Responses of inferior colliculus neurons to time-varying interaural phase disparity: Effects of shifting the locus of virtual motion

1. Motion of sound sources results in temporal variation of the binaural cues for sound localization. We evaluated the influence of virtual motion on neural tuning to one of these cues, interaural phase disparity (IPD). Responses to dichotic stimuli were recorded from single units in the inferior colliculus of the anesthetized cat and gerbil (Meriones unguiculatus). Static IPDs were generated by presenting dichotic tone pairs with a constant phase offset maintained for the duration of the stimulus. Time-varying IPDs were generated by simultaneously presenting a pure tone to one ear and a phase- modulated tone to the other ear. Sets of time-varying stimuli consisted of modulations through partially overlapping ranges of IPD, corresponding to movement of a sound source through partially overlapping arcs in the horizontal plane. 2. In agreement with previous results, neuronal discharge was typically a peaked function of static IPD resulting from both binaural facilitation at favorable IPDs and binaural suppression at unfavorable IPDs. Responses to time-varying IPD stimuli appeared to be shaped by the same facilitative and inhibitory mechanisms that underlie static IPD tuning. Modulation toward the peak of binaural facilitation increased the probability of discharge, and modulation toward the peak of binaural suppression decreased the probability of discharge. However, it was also clear that IPD tuning could be significantly altered by the temporal context of the stimulus. For the vast majority of units in response to modulation through partially overlapping ranges of IPD the discharge rate profiles were generally nonoverlapping. This shift in IPD tuning induced by the virtual motion reflects the fact that the binaural interaction associated with a given IPD depends on the recent history of stimulation. In addition, modulation in opposite directions through the same range of IPDs often elicited asymmetric responses. These nonlinearities imply that most inferior colliculus neurons do not unambiguously encode a specific IPD, but instead may encode small changes of IPD occurring virtually anywhere within their receptive fields. In a few cases modulation through overlapping ranges of IPD elicited contiguous response profiles, indicating that for these units responses were determined entirely by instantaneous IPD. 3. The nonlinearity of responses to time-varying IPD stimuli could not be attributed to monaural entrainment to the phase-modulated signals, did not depend on the phase modulation waveform, and occurred irrespective of which ear received the phase-modulated signal. Responses were similar in cats and gerbils, suggesting that the underlying mechanisms are common to binaural processing in diverse mammalian species. 4. The consistent shifts in IPD tuning displayed by most neurons in our sample suggests that sensitivity to dynamic spatial cues is a general property of neurons in the inferior colliculus. A measure was developed to quantify the magnitude of the shifts in IPD tuning across a large sample of units. This index of motion sensitivity was distributed continuously across the sample of units and was independent of frequency tuning, preferred static IPD, and sharpness of tuning to static IPD. 5. The initial neural encoding of IPD is believed to occur through a process of coincidence detection or cross-correlation at the superior olivary complex. The present finding that IPD tuning in the inferior colliculus is so dependent on the dynamic context suggests that the output of the original cross-correlator must be modified in the ascending auditory pathway. These data reveal that within the inferior colliculus the neural representation of IPD, and consequently sound location, is influenced by movement of a sound source.