Level Dependent Shifts in Auditory Nerve Phase Locking Underlie Changes in Interaural Time Sensitivity with Interaural Level Differences in the Inferior Colliculus

Abstract

Interaural time differences are initially analyzed in the medial superior olive (MSO) in the brainstem. Neurons in this nucleus act as coincidence detectors, only firing when the activity from the two ears reaches the cell within a small time window (Batra et al. 1997a, b; Goldberg and Brown 1969; Spitzer and Semple 1995; Yin and Chan 1990). Maximal values of interaural time difference (ITD) for humans are 700 μs, with just noticeable differences often of the order of a few tens of μs (Durlach and Colburn 1978; Hafter et al. 1979; Mills 1958). To achieve such accuracy requires a very precise time signal from the two ears, which is provided by the phase-locking in the auditory nerve fibers (Johnson 1980; Kiang et al. 1965; Palmer and Russell 1986), that is a direct result of the manner of activation of the inner hair cells by the vibration of the basilar membrane (see Ruggero and Rich 1987 for a review). The vibration of the basilar membrane is non-linear, resulting in shifts in the phase of vibration as a function of the level of tonal stimuli (e.g. review in Robles and Ruggero 2001). Such phase shifts can also be seen in the phase-locked activity of auditory nerve fibers (e.g. Anderson et al. 1971). An implication of these level dependent phase shifts is that the output of the MSO coincidence detectors should be sensitive to interaural level differences (ILDs) as these will cause a phase shift between their inputs from the two ears. This was tested (Kuwada and Yin 1983; Yin and Kuwada 1983) by recording the ITD sensitivity of inferior colliculus (IC) neurons (the target of the ascending projection from the MSO). They saw a continuum along which the ITD sensitivity of some neurons was unchanged by ILD, while others showed marked and systematic changes in phase. The relationship between the ITD, ILD and the CF was not explored in detail in the paper, so the degree to which the observations at the periphery (e.g. the auditory nerve – Anderson et al. 1971; Robles and Ruggero 2001) match those at the IC remains unclear. Here we