Processing of binaural stimuli by cat superior olivary complex neurons

Summary

A method was developed to record stereotactically from the cat Superior Olivary Complex (SOC) using glass micropipettes. Sound stimulation was given through a closed system that permitted independent variation of interaural time (Δtime) and intensity (Δint) differences. The most common binaural units found (n = 34) were ipsilateral excitatory, contralateral inhibitory (EI¹), cells of the Lateral Superior Olive (LSO). Some Medial Superior Olive (MSO) cells and presumed MSO ascending afferents were found but, as noted by other authors, we found it difficult to obtain single unit recordings from this nucleus. The LSO EI cells were mostly sensitive to higher frequencies and showed Peristimulus Time Histograms (PSTHs) consisting of a sharp “On” response followed by a plateau when stimulated with Best Frequency (BF) tone bursts or noise bursts. This “On” response was sensitive to Δtime and Δint such that ipsilateral time lead or intensity increase resulted in a stronger response. The response reached a minimum around zero Δtime or Δint. No sharp peaks or dips were seen in the physiological range needed for localization, instead the response increased with increasing ipsilateral lead or intensity to the maximum values tested (2048 μs Δtime, 30 dB Δint). In the physiological range the Δtime and Δint response were complementary (both increasing response as ipsilaterality was increased). Provided enough sound energy in the unit's sensitive region was present, the same Δtime curves were produced when BF tone bursts, masked tone bursts, “sharp onset” tone bursts or noise bursts were used. Changing the Δtime of the carrier of the tone burst alone had no effect (except for one cell with a BF of 560 Hz), only the relative time of arrival of the stimulus envelope seemed to be important. In contrast to these LSO EI cells MSO-type units showed EI or EE predominantly low frequency phase-locked responses. When stimulated with interaurally phase shifted (Δpha) BF tones the unit response was a cyclic function of Δpha. Some cells (all that were tested, n = 6 including the 560 Hz LSO EI cell) showed these cyclic responses when stimulated with noise bursts or non-BF tones. However, these “characteristic delays” were not necessarily in the physiological range, i.e. we could find no evidence that these units were responding to Δtime/Δpha values corresponding to a particular sound source direction. In both LSO and MSO it seems that integration of information higher in the CNS from a population of these cells is necessary for unambiguous coding of sound source direction. The time intensity trading ratios measured in two MSO type cells (11 and 26 μ/dB) were clearly different to those measured in LSO EI cells (n = 6, 99–550 μs/dB). These ratios correspond approximately to those of the psychophysical Δtime and Δint images measured by Hafter and Jeffress (1968).