Binaural interaction in low-frequency neurons in inferior colliculus of the cat. III. Effects of changing frequency

Abstract

The effects of changing stimulus frequency on the interaural phase sensitivity of neurons in the inferior colliculus (IC) were studied in barbiturate-anesthetized cats in order to reexamine the issue of characteristic delay (CD). Since the results obtained with the interaural delay and binaural beat stimuli are similar, we used the averaged interaural delay curves and binaural beat period histograms as comparable expressions of a neuron's interaural phase sensitivity. When the averaged interaural delay curves at different frequencies are plotted on a common time axis, for some cells the resulting superimposed delay curves show peaks or troughs that coincide at some CD. For most cells, though, this method of detecting a CD by visual inspection yields ambiguous and uncertain results. Composite curves, computed from the average of all the normalized superimposed delay curves, are also not helpful for showing CD. In order to provide a more objective means of analyzing the data, we plotted the mean interaural phase versus the stimulating frequency and computed the linear regression line, using the mean square error as a measure of linearity. The slope of the regression line is the CD for the neuron, and the phase intercept is referred to as the characteristic phase (CP). Cells that display a CD at the peak discharge have a CP = 0.0 cycles, while those that show a CD at the minimum discharge have a CP = 0.5. Cells that exhibit a CP at any value other than 0.0, 0.5, or 1.0 will have a CD at some relative amplitude other than the peak or trough. For cells that exhibit a CD at the peak or trough, results of the analysis procedure using the phase-frequency plot correspond to those obtained from visual inspection. For cells that do not show a common peak or trough, the analysis procedure not only specifies the location of the CD but also provides a statistical criterion of the linearity. From this analysis about 60% of the runs were identified as satisfying the criteria for CD at the P less than 0.005 level and 71% of these CDs are between +/- 300 micros. Most CD cells do not have the CD at the peak or trough of the response. Our results differ from those found in previous studies but they are in essential agreement with the original concept put forth by Rose et al. (31). Some cels exhibit little change in the CD or CP with variations in intensity, while others display marked systematic shifts in both CD and CP. In general, the peaks and troughs of the composite curves show less variability with intensity than the CD.(ABSTRACT TRUNCATED AT 400 WORDS)