Behaviour observed in multiple schedule components : implications of data for theories of contrast
Type of content
Direct Interaction theories of Hermstein and later Williams and Wixted see contrast as due to the direct effects of reinforcement in one component on responding in another component despite temporal separation in the multiple schedule. Competition theories of Staddon and later McLean and White view responding in a component as independent of reinforcement in the alternating components, but rather a result of competition among competing responses in the same component. Hence other behaviours and other reinforcers mediate the effect of reinforcement in one component on responding in the alternating component. A key area of contention between theories is when reinforcement in one component of a multiple schedule is made independent of responding, or is provided with non contingent signals. Here the traditional effect has been a decrease in responding in the non contingent condition, however reinforcement remains constant. Williams and Wixted argue that since the animal is free to engage in the same behaviours in the non-contingent procedure as in extinction, competition theories must predict contrast. However McLean claims that contrast will occur in this procedure only if behaviour in the non-contingent condition is maintained by extraneous reinforcers (thus if that behaviour is maintained by food reinforcement, or by reinforcement signals, contrast will not occur). Multiple Schedules were used in an arrangement that produces systematic variation in response rate in the varied component while firstly maintaining reinforcement levels and subsequently removing reinforcement in the varied component. Behaviour during the last three sessions of each condition was recorded using a video camera allowing analysis of behaviours associated with different reinforcement conditions. Results suggested that non contingent reinforcement, and non-contingent signals, maintain behaviours that occupy the animals time in the varied component to a similar extent as the traditional key peck response when reinforcement is contingent on responding. Subsequently these behaviours diminished with the introduction of extinction. The results support competition theorist's accounts, discrediting one vital criticism of competition theories by Williams and Wixted. The wider implications of the study are discussed, and suggestions for future research are proposed.