Hungry rats received food following lever-press durations exceeding a minimum value, which ranged from 0 to 6.4 sec. When no intertrial intervals separated successive presses, modal press durations remained at very short values as the minimum value required for food was increased. This was particularly true immediately after a food presentation. When an 8-sec intertrial interval followed each lever release, modal press durations were always at or beyond the minimum value required for food, and outcome of the preceding press had no effect on press duration. Possible reasons for the effects of intertrial intervals included punishment of short presses, increased delay of reinforcement of short presses, and reduced density of reinforcement. In addition, functions relating discrete-trials lever-press duration to minimum duration required for food were found to be qualitatively and quantitatively similar to the power functions recently proposed by Catania (1970) for interresponse time and response latency. This similarity was taken as support for a general psychophysical law of temporal judgments.

Reinforcement rate and differential reinforcement of IRTs were independently manipulated to assess their relative contribution to the control of interresponse times (IRTs). Modified percentile reinforcement schedules (Platt, 1973) allowed control of reinforcement rate while longest or shortest IRTs were selectively reinforced. In the absence of differential IRT reinforcement, mean IRT decreased with increasing reinforcement rate. Compared to this small effect of reinforcement rate, reinforcement of long IRTs produced large changes in mean IRT at constant reinforcement rates. No interaction of reinforcement rate and IRT reinforcement was detected. The demonstration of large IRT changes in the absence of reinforcement-rate changes indicates the precedence of IRT reinforcement over molar reinforcement-rate correlations in the determination of IRTs in these procedures.

Rats received food following lever-press durations between t and t + t' sec where t was 2, 4, or 8 sec and t' was 0.25t, 0.50t, or 1.00t sec. Modal press durations were greater than t but less than t + t' in all cases. Distributions of press durations were lower and broader for larger values of t. Lower t'/t ratios produced lower median press durations and relatively narrower press-duration distributions. Median press duration was a power function of t within a t'/t ratio condition, corresponding to previous results for latency, interresponse time, and response durations.The differentiation of temporal response properties has been used as a psychophysical method for the study of temporal discrimination (Catania, 1970;Ferraro and Grilly, 1970;Platt, Kuch, and Bitgood, 1973). Approached in this manner, response differentiation is a special case of stimulus discrimination in which the stimuli being discriminated are assumed to be correlated with a parameter of the response. For example, an organism that terminates a response t sec after the response is initiated might be thought of as responding to a temporal stimulus correlated with response duration. If an experimenter requires a minimum response duration of t sec for the delivery of reinforcement, and the organism comes to emit response durations approximating t, the organism can be said to be responding differentially with respect to this temporal stimulus. Functions relating emitted values of temporal properties of responding such as duration, latency, and interresponse time (IRT) to the reinforcement criteria resemble psychophysical functions relating human time estimations to the experimenter's standard stimulus (Catania, 1970 functions from both sources are linear with similar slopes and intercepts when plotted in logarithmic coordinates. This convergence of response and stimulus functions supports the view of response differentiation as a case of stimulus discrimination.Systematic studies of response duration differentiation are few. Platt et al. (1973) studied rats' lever-press durations with reinforcement contingent on durations larger than t, where t varied from 0.4 to 6.4 sec. They found that median press durations approximated t under conditions in which intertrial intervals (ITIs) occurred after each lever release. In a freeoperant procedure without ITIs, control of press duration by t was much poorer. Ferraro and Grilly (1970) used t values from 0.1 to 1.6 sec in a free-operant procedure, reinforcing durations between t and t + 0.2 sec. They found that mean durations approximated t, even though reinforcement probability decreased precipitously as t increased.The present study employed a duration differentiation procedure in which press durations longer than t, but shorter than t + t', were followed by reinforcement. A range of t and t' values was used to explore the effects of these variables on press duration. This procedure was used to obtain a more restricted specification of the response durations reinforced than is provided by a schedule in whi...

No abstract