Figure 2.

Frequency tuning of latency of synaptic input and spike output. A, Example current-clamp recording of a L4 pyramidal neuron. Left, Recorded voltage traces (100 ms each) in response to tones of various frequencies and intensities. Calibration: 50 mV, 100 ms. Middle, An enlarged voltage response. Gray arrows indicate how the spike latency and input latency are determined. Right, Color maps depict the frequency-intensity tonal receptive fields of input latency (bottom) and spike latency (top) for the cell. Latencies were averaged from six trials, with spike failures excluded. B, Left, Tuning curves of input (red) and spike (black) latencies at 70 dB intensity for the cell. Whiskers indicate SD. Solid gray and pink curves are the smoothed tuning curves. The blue arrow indicates the integration time at the best frequency. Dashed vertical lines mark the frequency range for spike response. Right, Tuning curves of relative latency (top, relative to the shortest latency) and integration time (bottom) within the spiking frequency range. Integration time is defined as the difference between the input and spike latencies. C, Average tuning curves of relative latency for synaptic input and spike output. Before averaging, tuning curves of individual cells were aligned according to the BF (determined as the frequency with the shortest spike latency), which was set as zero. Whiskers indicate SE. D, Average dynamic range of input and spike latency tuning. Error bars indicate SD. *p < 0.01; **p < 0.001 (paired t test; n = 15 cells). E, Average tuning curve of integration time.