Humans and other animals process temporal information as if they use an internal stopwatch that can be stopped and reset, and whose speed of ticking is adjustable (Church, 1978). Support for this idea comes from data in bees, fish, turtles, birds, rodents, monkeys, and human infants and adults (e.g., Bateson and Kacelnik, 1997, 1998; Brannon et al., 2002; Brodbeck et al., 1998; Gallistel, 1990; Lejeune and Wearden, 1991; Matell and Meck, 2000; Paule et al., 1999; Richelle and Lejeune, 1980; Talton et al., 1999). The internal clock concept was first introduced in the seminal work of Treisman (1963). According to Treisman (1963), pulses emitted at regular intervals by a pacemaker are stored in an accumulator whose content represents the current subjective time. The rate of the pacemaker was proposed to be related to participant's arousal, and the response of the participant was proposed to be controlled by a comparator mechanism. Although the structure of Treisman's (1963) internal clock can still be recognized in current models of interval timing as applied to a variety of species, these models differ in regard to attentional processing. This is not surprising, considering the differences in the behavioral protocols used in the fields of human cognition and animal behavior. For example, data obtained from human participants performing in dual-task paradigms show that the use of a second task performed in parallel with a temporal task results in temporal underestimation. This effect has been observed with various nontemporal tasks, such as perceptual, mental arithmetic, or motor tracking (e.g., Brown, 1985, 1995, 1997; Hicks et al., 1976, 1977; Macar et al., 1994; Zakay, 1989; Zakay et al., 1983). It is inferred from these results that increasing the demand in processing results in fewer temporal pulses to be accumulated, possibly due to the sharing of attention between the internal clock and other processes (see Fortin, this volume; Lustig, this volume; Penney, this volume). Cognitive models of interval timing developed to explain the effects of nontemporal parameters assume that (a) temporal information is obtained from both a timer (temporal processor) and other sources (general processor), and (b) attention is shared between these processors, a mechanism that we will refer to as attention sharing (e.g., Block and Zakay, 1996; Fortin and Masse, 2000; Thomas and Weaver, 1975; Zakay, 1989, 2000). This view predicts that manipulations that would increase the resources allocated to general processing would result in the time processor losing resources, and consequently losing its ability to accumulate pulses and time accurately.

In contrast to human studies of interval timing, studies of interval timing in rats and pigeons are typically conducted in single-task paradigms designed to explore the specifics of the time processor, but less apt to address the possible relation with general processing. The use of such paradigms promotes psychological theories that attempt to explain the effect of all manipulations as changes in the time processor, rather than possible interactions with the other processes. Here we review behavioral and pharmacological data supporting the dissociation of the pacemaker of an internal clock from attention sharing between the timer and other processes. The dissociation of these processes at the behavioral level is critical, because both processes depend on intact dopaminergic function (Buhusi and Meck, 2002; Malapani and Rakitin, this volume).

0 0

Post a comment