The purpose of this study is to examine the influence of social deprivation-satiation operation on subject's awareness of response-reinforcement contingency. Experiment I showed that a subject's awareness is an important factor in considering the effect of social reinforcement. Twenty-four preschool boys took part in a 10-minute social satiation session, in which they received the stimulus word,“orikousan lane”, either 2 (deprivation) or 16 (satiation) times. This was followed by a discrimination test of 75 trials, in which the same stimulus word was given to all correct responses. The test was administered by the experimenter in a preceeding session or by another individual unknown to the subjects. At the end of the test, they were asked about their awareness of response-reinforcement contingency. Ss in the other experimenter group showed more correct responses than in the same experimenter group, and Ss in the aware group more than in the unaware group. It was discussed that the differences in test performance reflect the difference in the experimenter's behavior in the two sessions, and that the efficacy of reinforcement relates to a subject's awareness of responsereinf orcement contingency. Experiment II showed that the low effect of reinforcement in the same experimenter group is based on the experimenter's behavior in the preceeding session, and the subject's unawareness' Thirtysix preschool boys took part in a social satiation session, in which they had no interaction with the experimenter or received the stimulus word,“orikousandane”, either 2 (deprivation) or 16 (satiation) times. It was followed by a discrimination test, in which the same stimulus word was given to all correct responses. A half of subjects were given the information about the correct response in the test by the experimenter. The following was predicted on the basis of the finding of the experiment I; the mean scores in informaton group will be higher than in the no-information group, and those in the no-interaction group will be higher than in the group given the reinforcing stimulus in preceeding session. The results confirmed this prediction, and subjects in the information group belonged to the aware group more than in the no-information group, and the mean scores were higher in the aware group than in the unaware group. It was emphasized that the receiving of the information about correct response leads to increment of correct one, and that a subject's awareness of response-reinforcement contingency encourages discrimination learning. These results were interpreted as indicating that it is necessary to consider the variables mediating social reinforcement effect with performance. Also dicussed was the necessity of examining the relation of the task in a discrimination test with response-reinforcement contingency.