significant phase locking. Systematic and large increases in phase-lead commonly occurred as SPL increased. rMTFs measured at multiple carrier frequencies (F(c)s) showed that the suppressive region was not the result of sideband inhibition. There was no systematic relationship between BMF and F(c) of stimulation in the cells studied, even at low carrier frequencies. The results suggest various possible mechanisms that could create IC MTFs, and strongly support the idea that inhibitory inputs shape the rMTF by sharpening regions of enhancement and creating a suppressive region. The paucity of BMFs above 100 Hz argues against simple rate-coding schemes for pitch. Finally, any labeled line or topographic representation of modulation frequency is unlikely to be independent of SPL.