PHYSIOLOGICAL CORRELATES OF THE PRECEDENCE EFFECT AND SUMMING LOCALIZATION IN THE INFERIOR COLLICULUS OF THE CAT

The precedence effect (PE) describes an illusion produced when two sim ilar sounds are delivered in quick succession (interclick delays of 2- 8 msec) from sound sources at different locations so that only a singl e sound is perceived. The localization of the perceived sound is domin ated by the location of the leading sound. If the delays are very shor t (<1-2 msec), summing localization occurs and a phantom source is per ceived whose location is toward the leading sound. The purpose of thes e experiments was to look for physiological correlates of the preceden ce effect and summing localization by recording from single neurons in the inferior colliculus of the anesthetized cat. Click stimuli were d elivered under two different situations: over headphones in dichotic e xperiments and through two speakers in an anechoic room in free-field studies. In the latter case the cat was placed midway between the spea kers and a single click stimulus was delivered to each speaker with va riable interclick delays (ICDs). Most cells, under both dichotic and f ree-field conditions, exhibited a form of the precedence effect in whi ch the response to the lagging click was suppressed when ICDs were sho rt. The suppression of the lagging click, or echo, was measured by rec overy curves, which plotted the response of the lagging click as a fun ction of ICD. There was considerable variability in the recovery curve s from different cells: the ICDs at which the recovery reached 50%, wh ich is a measure of the echo threshold for the cell, ranged from 1 to 100 msec with a median of 20 msec. Human psychophysical experiments re port echo thresholds far clicks ranging from 2 to 8 msec. If we assume that absolute echo threshold is determined by the cells with shortest recovery curves, then the thresholds for single cells are in accord w ith the psychophysical results. The possible sites of generation of th e echo suppression are also considered. Changes in the relative level of the leading and lagging clicks produced the expected shifts in the recovery curves. With short ICDs in the summing localization range (be tween about +/-2 msec), cells arse showed responses consonant with the human psychophysical result that the sound source is localized to a p hantom image between the two speakers and toward the leading one. The location of the image varied systematically with the relative levels o r ICDs of the clicks. These results suggest that the neural substrate for many of the psychephysical effects of the PE and summing localizat ion are seen at the level of the ICC and therefore lend support to the two basic assumptions: that the cat also experiences the PE and that the discharge of at least some cells in the ICC is related to the perc eived location of the sound source.