Figure 3.

Glutamate receptor contributions varied with first-spike latency. Each data point shows the percentage of reduction in spikes of one unit resulting from drug application as a function of the first-spike latency of the unit. A, NBQX reduced spikes substantially for all units with latencies <12 ms but never for units with longer latencies. B, CPP always reduced spikes substantially for units with latencies ≥12 ms but had variable effects on units with shorter first-spike latencies. The horizontal dashed line indicates the criterion for substantial drug effect based on 50% spike reduction.