Publication Date


Document Type


First Advisor

Lotsof, Erwin||James, William Homer

Degree Name

M.A. (Master of Arts)

Legacy Department

Department of Psychology


Social psychology; Psychology--Experiments


Goal directed behavior is conceptualized within Rotter's Social Learning Theory as a function of three determinants; they are the value of the goal to the individual (Reinforcement Value), his expectancy for attaining the goal as a function of certain behaviors (Expectancy), and his perception of the situation in which these behaviors are emitted (Psychological Situation). Since Rotter operationally defines the empirical coordinates of these theoretical concepts, Social Learning Theory is developing into a scientific behavior theory. Workers in Social Learning Theory have offered that individuals may, as a function of certain determinants, expect the attainment of a goal to be within (internal to) or beyond (external to) their control. Such expectancies may be associated with situational and/or individual difference variables; that is, situations may be characterized as offering rewards which are controlled by the individual through some skill he must possess, or by chance factors; and individuals may possess characteristic attitudes of feeling responsible for (internal control) or of feeling out of control of the attainment of certain goals. This conceptualization of individual differences seems to be a personality characteristic which affects behavior in a wide variety of situation. The present paper was concerned with the expectancies of internally (I) and externally (E) oriented subjects in a group game of chance as a function of two reward conditions. The acquisition and extinction of Expectancy were empirically defined by the rise and fall in the subjects betting behavior. Expectancies were hypothesized to be a function of both the subjects' general attitudes concerning the locus of control of reinforcement and the per cent of bets that he won. The general attitudes concerning internal and external control of the one hundred and ninety-three subjects were assessed with the James I-E scale. Subjects who scored high on this instrument tend to believe that the occurrence of rewards and punishments is external to their control while the reverse is believed by low scorers. The experimental game was culturally defined as chance; it was played by groups of ten to fifteen and resembled Bingo. During the game the experimenter called numbers, some of which were on the subjects' game cards. The subjects bet prior to each call to indicate how certain they were that the number would be on their card. During the acquisition trials half of the subjects won fifty per cent and half won eighty per cent of their first thirty bets and during the extinction trials no bets were won. Subjects who obtained high (external) and low (internal) scores on the I-E scale were assigned in equal numbers to both reward conditions. As predicted the 80 per cent condition was associated with higher expectancies during the acquisition trials and lower expectancies during the extinction trials than the 50 per cent group. It was therefore concluded that this experiment supported the psychological literature on partial reinforcement percentages and thus it had extended the generality of these hypotheses to a well known externally controlled group situation. The results associated with the I-E construct were also in the predicted direction; that is, the Es tended to have higher expectancies than the Is during both acquisition and extinction., In all but one case, however, the results did not reach statistical significance. The comparison which was statistically significant was between the acquisition trials of the twenty subjects who obtained the highest and the twenty subjects who obtained the lowest scores on the I-E scale. This analysis demonstrated that extreme Es acquired higher expectancies during acquisition than extreme Is. On the basis of this moderate support it was concluded that the I-E construct warrants further investigation as a major behavioral variable. The results of the present study (l) extend the results of studies of partial reinforcement to the betting behavior of subjects in a group game of chance, (2) point to the importance of situational cues in determining expectancies and (3) give moderate support to the I-E construct as an important determinant of individuals' expectancies. Since the results are generally in agreement with Rotter's formulations, the present study has contributed to the scientific significance of Social Learning Theory.


Includes bibliographical references.


xi, 85 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type