Inertia in Cognitive Processes: The Role of Accumulated Information in Attitude Change

Abstract

Research evidence is presented to support a new and simple theory of attitude formation and change. This theory posits that the attitude of any individual converges overtime on the arithmetic mean of the attitude-pertinent information received by the individual. Consequently, the stability of an attitude is dependent on the number of messages out of which that attitude was formed. This formulation also implies that the emotional state or feelings of an individual and the degree of heterogeneity of influences to which he or she was exposed are unrelated to attitude change. Using a multi-stage, multi-time procedure, and instruments designed to detect and measure interpersonal influence, data provided by 135 high school students over a six-month interval support these hypotheses.While the notion that cognitive processing of information may be too complex and unpredictable to be represented by a few simple quantitative laws or principles remains a suspicion even among some social scientists, the last decade of empirical research has produced many cases in which some of these processes have been described quite accurately with very simple quantitative formulations. Cliff (1972), for example, asked 31 subjects to estimate the favorability of 10 adjectives describing personality traits on a series of nine point scales. Subjects also rated the favorableness of all possible pairs of these adjectives on the same scales. Cliff found that the favorability of the pairs of adjectives could be predicted very accurately by the average favorability of the two component adjectives, for example, L where a, = the favorableness of the first adjective, a2 = the favorableness of the second adjective, aI2 = the favorableness of the pair taken together.On the average, 96 percent of the variance in the favorability score of the pairs was accounted for by the simple average of their components. Monotonic transformations to account for potential curvilinearity were able to explain on the average only .025 additional percent of the variance.In a more complex survey design, Woelfel and Haller (1971) measured the status aspirations (i.e., levels of education and occupational prestige to which subjects aspire) of a set of 99 high school students, and showed that about half of the variance in these expectations could be accounted for by the average status expectations of a set of independently identified ''significant others" with whom the respondents reported communicating. In a study of the determinants of marijuana use, Woelfel and Hernandez (1972) found that 86 percent of the variance in rate of marijuana use by 341 U.S. and Canadian college students could be accounted for by a weighted average of the information about marijuana they reported having received. In this same vein, Reeves (1974) found that nearly half the variance in the extent to which young children see TV characters as real or fantasy can be accounted for by a weighted average of the opinions of a subset of their significant others. Woelfel, Woelfel, Gilham, and McPhail(l974)...

The Comparative Statics and Dynamics of Beliefs: The Effect of Message Discrepancy and Source Credibility

“…Information integration theory (Anderson, 1971;Anderson & Farkas, 1973) makes the same prediction (see also Himmelfarb, 1974;Saltiel & Woelfel, 1975). However, research on the effect of discrepancy on belief change generates results at odds with the distance-proportional model.…”

A dynamic model of the effect of discrepant information on unidimensional attitude change

“…A4: As in Newtonian mechanics, we assume that the amount of acceleration of a concept in the cognitive space will be equal to the amount of force acting upon the concept divided by the mass of that concept. A5: Moreover, the inertial mass of the concept is assumed to be a monotonically increasing function of the information the actor possesses about that concept (see Saltiel &Woelfel, 1975 andDanes, Hunter, &Woelfel, 1978 for evidence supporting this assumption).…”

“…A number of investigators (Saltiel & Woelfel, 1975;Danes, Hunter, & Woelfel, 1978;Laroche, 1977;Anderson, 1974;Hunter & Cohen, 1972;Fink, Kaplowitz & Bauer, 1983) have proposed mathematical models which predict attitude change as a function of the discrepancy between one's initial position and the position advocated in a message, plus perhaps such other variables as source credibility and prior information about the topic.…”

Exploring persuasive message strategy for international aid campaigns: potential donors’ inter-attitudinal structure using Galileo model

“…The explanation of the procedure is as follows: Messages act as forces that alter the position of concepts in the attitude space. When two words or phrases are associated in the same message, they approach each other in the space (Saltiel and Woelfel 1975;Woelfel and Saltiel 1988). When a number of concepts are associated in the same message, all concepts in the message approach the common center of the concepts in the message.…”