FIG. 1.

A Two temporal patterns of noise bursts, Rhythm 1 and Rhythm 2. B Timing of response windows. This example shows the Hold 2 condition. The target and masker sequences were interleaved in time. The masker pattern was complementary to that of the target. The earliest detectable change from Rhythm 1 to Rhythm 2 occurred 600 ms after the onset of Rhythm 2. That time marked the beginning of a 1200-ms time window in which the cat could release the pedal to score a “hit” and receive a food reward. Releases within 1200 ms prior to the beginning of the hit window were scored as “false alarms.” Releases even earlier were scored as “early releases.” Releases after the hit window were scored as “misses”.