Longinterval Timing

The observation that timing data superimpose when plotted in relative time, rather than in absolute time, has played a formative role in the development of theories of interval timing (e.g., Gibbon 1977, 1991; Gibbon and Church, 1984). Most of these data came from short-interval timing (i.e., in the range of seconds and minutes), with relatively less information available about timing of long intervals (i.e., in the range of hours). A central question about long-interval timing is the extent to which the principles of timing that apply to short-interval timing also apply to long-interval timing.

One way to assess the timing of long intervals is to deliver a daily meal at a fixed interval after a salient cue, such as the transition from light to dark in the testing environment. In these experiments, the rats lived in operant boxes continuously for a month at a time and earned all of their food in this context. A standard 12-12 light-dark cycle was in effect (with lights on at 0800 and off at 2000). Meals started 3 or 7 h after light offset for independent groups of rats (n = 3 per group). Each 45-mg food pellet was contingent on a photobeam break after a variable interval during 3-h meals. The variable interval was adjusted to maintain food consumption between 15 and 20 g per meal. Anticipation is demonstrated by the fact that the rats started to inspect the food source before the start of the meal. Anticipatory responses as a function of time prior to the meal (absolute time, left panel) and as a function of time divided by the intervals 3 and 7 h (relative time, right panel) are plotted in Figure 3.2. Anticipatory responses began earlier (in absolute time) for the 7-h condition than for the 3-h condition (Figure 3.2, left panel). Thus, although the intermeal intervals were 24 h for both groups, the variability of the timing functions was different. The variable that controls this difference was identified by examining response rate as a function of relative time. The data superimposed when plotted in relative time (Figure 3.2, right panel). In particular, the spread of the response distribution (i.e., width) was a constant percentage of the interval between dark onset and the meal. These data suggest that (a) the rats timed with respect to dark onset (i.e., 3- and 7-h intervals), and (b) the rats timed the intervals proportionally (Crystal, 2001a).

The superposition result in Figure 3.2 (right panel) would be a standard finding if the intervals were in the short-interval range (e.g., 30 and 70 sec). The observation that the scalar property applies to intervals in the range of a few hours suggests that similar principles of timing apply to short-interval and long-interval timing. The continuity of principles is noteworthy particularly because the 3- and 7-h intervals were obtained from anticipation of daily (i.e., circadian) meals.

0 0

Post a comment