Effects of message repetition and position on cognitive response, recall, and persuasion.

Although the mere exposure effect has been researched widely, surprisingly little is known about the attitudinal and cognitive effects of message repetition. It was hypothesized that the sequence of topic-relevan t thoughts generated in response to a (repeated) persuasive message would parallel attitude change. To test this prediction, two experiments were conducted. In Experiment 1, individuals heard a communication either zero (control), one, three, or five times in succession, rated their agreement with the advocated position, and listed the message arguments they could recall. In Experiment 2, individuals heard a communication either one, three, or five times, rated their agreement, listed their thoughts, and listed the message arguments they could recall. In both experiments, agreement first increased, then decreased as exposure frequency increased (regardless of the position advocated), but agreement was unrelated to the recall of the message arguments. In Experiment 2, analyses of the listed thoughts revealed that counterargumentation decreased, then increased, whereas topic-irrelevant thinking increased as exposure frequency increased; as expected, only topic-relevant thoughts were related to agreement. These results are interpreted in terms of an attitude-modification model in which repetition and content of a persuasive advocacy affect the type and number of thoughts generated; these thoughts, in turn, affect the attitudinal reaction to the advocacy. In this article, we will consider the atti- using children as well as adults (Heingartner tudinal effects of repeated exposure to per- & Hall, 1974), employing between-subjects suasive communications, an area that has (Moreland & Zajonc, 1976) as well as withingenerated surprisingly little research by social subjects (Crandall, 1972) designs, in field psychologists despite its frequent occurrence (Zajonc & Rajecki, 1969) as well as laborain and importance to everyday life. This area tory (Matlin, 1970) settings, employing picis not well understood (cf. Harrison, 1977), torial magazine advertisements (McCullough in part because most research has focused on & Ostrom, 1974) as well as nonsense syllables