, 2011). This role is in agreement with the strong increase in spontaneous

activity, clear broadening of ITD tuning and strongly reduced effect of ITD on spike rate observed upon application of the glycine receptor antagonist strychnine (Brand et al., 2002; Pecka et al., 2008) and the relatively slow kinetics of glycinergic synaptic potentials compared to the glutamatergic synaptic potentials (Magnusson et al., 2005). Apart from the lack of evidence for a role of well-timed inhibition, we also did not find support for the two other models that propose that MSO neurons contribute to the creation of internal delays. The suggestions that interaural asymmetries in synaptic potentials (Jercog et al., 2010) or cellular morphology (Zhou et al., 2005)

may contribute to ITD tuning of MSO cells are contradicted by our observation that the slopes of subthreshold inputs were similar for both ears (Figure 7A), in agreement with recent slice studies (Fischl selleck inhibitor et al., 2012; Roberts et al., 2013), and we obtained a similar result for the EPSP-AP latencies (Figure 7B). The interaural symmetry of EPSP-AP latencies agrees with the observation that in the gerbil MSO axons typically emerge directly from the soma (Scott et al., 2005). Our data therefore indicate that ITD tuning depends critically on the exact timing of the excitatory inputs to the MSO neurons, and that the MSO neuron find more itself does not make a large contribution to the internal delay. ITD tuning was complex. Two features were remarkable. First, at low sound frequencies we observed multiple preferred latencies in the responses for both ears.

Most likely, this is inherited from the SBCs. Spike timing-dependent plasticity has been suggested as a possible mechanism for the coincidence of these inputs (Gerstner et al., 1996), and our results suggest that, if it is, it can work for multiple preferred latencies, indicating a hitherto unknown complexity to the tuning of the MSO neurons. It should be noted that these multiple latencies were typically obtained at low frequencies and high intensities, so their contribution to natural stimuli Calpain remains to be established. Behaviorally, localization is poorer for pure tones than for more “natural,” wideband sounds. Future work using wideband stimulation is required to test how our findings generalize to a wider range of stimuli. A second property that added to the complexity of the tuning was that a comparison of the inputs from both ears indicated that ITD tuning was frequency dependent. This observation by itself argues against the original Jeffress model (Jeffress, 1948), in which a delay line was the only source for ITD tuning. Since we did not observe any evidence for a contribution of the MSO neurons themselves to the delay line, this is compatible with the idea that cochlear tuning disparities contribute to the creation of internal delays (Day and Semple, 2011; Joris et al., 2006).