Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Dr. Jack Michael

Second Advisor

Dr. Galen Alessi

Third Advisor

Kay Malott

Fourth Advisor

Dennis Simpson


The latency between discriminative stimulus onset and initiation of a corresponding ratio requirement (RR) was studied in pigeons using a two-key discrete-trial procedure with three-component multiple schedules. The two-key procedure allowed measurement of response-reinforcer relations on a "constant" key, and stimulus-reinforcer relations on a "stimulus" key. The first experiment showed regular between-session effects as equivalent fixed-ratio (FR) components were raised over several phases and then lowered. A direct relation was observed between latency and RR for each of ten subjects. Initial stimulus-key pecking showed several trends, the most prevalent being an inverse relation to RR.

Experiment 2 demonstrated within-session separations and reversals of median latency for eight of ten subjects as different FR values were applied to each of the three schedule components. A direct relation between latency and RR was again noted. Initial stimulus-key pecking showed an inverse relation to RR, with a high proportion of such responses occurring in the smallest FR component. These effects were replicated in two of the previously successful subjects when equivalent FRs were used with differing percentage reinforcement in each component.

Experiment 3 failed to duplicate the within-session latency separations of the former experiment when variable-ratio (VR) schedules were substituted for FR components with three of Experiment 2's successful subjects. Each schedule included one element of continuous reinforcement in the distribution of RRs comprising its VR list. Latency was found to mainly be controlled by this lowest RR, with minor differences appearing for each of the VR averages. Initial stimulus-key pecking maintained the same inverse relation to RR as in Experiment 2. These effects were replicated in one of two other subjects when equivalent VRs were used with differing percentage reinforcement in each component.

These results were analyzed in terms of the pausing typically found under ratio-based reinforcement, and were further discussed according to their implications for the determination of response latencies in operant relations. Recommendations were made for continued investigation of these effects, and for a clearer terminology regarding temporal characteristics of discriminated operant performance.

Access Setting

Dissertation-Open Access