THE ROLE OF ASSOCIATIVE FACTORS IN AROUSAL PHENOMENA

Inverted U-shaped functions have been found to relate instrumental response strength to level of deprivation-induced or shock-induced intraorganismic arousal. Such findings have been interpreted as showing that a moderate amount of this arousal is optimal for the performance of instrumental behaviors (e.g., Malmo & Belanger, 1967). The present research investigated the notion that the effects of arousal on performance are dependent, at least in part, on learning about internal stimuli which characterize different arousal levels. Two experiments attempted to induce changes in bar pressing behavior, attributable to changes in arousal, by administering shock of various intensities immediately prior to placement in the conditioning situation. In one experiment, water deprived rats were required to solve a successive free-operant discrimination problem based only on internal cues produced by previous shock. For one group, responses were reinforced only on days in which a shock was delivered prior to training. On days without shock bar presses were nonreinforced. Another group received a reversed shock-reinforcement contingency. All rats were subsequently tested with six different intensities of shock. The results showed that both groups responded more on their rewarded than on their nonrewarded days, indicating that responding was under the discriminative control of internal shock produced cues. Test phase performance did not differ between the two groups. The second experiment examined whether or not prior training with intense shock or high water deprivation could alter the shape of the function relating shock intensity to bar press performance. It was reasoned that disruptions in performance following shifts to high arousal might be due to stimulus generalization decrement resulting from the introduction of unconditioned internal stimuli. If so, these disruptions might be reduced or eliminated through conditioning of the internal stimulus consequences of high arousal to the instrumental response. To test this notion, a group of 23 hr water deprived rats (Group S) was trained to bar press following delivery of .50 ma shock (administered in a different situation), and another group (Group HD) was trained under 71 hr water deprivation. A control group (Group C) was trained under 23 hr water deprivation in the absence of shock. All rats were then tested under 23 hr water deprivation, following .00, .16, .50 and .60 ma shocks, which were delivered in three different sequences of intensity. The results showed that an inverted U-shaped shock intensity-performance function was obtained only for the control group. In contrast, a cubic function related performance to shock intensity for Group HD, while the shock intensity-performance function for Group S was essentially flat, with level of performance below that of Group HD and C. Sequence of test shock intensity had negligible effects on performance. The results of these experiments indicate that internal cues produced by prior shock can be conditioned to instrumental responses and that the function relating response strength to shock intensity is dependent upon previous learning. Additionally, the results suggested that stimulus generalization decrement produced by the introduction of the internal stimulus consequences of intense shock was reduced or eliminated by prior conditioning of internal cues arising from high deprivation to the bar press response. This indicated that internal states produced by high deprivation and intense shock share some common stimulus properties. It was concluded that the role of learning about internal stimuli is importantly related to the behavioral effects of arousal-inducing variables