The perception of empty and filled time intervals by rats.

In Experiment 1, rats were trained in a within-subjects design to discriminate durations of a filled interval, and durations of an empty interval (an unfilled interval marked at the beginning and end by a 500 ms tone). Training and psychophysical testing was conducted with three sets of anchor durations. Rats made more long responses for filled than for empty intervals at signal durations greater than the geometric mean. In Experiment 2, Group Same was trained similarly to the rats in Experiment 1 with the ambient conditions (house light illumination) remaining the same during the intertrial interval and the empty intervals. Group Different was trained with the house light turned off during empty and filled intervals. The similarity of ambient conditions during the intertrial interval and the empty intervals did not significantly affect timing. Filled intervals were timed more precisely and they were perceived as longer than empty intervals of the same duration. The psychophysical functions superimposed across anchor duration sets. These results are the first clear evidence of a filled interval illusion in rats, and they suggest that this difference may reflect a clock rate effect (greater for filled intervals) rather than a switch latency effect (slower for empty intervals).