Studies of performance on DRL and FI schedules of reinforcement have shown that organisms can adjust the distribution of their responses in time in
accordance with the constant temporal characteristics of these schedules. In the present research an attempt was made to find out whether such adjustment
also occurs when the temporal distribution of the presentation of reinforcements undergoes orderly change.
In Part I the relevant literature relating to performance on DRL and FI schedules is reviewed.
In Part II a series of experiments is described in which the behavioural effects of progressive-interval (PI) schedules were investigated.
In these schedules only the first response was reinforced in each of consecutively increasing time-intervals; the durations of the intervals increased
by (i) arithmetic progression, or (ii) geometric progression. It was found that behaviour adjusted to the changing conditions of these schedules. The
evidence was as follows :
1. The number of responses emitted in each interval of a schedule was relatively constant;
2. This number was less variable under geometric PI schedules than it was under arithmetic PI schedules;
3. The durations of the post-reinforcement pauses gradually increased in the course of the progression of a schedule;
4. The difference between the duration of an interval and the post-reinforcement pause in that interval became greater in the latter part of a schedule ;
5. This difference was less under geometric PI than under arithmetic PI;
6. These relations were maintained in long sessions some of which lasted 12 hours;
7. There was some evidence that prior experience on PI schedules resulted in improved accuracy of performance on DRL.
It was suggested that the practice of referring to performance on temporally based schedules as temporal discrimination is inappropriate since
discrimination implies the presence of a stimulus to be discriminated, whether external or internal. Performance on PI schedules could not be based on
such stimuli unless a stimulus of different duration was postulated for every interval of the schedule.Temporal orientation of behaviour was proposed as an
alternative term, reserving temporal discrimination for those situations in which the duration of an explicit stimulus is discriminated.