Timing inter-reward intervals

From minutes to days-The ability of sows (Sus scrofa) to estimate time intervals

Time estimation helps allocating time to different tasks and to plan behavioural sequences. It may also be relevant to animal welfare if it enables animals assessing the duration of a negative situation. Here, we investigated the ability of dry sows to estimate short and long time periods. We used a variant of the peak-interval procedure and the choice between 2 resources of different quality and replenishment rates to assess time periods in the order of minutes and days, respectively. In the minute-experiment, the sows were trained to expect an interruption while feeding at the end of an interval. Heart rate and heart rate variability slightly and continuously increased and decreased, respectively, towards the end of that interval. In the day-experiment, lasting about 60days, the sows were increasingly more likely to open the door to a high food reward on the correct day when this food reward was presented every fifth day. We conclude that the sows learnt to estimate time intervals of 5days after lengthy training but did not accurately learn intervals in the range of minutes. Therefore, they might re-visit replenishing resources after several days, but may not base short-term decisions solely on the passing of time.

Scalar properties in animal timing: conformity and violations

The article reviews data from animal subjects on a range of timing tasks (including fixed-interval and temporal differentiation schedules, stimulus timing, aversive conditioning, and Pavlovian methods) with respect to conformity to the two scalar properties of timing behaviour: mean accuracy and scalar (Weberian) variance. Systematic deviations were found in data from temporal differentiation schedules, timing of very short (<100 ms) or very long (>100 s) durations, effects of "task difficulty", and some special cases where circadian and interval timing seemed to interact, or where some specific durations seemed to be timed more precisely than others. Theoretical reconciliation of some of these deviations with underlying scalar timing can be achieved, but a number of problematical cases remain unexplained.

Discrimination of what, when, and where is not based on time of day

Rats (n = 6) visited four baited locations (randomly chosen on each trial; study phase), one of which was randomly selected to provide chocolate. After short (1-h) or long (25-h) retention intervals (RIs), eight locations were available, and the four locations not available in the study phase provided food (test phase); the chocolate location also provided food after long RIs. More visits to the chocolate location occurred after long RIs than after short RIs. Next, chocolate was paired with LiCl during the long RI (i.e., after encoding the chocolate location). Fewer revisits to the chocolate location occurred after LiCl than in previous testing with the long RI. The rats demonstrated complete transfer when grape replaced chocolate after LiCl-chocolate pairing. The discrimination of what, when, and where could not be based on adopting different revisit strategies at different times of testing.

The role of temporal generalization in a temporal discrimination task

Two experiments trained rats to discriminate two or three stimulus durations using a temporal discrimination task. A standard peak shift effect was observed when training was administered with short versus long signals in Experiment 1. Both discrimination accuracy scores and the generalization gradients revealed that shorter intervals were discriminated more accurately, which may be due to the scalar property of timing. In Experiment 2, three signals (short, medium, and long) were associated with three different responses, or two of the intervals were associated with one response (short and long or short and medium) and the other interval with a different response. Here, the short/medium versus long discrimination was learned most readily of the three tasks. The results of both experiments indicated a strong contribution of learning of individual durations combined with scalar generalization gradients, but Experiment 2 indicated that categorical encoding of durations may have also been operating.

Timescale invariance and Weber's law in choice

Pigeons were exposed to concurrent schedules for which reinforcement was alternately available at different times for each of two choices. In Experiment 1 (in which reinforcement times progressed arithmetically), overall, but not relative, response rate was timescale invariant. In Experiment 2 (in which reinforcement times progressed geometrically and were more spaced out), there was temporal control at all reinforcement times, but the amplitude of left-right response alternation decreased as time in the trial increased. These results indicate that the temporal regulation of both overall and relative response rates conforms to Weber's law although relative rate is heavily influenced by processes other than timing. It also adds support to the idea that overall and relative response rate reflects the operation of two independent processes.