Consumer covariation judgments: theory or data driven?

We examined the issues of when, and how, consumers' prior beliefs or "theories" might bias their judgments of the association between price and quality using a task that involved taste testing orange juices. When the task was manipulated so that subjects had difficulty in comparing and contrasting the taste quality of the juices, their covariation judgments tended to be theory driven: while these judgments were somewhat sensitive to the actual price-quality relationship in the "data," they were biased by individuals' prior beliefs about the covariation. Follow-up studies indicated that this effect was not due to an encoding bias in that subjects' prior beliefs about price and quality did not significantly distort their taste perceptions per se. Instead, when subjects found it difficult to discriminate among the stimuli with regard to taste quality, they appeared to have formed their covariation judgments by making heuristic use of their prior beliefs. Implications are discussed for the covariation judgment process and consumer learning.