In the present experiment, an attempt was made to extend the base of evidence for the assumption of the behavioral theory of timing that pacemaker rate is determined by reinforcement rate. Pigeons discriminated the first half from the second half of a 50-sec trial in a free-operant psychophysical procedure. Left-key responding was reinforced at variable intervals during the first 25 sec, and right-key responding was reinforced at variable intervals during the second 25 sec. The rate of "extraneous" reinforcers delivered at variable intervals following responses to a center key was manipulated independently of performance in the temporal discrimination. Quantitative estimates of pacemaker rate were not directly proportional to extraneous rate of reinforcement, whether extraneous reinforcers were available during the intertrial interval, the entire session, or the trial only. Instead, estimates of pacemaker rate were inversely related to the rate of extraneous reinforcement, which suggests that pacemaker rate is determined by the ratio of the rate of reinforcement for the timing response relative to other sources of reinforcement.The behavioral theory of timing (BeT) provides a successful account of animal timing (Killeen, 1991;Killeen & Fetterman, 1988. Several propositions are fundamental to BeT. First, individuals possess an internal pacemaker. Second, pacemaker pulses cause transitions between hypothetical sequential "adjunctive states." Third, Killeen and Fetterman (1988, 1993) have suggested that adjunctive behavior becomes correlated with the sequential states. In particular, the adjunctive behavior correlated with state n serves as a discriminative stimulus that occasions the timing response required at time t. The adjunctive behavior gains utility as a discriminative stimulus by virtue of the selective action of reinforcement that follows the timing response. Mathematically, the model predicts the time at which state n occurs. Behaviorally, the timing response is occasioned by the adjunctive behavior that occurs at about time t, that is, when in state n. The time at which state n is reached depends on the pacemaker period. Thus, timing consists of a conditional discrimination based on the adjunctive behavior. For example, if an individual is in a state correlated with a "long" duration as opposed to some shorter duration and is asked to make a temporal -Accepted by previous editor, Vincent M. LoLordo discrimination, it would be more likely to respond "long" ifit is engaged in behavior previously correlated with the "long" state than if it is engaged in behavior that is not well correlated with being in the "long" state.BeT assumes that pacemaker pulses cause transitions between hypothetical sequential states according to the Poisson process given by p N(t) = n = .
n! P[N(t)= n] gives the probability that by time t the number of pulses registered is n. The average pacemaker period is r, and pacemaker rate is 1/t:A major assumption of BeT is that pacemaker rate varies as a positive linear function of rein...