Neurons sensitive to interaural phase disparity in gerbil superior olive: Diverse monaural and temporal response properties

1. We assessed mechanisms of binaural interaction underlying detection of interaural phase disparity (IPD) by recording single-unit responses in the superior olivary complex (SOC) of the anesthetized gerbil (Meriones unguiculatus). Binaural responses were obtained from 58 IPD-sensitive single units, 44 of which were histologically localized. Monaural responses were also obtained for 52 of 58 IPD-sensitive units. Additionally, responses were recorded from 16 units (best frequency < 2.4 kHz) in lateral SOC that were excited by ipsilateral stimulation and inhibited by contralateral stimulation (EI), none of which was IPD sensitive. Our results are consistent with a mechanism of binaural interaction involving detection of coincident excitatory inputs from the two ears. There was no compelling evidence of binaural sensitivity arising from IPD-dependent interactions of phase-locked excitatory and inhibitory inputs from the two ears. Despite the uniformity of binaural interactions, considerable diversity of temporal and monaural response properties was observed. 2. Monaural and binaural responses of 35 of 58 IPD-sensitive units were phase locked to the period of low-frequency (<2.5 kHz) tones. Most phase-locking units were bilaterally excitable and, consistent with the coincidence-detection model, their IPD selectivity could be predicted from the difference between the mean phases of the monaural responses. The remaining units (23 of 58) did not phase lock in response to monaural or binaural tones. Most non-phase-locking units failed to respond to monaural stimulation of one or both ears (monaurally unresponsive units). 3. Some IPD-sensitive units were inhibited by monaural stimulation of the ipsilateral ear or both ears. A few units responded only at the onset of monaural and binaural tones. Phase locking was present in responses of some, but not all, of these monaurally inhibited and onset units. 4. Most IPD- sensitive neurons were encountered at sites within or immediately adjacent to the cell column of the medial superior olive (MSO). IPD-sensitive units were also recorded in the lateral superior olive (LSO), in the superior paraolivary nucleus (SPN), and within a region forming a medial-dorsal cap around MSO. Bilaterally excitable units were concentrated around MSO, but were also encountered in SPN, the medial-dorsal region, and LSO. Some monaurally unresponsive units were recorded in the vicinity of MSO, but most were located in the medial-dorsal region. Monaurally inhibited units were localized to the medial border of the MSO cell column or to SPN. Onset units were localized to SPN and the medial dorsal region. EI units were located exclusively in LSO. 5. Despite the diversity of monaural and temporal response properties, binaural interactions underlying IPD tuning were remarkably uniform. Responses were generated at best IPD by facilitation or summation (45 of 46 units) and at worst IPD by suppression or occlusion (35 of 41 units). No differences were evident in the binaural interactions underlying IPD tuning of units with different monaural and temporal response properties or locations. 6. IPD tuning was characterized at multiple tone frequencies for 34 units. For most units, the interaural time difference (ITD) tuning functions computed at multiple frequencies coincided at their peaks. There were no significant differences between means for units with different monaural and temporal response properties. Most units (81%) had characteristic delays corresponding to delays of the stimulus at the ipsilateral ear (mean 241 μs). 7. Mean binaural response latencies for phase-locking and nonphase-locking units were 5.1 ± 0.2 (SE) ms and 12.5 ± 1.6 ms, respectively. Mean latencies of non-phase locking units were significantly different from those of phase-locking units (P < 0.05), but were not significantly different from those of inferior colliculus neurons. 8. These data are consistent with a neural mechanism of IPD coding involving detection of coincidence of excitatory events originating from the two ears. Synaptic inhibition within MSO may participate in IPD coding by modulating this excitatory-excitatory interaction. The diversity of monaural and temporal response properties may be indicative of functionally and anatomically distinct populations of IPD-sensitive neurons within the gerbil SOC. It is proposed that phase-locking units correspond to primary binaural comparators and that non-phase locking units correspond to higher order binaural neurons.