THE INFLUENCE OF POSITIVE AND NEGATIVE REINFORCEMENT ON SELECTIVE ATTENTION IN THE RAT: ASSOCIATIVE FACTORS

When rats are trained to respond during a tone-light compound stimulus to receive food they will subsequently attend to the light element of the compound. However, if they had been trained to avoid shock by responding during the compound stimulus they are much more likely to attend to the tone element. In Experiment 1 it was shown that this predisposition to attend to light when food was the reinforcer and to attend to tone with shock avoidance could be reversed with pretraining to the less relevant stimulus (i.e., tone with food and light with avoidance). That is, in the standard blocking procedure the pretrained stimulus continued to control responding even after it was compounded with a more relevant stimulus. In Experiment 2, two groups of rats were trained to respond for either food or shock avoidance during the tone-light compound stimulus; however, food or shock respectively was presented independent of responding during the absence of the compound stimulus. With this procedure, only stimulus-response factors could contribute to any differential stimulus control between the food and avoidance procedures since the same reinforcer was received during both the compound and its absence. Since no differential control was observed between the food and avoidance procedures, it was concluded that the stimulus-reinforcer relation is essential to the establishment of differential control between them. In the final experiment food was presented during the absence of the tone-light compound for rats responding during the compound to avoid shock. Rats responding during the compound to receive food had shock presented during the absence of the compound. For these procedures any nonassociative effects of either food or shock would be equivalent across groups since both groups receive both food and shock. Since the rats continued to attend to light when they had responded for food and to tone when they had responded to avoid shock, it was concluded that nonassociative factors are of little or no importance to the differential stimulus control observed across food and avoidance procedures. Therefore, it may be concluded that the original stimulus-reinforcer interaction where light was attended to with food procedures and tone was attended with avoidance procedures must be considered as an example of associative learning.