Date of Award:


Document Type:


Degree Name:

Master of Science (MS)




Edward K. Crossman


In mixed fixed-ratio schedules, reinforcement is delivered according to two or more fixed-ratio components, which may be programmed in simple alternation or according to a random sequence. If the components simply alternate, the behavior is similar to that obtained in a multiple schedule. Pauses are long before long fixed ratios and short before short fixed ratios. If a random sequence is used, responding is characterized by short pauses after reinforcement independent of the size of the ratio and primes occur at the beginning of many of the fixed ratios. The purpose of the present study was to investigate the possibility of other patterns of responding being controlled by component sequences ether than simple or random alternation in mixed fixed-ratio schedules.

Two pigeons were trained to key peck on a fixed-ratio 100 (FR 100) schedule. Following this training, the schedule was based on a superimposed fixed-interval contingency which specified that an FR 10 would occur after the completion of the first FR 100 following a fixed period, timed from the last FR 10. Thus, the schedule was a mixed FR 100 FR 10 with FR 10 frequency controlled by the superimposed fixed-interval contingency. The value of the FI was varied for both birds. To control for the possibility that the pattern of responding was related to priming, two additional pigeons were trained to respond on comparable multiple schedules in which primes were absent.

The mixed schedule shortened the pre-FR 100 pause. Analysis of the sequence of pauses following an FR 10 showed that the pause following an FR 10 was long relative to other pauses in the sequence which were short and approximately equal to the pre-FR 10 pause. Primes, a run of approximately ten responses at the beginning of a ratio followed by a pause, were relatively infrequent in the first FR 100 after an FR 10, but were more frequent in other positions in the sequence. In the multiple schedule, all pre-FR 100 pauses were long and approximately ·equal in length, but the pre-FR 10 pauses were short.

These results demonstrate that the pattern of responding obtained in the mixed schedule was related to priming. First, short pauses preceded FR 100s which contained primes in the mixed schedule. Primes did not occur in the multiple schedule, and the pre-FR 100 pauses were long. Second, primes were frequent only in those ordinal positions in the mixed schedule which were preceded by short pauses. Third, short pauses and primes developed simultaneously.

These three effects suggest a new account of priming as a unit of behavior. This unit consists of a short pause followed by a run of ten responses. This run of ten responses is intermittently reinforced by food delivery when the FR 10 occurs. The pause after this run, which occurs if the schedule is FR 100, is caused by the S-delta condition produced by the information that the schedule is not FR 10.

The pattern of responding in the mixed schedule was also related to the sequential nature of the schedule. Since FR 100s always followed FR 10s, FR 10s were followed by long pauses and few primes. FR 100s, on the other hand, were occasionally followed by FR 10s. Thus, the completion of an FR 100 was a discriminative stimulus for "FR 10-like behavior".



Included in

Psychology Commons