The Importance Of Interval Timing In Adaptation And Learning

It is becoming increasingly evident that interval timing is crucial for many basic forms of adaptation and learning (e.g., Staddon and Higa, 1996, 1999). One of the clearest cases comes from the field of optimal foraging, which studies the extent to which animals' foraging decisions are the direct product of natural selection (for a review, see Krebs and Kacelnik, 1984). In most cases, to make decisions that maximize fitness, an animal needs to measure its rate of food intake in one or more environments, and measuring rate requires measurement of time. Recent work has shown that European starlings are deftly sensitive to their rate of food intake and appear to record the interval of time between each prey they capture and consume (see Bateson, this volume; Bateson and Kacelnik, 1997, 1998). Even nematodes such as Caenorhabditis elegans engage in complex foraging behaviors that are temporally sensitive — suggesting that interval timing is a very basic process that can be fruitfully explored at the molecular level (e.g., Brockie et al., 2001).

Song learning in passerine birds (songbirds) has also been a central focus of ethological research, and the study of the neurobiology of song learning, song production, and song perception has stimulated many seminal contributions to our understanding of brain-behavior relationships and how learning influences these patterns. The study of the temporal hierarchical control of singing in birds is clearly an area ripe for the application of principles learned from the psychophysical analysis of timing and time perception (see Hahnloser et al., 2002; Hills, this volume; Fee et al., 2002; MacDonald and Meck, this volume; Yu and Margoliash, 1996).

Associative learning or conditioning is another widespread and elemental form of animal behavior for which computational accounts of interval timing are becoming enormously influential (see Arcediano and Miller, 2002; Bugmann, 1998; Gallistel, 1990; Gallistel and Gibbon, 2000, 2001; Hopson, 1999, this volume; Migliore et al., 2001). For example, in classical conditioning it is well established that the efficiency of learning about a conditioned stimulus (CS) is affected by the time interval between the CS and the unconditioned stimulus (US): in general terms it is found that the shorter the CS-US interval, the faster and better the conditioning that occurs (Balsam, 1984). In fact, it is not the absolute duration of the CS-US interval that is important, as was initially thought, but the ratio of the CS-US interval to the interval between successive US (e.g., Balsam, 1984; Gibbon et al., 1977; Gibbon and Balsam, 1981; Jenkins, 1984; Meck, 1985). The observation that temporal factors of this sort dramatically impact the acquisition of conditioned responding suggests that learning about time may be a necessary condition for associative learning (e.g., Balsam et al., 2002; Roberts and Holder, 1984).

0 0

Post a comment