ラッテの白黒弁別学習とその逆転学習に及ぼすchlordiazepoxideの効果

Two groups of male albino rats were run 20 trials per day on a black-white discrimination, motivated by electric shock (Fig. 1) to a learning criterion of 18 correct responses within a day in which the last 10 responses must be correct. One group was injected with 2ml/kg of physiological saline and the other with 20mg/kg of chlordiazepoxide (CDP) (2ml/kg of a 10mg/ml solution) both intraperitoneally 30min prior to the first daily trial.<br>Two days after reaching criterion, the rats were given reversal training to the same criterion or a maximum of 200 trials: half the animals in each group were changed in drug state and the rest were not changed, and thus 4 groups (SAL-SAL, SAL-CDP: CDP-SAL, CDP-CDP) were run as in acquisition except the stimulus incentives were reversed.<br>Mean trials to criterion in acquisition and reversal are shown in Table 1 for each subgroup. Acquisition was slightly but significantly depressed by CDP (also see Fig. 2), while reversal learning was substantially retarded by the drug. So far as the trial variable was concerned, these was no significant interaction or no indication of it (and thus no drug-learning dissociation) between drugstates during acquisition and reversal.<br>Fig. 3 shows per cent errors during reversal training: per cent errors on the first reversal day clearly indicated a dissociative effect: that is, the rats without drug-state change (SAL-SAL, CDP-CDP) made more errors than those with such change (SAL-CDP, CDP-SAL). In addition, those rats given drug during acquisition were better in performance than those which had been treated with saline. This finding may be at least partially explained by a negative correlation within each group between trials to criterion in acquisition and the number of errors on the first reversal day (Table 1). That is, response strength would be weaker when more trials were needed to the identical learning criterion.<br>No significant effects of CDP upon running times were found either during acquisition or reversal except that the drug significantly retarded wrong-response speed on the first acquisition day (Table 2).<br>A depressant effect of CDP upon discrimination learning was explained in terms of the drug's inhibitory action on negative incentives. It may be assumed that this effect was slight during acquisition but strong during reversal because the formerly-rewarded and thus strongly-reinforced response was punished in reversal and this yielded an approach-avoidance conflict: CDP attenuated the conflict and facilitated the approach response to the punished, negative stimulus.<br>The dissociated performance found on the first reversal day may be explained in terms of generalization decrements since the internal stimulus was changed by shift in drug-state, but this dissociative effect was only transitory, being replaced by the drug's inhibitory effect on negative incentives.<br>The drug's dissociative effect and the depressant effect upon negative incentives seemed to be mediated by different mechanisms. The latter psychotropic effect might be ascribed to the CDP's inhibitory action, direct or indirect, upon the hippocampus. Behavioral studies have shown that hippocampal lesions impair passive avoidance (which is essentially an approach-avoidance conflict), experimental extinction and discrimination reversal (Kimble, 1968). Electrophysiologically, CDP was found to inhibit a prolonged after-discharge, produced by electrical stimulation of the limbic system (Sternbach et al., 1964), as well as to increase seizure threshold of the hippocampus (Hara and Miyake, 1964). Our observation also suggested some and probably depressant effects of CDP upon the hippocampus. Both 10 and 40mg/kg, i. p. of CDP produced continuous rythmic activities of 20-30Hz in cortical and subcortical structures in the waking state. Theta waves, usually associated with gross body movements were not observed under CDP (Fig. 4). Slow-wave sleep was often replaced by high-amplitude