intervals, responding becomes relatively flat as the animal transitions to the steady-state FI scallop (Ferster and Skinner, 1957).

The performance of the general timing model under these conditions is illustrated in Figure 2.17. The model also begins with an inverted scallop pattern and gradually transitions to the FI scallop, as shown in Figure 2.6. The learning process shown happens over fewer trials than it does in animals, but the progression and dynamics are very similar.

2.4.3 Complex Timing Phenomena Gap Timing

The gap procedure represents one of the most popular paradigms for exploring the connection between learning and timing (see Buhusi, this volume; Fortin this volume). Based on the PI procedure, the gap procedure adds an additional type of probe trial called a gap trial. In a gap trial, the timing stimulus is turned off for a short portion of the interval and then turned back on for the rest of the trial. As in other probe trials, no reinforcement is provided during a gap trial.

The essential result of the gap procedure is that the peak response time is shifted by slightly more than the length of the gap (e.g., Roberts and Church, 1978; Roberts, 1981). This implies that the subject's internal clock is paused or stopped for the duration of the gap and then resumes when the stimulus is turned back on. In longer gaps, the peak is shifted by just less than the length of the gap plus the portion of the interval before the gap, as if the subject's internal clock had reset during the gap and started timing anew once the stimulus was represented.

This stop vs. reset pattern is very attractive to proponents of stopwatch-like models, since those are precisely the functions that a mechanical or digital stopwatch

0 0

Post a comment