No abstract

Adult human Ss, reinforced for key pressing on a DRL 5-sec schedule of reinforcement, regularly verbalized hypotheses about the behavioral requirements necessary for obtaining reinforcement. High frequencies of overt collateral behavior accompanied response-based hypotheses and low or absent rates of overt collateral behavior accompanied time-based hypotheses. The findings support the conclusion that overt collateral responding can be functionally equivalent to mediative counting during temporally spaced responding.Previous studies examining DRL performance in humans (e.g., Bruner & Revusky, 1961;Laties & Weiss, 1962;Randolph, 1965) have reported regular occurrences of collateral behavior between successively reinforced responses. Recently, Stein & Landis (1973) documented the mediating role human collateral behavior can serve in regulating temporally discriminated DRL performance. In each of these studies, collateral behavior was expressed either overtly, as a chain of motor responses, or covertly, in the form of verbal mediators which emphasized either a time-or response-based self-instruction about the contingency of reinforcement. The present experiment examined the interrrelationships between overt and covert forms of collateral behavior and DRL performance as they occurred at regular intervals throughout training. METHOD SubjectsThe Ss were 10 female introductory psychology students, who received class credit for participation. None had had previous experience with DRL schedules from either coursework or other experiments. ApparatusThe 5s were seated in a room and fitted with a headphone, over which continuous white noise at 60 dB (re .0002 dynes/cm 2 ) was presented to mask extraneous sounds. In front of the S was a chassis containing four adjacent telegraph keys, placed 9.5 em apart, each requiring a downward force of 438 g to be electrically recorded as a response. Directly above each key was a separate light which could be illuminated remotely. A four-digit counter was mounted in the middle of the chassis to indicate the cumulative number of reinforcements (points). A series of relays, timers, and recorders located in an adjacent room was employed to program events and record responses. A one-way mirror between the programming and S rooms allowed continuous visual monitoring of S.Each keypress activated a separate channel of an event recorder, \vith another channel recording reinforcements. The second key from S's right (Key 3) was programmed to produce reinforcement on a DRL 5-sec schedule. Responses on the *Sponsored by F. Robert Brush, Syracuse University, who takes full editorial responsibility for it.tReprints are available from

No abstract

No abstract