選択行動と対人的知覚の研究 (V)

Abstract

We have concerned ourselves with the relationship among three relations (Rs) contained in the triad: p′s attitude toward q (R1), p′s attitude toward o (R2) and q′s attitude toward o as perceived by p (R3). Up to now a constant discrepancy has repeatedly been found between our results and the prediction derived from Heider's theory of balance. We are then to inquire into the matter more deeply. The purpose of the present study is thus to examine if any difference in methods of collecting and analyzing data may affect the results.In the preceding investigations subjects reported their cognition about q′s attitude toward o in terms of [+], [0], and [-]. The sign of R3 thus obtained was set against those of R1 and R2. Instead, in the present study, each subject was given sets of three stimulus-signs as though they represented three Rs in the triad composed of o, q and p (himself). Some of the stimulus-signs were made to represent rightly the “objective sign”, the real attitude of one person toward another, but the others were not. The subject was asked to correct wrong stimulus-signs presented, if any, according to his awareness, to constitute the “subjective configurations”. In addition he rated the degree of appreciatedd pleasantness of each configuration on a five-point-scale. Some 200 sixth graders served as subjects. The major findings are as follows:1. The subjective configurations are categorized into twenty-seven types by the permutation of three signs contained. In the types where all three signs were coincident the observed frequencies were significantly larger, while in the types where all three signs were different the observed frequencies were smaller than the theoretical ones. In the types where only two signs were coincident, the observed frequencies were either larger or smaller than the theoretical ones, depending on whether the sign of R2 coincided with that of R3 or not. This is completely compatible with our previous results, but not with Heider's theory in several respects. Then we may conclude that the difference in the procedures we employed has no virtual effects upon the results.2. The objective sign was often subjectively changed but not uniformly. In R3 the change occured more frequently than in R1 and R2. When the objective configuration was in an imbalanced state the change was easier than when it was in a balanced state, especially in R3. And in R1 and R2 the change was easier for the objective sign [-] than for [+]. We may thus conclude that the locomotion from an imbalanced to a balanced state is achieved through the path of R3, through mainly cognition rather than action.3. The degree of pleasantness of the configurations as rated by the subjects corresponded roughly to the states of the configurations: the balanced configurations tended to be rated as more pleasant than the imbalanced ones. Another factor, however, seems to be rather important in determining the pleasantness: the subjective sign [+] in any Rs serves to make the rating more positive and the sign [-] more negative, irrespective of the states of the configurations.