Three examples of different SU and LFP pairs co-recorded off the same electrode in response to pup calls and tones demonstrate the quality of our recordings.
(A–C top row) Waveforms of all spikes (left) and inter-spike interval distributions (right) for pup call stimulation. Note the absence of refractory spikes within 1ms.
(A–C middle row) Responses to pup calls. Trial-by-trial SU spike raster (upper) and 10 ms-binned PSTH (middle) for call-excited (A and B) and -inhibited SUs. The mean call-evoked LFP responses (bottom) at the same site are also shown. Stimuli presented during the interval denoted by the horizontal black bar.
(A–C bottom row) Tonal tuning curves for SU spike count (solid black line, left axis) and the amplitude of the negative LFP deflection (dotted black line, right axis. Correlation coefficients (CC) between the full frequency tuning curves for SU and LFP varied from −0.58 to 0.95 over our tone-excited population of SUs, suggesting these neural signals do not reflect the same processes.

The timing of when the call-evoked LFP reached specific phases was significantly different between mothers and virgins for sites around call-inhibited but not -excited SUs.
(A1 and B1 box plots) Group comparison of the times at which the Hilbert phase of a site’s average call-evoked LFP reached values of π and 1.16π for call-excited and -inhibited SU sites. These times were not different between mothers and virgins for call-excited sites (Mann-Whitney, N_mothers=16, N_virgins=10, π: U=56, p>0.05, 1.16π: U=52, p>0.05, 2-tailed), but were for phases between π and 1.25π at call-inhibited sites (Mann-Whitney, N_mothers=14, N_virgins=11, π: U=45, p>0.05, 1.16π: U=30, p<0.05, 2-tailed). These phases corresponded to the initial rise from the minimum of the LFP. Note: Sample sizes for LFPs may be less than or equal to that of SUs if more than one SU was isolated at a site.
(A2 and B2) Comparison of the population-averaged LFP for all call-excited or -inhibited sites. The location of π (circle) and 1.25π (square) Hilbert phase values are marked. Notice that the timing difference in the Hilbert phase seen in (B1) manifests as a sizeable shift in the timing of the valley between mothers and virgins.
(C) Variability in the LFP shape depicted as a time-dependent probability histogram of trial-by-trial Hilbert phases. Whiter (Darker) colors indicate higher (lower) probabilities for a specific phase at a specific time.

Significant differences in LFP phase precision mainly arose from call-inhibited sites with LFP BF<50 kHz. SUs from these same lateral band sites had greater call-evoked inhibition in mothers than in virgins.
(A1 and A2) Population-averaged phase precision trajectories for LFPs at call-excited SU sites, grouped by LFP BF. No significant differences in trajectories were found between mothers and virgins for either lateral (A1) or high-ultrasonic (A2) band sites. Similarly, differences in the normalized, integrated SU firing (see Insets of Fig. 2A3 and B3) did not reach significance for either the lateral (t-test, t=.98, df=10, p>0.05, 2-tailed) or the high-ultrasonic band (Mann-Whitney, U=16, N_mothers=9, N_virgins=5, p>0.05, 2-tailed).
(B1 and B2) Population-averaged phase precision trajectories for LFPs at call-inhibited SU sites, grouped by LFP BF. The trajectory for mothers was significantly higher than virgins at lateral (B1) but not high-ultrasonic (B2) band sites. Similarly, differences in the normalized, integrated SU firing were significant for the lateral (t-test, t=2.3, df=13, p<0.05, 2-tailed) but not the high-ultrasonic band (t-test, t=1.7, df=11, p>0.05, 2-tailed; N_mothers=9, N_virgins=4).
(A3 and B3) Group comparison of call-evoked LFP phase precision rise times for lateral (hatched) and high-ultrasonic (solid) bands. No differences between mothers and virgins were found at call-excited sites for both the lateral (Mann-Whitney, U=12, N_mothers=7, N_virgins=5, p>0.05, 2-tailed) and high-ultrasonic bands (t-test, t=1.6, df=12, p>0.05, 2-tailed). However, there was a significant difference at call-inhibited sites in the lateral (t-test, t=2.3, df=12, p<0.05, 2-tailed), but not high-ultrasonic band (t-test, t=0.05, df=9, p>0.05, 2-tailed). When comparing across frequency ranges within the same group, there was no significant differences between the LFP phase precision rise times for mothers for excited (Mann-Whitney, U=12, N_BF>50kHz=9, N_BF<50kHz=7, p>0.05, 2-tailed) or inhibited (t-test, t=1.7, df=12, p>0.05, 2-tailed). However, the virgins showed differences between LFP sites with BF<50 kHz and BF>50 kHz for both excited (t-test, t=2.5, df=8, p<0.05, 2-tailed) and inhibited (t-test, t=3.0, df=9, p<0.05, 2-tailed) SUs.

CN: cochlear nucleus; SOC: superior olivary complex nuclei; IC: inferior colliculus; MGB: medial geniculate body. See text for details.

Pure SU inhibition, but not excitation, evoked by calls was systematically earlier, longer, and stronger in mothers compared to virgins.
(A1 and B1) Examples of raw (gray bars, 5 ms bins) and Gaussian-smoothed (black line) PSTH’s for a call-excited and -inhibited SU. The stars indicate the half-max or half-min values for calculating the excitation or inhibition latency and duration.
(A2 and B2 scatter plot) Call-excited (upward triangles) and -inhibited (downward triangles) latencies and durations for mothers (black in this and other figures) and virgins (gray in this and other figures). Stars refer to SUs depicted in A1 and B1. Excited N_mothers=16, N_virgins=10. Inhibited N_mothers=17, N_virgins=11.
(A2 and B2 bottom box plots) Group comparison of call-evoked excitatory and inhibitory latencies. For this and later figures, box plots show lines at the lower quartile, median and upper quartile, and whiskers extending out to extreme data points that are not outliers. The difference between mothers and virgins for call-excited SUs was not significant (Mann-Whitney, U=80, N_mothers=16, N_virgins=10, p>0.05, 2-tailed), but was significant for call-inhibited SUs (t-test, t=2.9, df=26, p<0.01). For this and later figures, n.s./asterisk indicates a non-significant/significant comparison.
(A2 and B2 right box plots) Group comparison of call-evoked excitatory and inhibitory durations. Mothers and virgins were not significantly different for call-excited SUs (t-test, t=.54, df=24, p>0.05), but were for call-inhibited SUs (t-test, t=4.7, df=26, p<0.0001).
(A3 and B3) Population-averaged time course of spike rate normalized by the spontaneous rate, for call-excited and -inhibited SUs. In this and later figures, the gray rectangles marked by asterisks denote regions where significant differences (anovan, followed by protected multiple t-test comparisons, p<0.05) were found between traces, and the error bars represent standard errors. Significant differences occurred only for call-inhibited responses. Dotted lines represent the baseline spontaneous rate. Insets show the normalized rate derived by integrating the SU spike count over the stimulus period and dividing by the spontaneous level. Call-excited SUs were not different between mothers and virgins (Mann-Whitney, U=51, N_mothers=16, N_virgins=10, p>0.05, 2-tailed), but call-inhibited SUs were significantly different (t-test, t=2.8, df=26, p<0.01, 2-tailed).

Mothers have an earlier, longer, and greater LFP phase precision at call-inhibited but not-excited SU sites.
(A1 and B1) Example LFP phase precision trajectories and their rise times and durations (stars) for both a call-excited and -inhibited site in a mother and virgin. Phase precision values lying above the dashed line are significant (see Experimental Procedures). (A2 and B2 scatter plot) Rise times and durations of the LFP phase precision at call-excited (upward triangles) and -inhibited SU sites (downward triangles) for mothers and virgins. Excited N_mothers=16, N_virgins=10. Inhibited N_mothers=14, N_virgins=11.
(A2 and B2 bottom box plots) Group comparison of call-evoked LFP phase precision rise times. Differences between mothers and virgins were not significant at call-excited SU sites (Mann-Whitney, U=70, N_mothers=16, N_virgins=10, p>0.05, 2-tailed), but were significant at call-inhibited SU sites (t-test, t=2.0, df=23, p=0.05, 2-tailed).
(A2 and B2 right box plots) Group comparison of call-evoked LFP phase precision durations. Differences between mothers and virgins were not significant at call-excited SU sites (Mann-Whitney, U=54, N_mothers=16, N_virgins=10, p>0.05, 2-tailed), but were significant at call-inhibited SU sites (t-test, t=2.3, df=23, p<0.05, 2-tailed).
(A3 and B3) Population-averaged phase precision trajectories. LFPs from call-excited SU sites did not show a significant difference between mothers and virgins. LFPs from call-inhibited sites in mothers had a significantly higher phase precision trajectory than virgins beginning near sound onset until more than 100 ms after sound offset.

Differences between mothers and virgins were particularly enhanced for tone frequencies falling in the natural pup call range.
(A1 and B1) Comparison of 60–80 kHz (upper panels) and 30–40 kHz (lower panels) tone-evoked LFP phase precision trajectories, averaged only across lateral band LFPs. The magnitude of phase precisions was higher for the latter frequencies since lateral band sites should be more responsive to them. No significant differences between mothers and virgins were found for call-excited sites for either tonal stimulus. However, mothers and virgins did differ significantly at call-inhibited sites for both tonal stimuli. Excited N_mothers=7, N_virgins=5. Inhibited N_mothers=7, N_virgins=6.
(A2 and B2) Group comparison of tone-evoked LFP phase precision rise times. Bars depict the mean and standard error of the rise times for each group. No significant differences between mothers and virgins were found for call-excited sites, regardless of the frequency of tonal stimulation (60–80 kHz Mann-Whitney, U=13, N_mothers=7, N_virgins=5, p>0.05, 2-tailed; 30–40 kHz: Mann-Whitney, U=14, N_mothers=7, N_virgins=5, p>0.05, 2-tailed). The same was true for call-inhibited sites (60–80 kHz: Mann-Whitney, U=14, N_mothers=7, N_virgins=6, p>0.05, 2-tailed; 30–40 kHz: t-test, t=0.78, df=11, p>0.05, 2-tailed).
(C) Frequency-dependence of tone-evoked LFP phase precision enhancement in mothers compared to virgins for call-excited (blue) and -inhibited (red) sites. Each point pooled trials from 5 logarithmically-spaced tone frequencies centered on that point. Population-averaged phase precision trajectories were computed separately for call-excited and -inhibited SU sites from each animal group. The average difference (solid lines) between trajectories (mother – virgin) over the duration of the tone quantified the precision enhancement relative to virgins. Shaded bands represent 95% confidence intervals computed by bootstrapping across sites. We found a significant increase only for LFPs at call-inhibited SU sites, with the greatest differences for frequencies falling in the natural pup call range (dashed vertical lines). A smaller but still significant difference was also apparent for tone frequencies above ~30 kHz.