A test of cognitive balance theory implications for social inference processes.

An Analysis of the Conceptual Representation of Relations: Components in a network model of cognitive organization1

“…For relations expressing positive affect, the schema includes the assumption of transitivity. If a likes b and b likes c, the person will assign a high probability to the fact that a will like c. This component of the schema is supported not only by the work of DeSoto and Kuethe but also by studies of patterns of completion of triadic structures: Given two positive affective relations, subjects complete the triad with a third positive relation (Atwood, 1969;Fuller, 1974;Rodrigues, 1968;Vickers & Blanchard, 1973;Wyer & Lyon, 1970).…”

Some implications of the "Socratic effect" for alternative models of cognitive consistency1

“…Although some research on social inference processes has not strongly supported the validity of balance/congruity principles in describing these processes (cf. Rubin & Zajonc, 1968, Wyer and Lyon, 1970, see also Newcomb, 1968, Zajonc, 1968, in other cases these prmciples have been applied to social inference phenomena quite successfully (cf. Feafiier, 1967, Tannenbaum & Gengel, 1966 A distinction must be made between the process of inferring unknown relations among unfamihar objects on the basis of new information about them and the process of organizing previously formed cognitions about familiar objects and events.…”

Factors affecting balance, agreement and positivity biases in POQ and POX triads

“…Few studies have specifically looked at the differences that this variable might cause. However, amongst those that have are Crockett (1974) Rossman and Gollob (1976); Rubin and Zajonc (1969), and Wyer and Lyon (1970). In one of the conditions of the Rubin and Zajonc experiment a POQ triad was used and in a second condition a POX triad.…”

“…The data showed no difference with regard to the balance bias, though the agreement bias (actually termed a 'popularity' bias) had a larger effect in the POQ condition. Wyer and Lyon (1970) asked subjects to infer relationships between entities in a variety of contexts, and suggested that balance theory is more applicable for inference of POX triads than POQ triads. There appears to be little supporting evidence for this admittedly tentative conclusion.…”

“…A few studies have specifically looked at such structural differences (e.g. Crano and Cooper, 1973;Wyer and Lyon, 1970), but decisive evidence has not yet been given. One may question, however, whether such differences, both in methodology and in structural characteristics, may be influential in causing particular biases to operate.…”