The Role of Own Cognitive Responses in Persuasion: a Conceptual Overview

Richard M. Perloff, Ohio State University
Timothy C. Brock, Ohio State University
[ to cite ]:
Richard M. Perloff and Timothy C. Brock (1980) ,"The Role of Own Cognitive Responses in Persuasion: a Conceptual Overview", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 741-744.

Advances in Consumer Research Volume 7, 1980     Pages 741-744


Richard M. Perloff, Ohio State University

Timothy C. Brock, Ohio State University

[The authors thank the ACR Convention reviewers for their very thoughtful and perceptive comments on an earlier version of this paper.]


Recent research indicates that own cognitive responses to communications are more influential than recalled message-arguments. Two theoretical interpretations for this finding are offered. This line of research is also applied to personality and persuasibility studies.


In recent years, the cognitive response approach to persuasion has captured a prominent place in the communication and consumer behavior literatures (e.g., Roberts and Maccoby, 1973; Calder, 1978; Petty, Ostrom and Brock, in press). Theoretical refinements have been increasingly proposed, and cognitive response models have been increasingly articulated to account for a wide variety of communication effects (Calder, 1978; Tybout, Stern-thai and Calder, 1978; Sternthal, Dholakia and Leavitt, 1978; Cacioppo and Petty, 1979). In this paper we attempt to extend this theorizing by offering some theoretical conjectures about an important finding in the cognitive response area: namely, the consistent tendency for subject-generated cognitive responses to a communication to be more significantly associated with attitude change than recalled message elements.

A variety of investigators, using different operationalizations, have reported low or nonsignificant correlations between recall of arguments contained in the message and agreement with the position advocated in the communication; at the same time, the number of self-generated proarguments in favor of a communicated position has been consistently associated with acceptance of a communicator's point of view, while the number of self-produced counterarguments has been inversely correlated with attitudinal acceptance of the advocated position (Cacioppo and Petty, 1979; Petty, Ostrom and Brock, in press). For example, Cacioppo and Petty (1979) reported that when both counterarguments and favorable thoughts served as covariates, the original significant main effect of the independent variable on agreement was reduced to nonsignificance. However, the F for agreement remained significant when message recall was the covariate.

Petty (1977) has provided even more convincing support for the proposition that cognitive responses are more influential than message-arguments. In this experiment, subjects read five high or low persuasive arguments on the topic of raising the driving age to 21. Subjects listed five thoughts on the subject and were instructed to memorize either their thoughts or the arguments. (Attitude change was assessed both immediately and a week later by comparing subjects' attitudes on the driving issue with those of a control group that read five neutral statements on the same topic.) The results on the immediate measurement indicated that regardless of whether they memorized thoughts or arguments, subjects who read the highly persuasive arguments generated relatively favorable thoughts about increasing the driving age to 21, and held the most positive attitudes on the issue. However, subjects who read the highly persuasive arguments, and memorized their thoughts, were most favorable to raising the driving age when attitudes were assessed a week later. The findings support the cognitive response prediction that rehearsal of owe thoughts is a more potent determinant of persistence of persuasion than rehearsal of message-arguments. Apparently, a highly persuasive communication elicits favorable thoughts which help determine immediate post-communication attitudes on the issue; to the extent that these favorable cognitive responses continue to be salient, attitudes at the delayed measurement are similar to post-message attitudes.

Why are own cognitive responses apparently more influential than message-arguments? In the next sections we offer two theoretical interpretations for this finding, and attempt to apply our approach to personality and persuasibility research.


Individuals remember more of their self-generated arguments to a communication than those improvised by another subject (Greenwald, 1968). They also remember a greater proportion of the thoughts they generate about a message than of the arguments contained in the communication itself (Petty, 1977). Additional support for the finding that people learn their own cognitive responses better than equivalent statements provided by someone else has been recently provided by Slamecka and Graf (1978).

In a series of intriguing experiments, Slamecka and Graf demonstrated that individuals better remembered words when they generated the words themselves than when they were simply presented to be read. This finding held for measures of cued and uncued recognition and free and cued recall. It persisted across variations in rules for encoding, self-paced or timed presentation, presence or absence of test information and between -or within-subject designs. These results extend the generalizability of the cognitive response findings to the arena of human learning and memory, suggesting the fundamental nature of these results.

Why are self-generated cognitive responses better learned than the statements contained in a message or provided by an experimenter? One explanation may he located in recent research on the self (Markus, 1977; Rogers, Kuiper and Kirker, 1977). Markus has asserted that people possess a more well-developed cognitive schema for self-related information than for other types of information. Thus, information about the self or associated with the self (such as one's thoughts) can be more extensively processed than information from other sources (such as from the message).

In essence, one's own pro or counter arguments are more meaningfully and extensively processed than the arguments that are contained in the message. An example from everyday experience may clarify this point. We have all read over our colleagues' manuscripts and been dismayed to discover obvious typos. And yet it is our colleagues who have the last laugh, for when we read over our own papers, we invariably skip over typographical errors as well. Why are we our own worst proofreaders? Because we read our own materials for content and meaning, not for surface attributes. Analogously, when we rehearse our own thoughts about a persuasive communication, we engage in more extensive processing than when we rehearse the arguments in the message.

Other interpretations may also be advanced to explain why subjects remember self-generated material (e.g., thoughts) better than information produced externally (for example, message-arguments). Slamecka and Graf (1978, p. 603) explained their findings on this subject by suggesting that:

. . . Initial recall confers beneficial consequences upon a subsequent memory test on the same material. As applied to the present paradigm, it would suggest that the act of generation is really an instance of recall, with the source being semantic memory . . . In contrast, a reading task involves no recall-based episodes, since all responses are given.

In addition, Slamecka and Graf suggest that generation involves extensive "tagging of nodes in the associative network, thus increasing access routes" (p. 603).

Whatever the reasons, there is evidence suggesting that self-generated thoughts appear to be more extensively processed and better remembered than message-arguments. We now turn to the issue of how these findings relate to the evidence that own thoughts are more influential than message-arguments.

Brock (1962) showed clearly that own thoughts processed for meaning were more influential than own thoughts processed for grammar and syntax. Given that own thoughts about a communication are more likely to be processed for meaning than the arguments in the message (Markus, 1977; Rogers, Kuiper, and Kirker, 1977), Brock's (1962) research suggested that thought rehearsal should be more persuasive than rehearsal of message-arguments. Stated somewhat differently, attention is a necessary condition for persuasion (Osterhouse and Brock, 1970); because we probably direct more and "deeper-level" attention to our own thoughts than to the arguments in the message, we more thoroughly learn the persuasive arguments contained in our own thoughts than the arguments that are contained in the persuasive message provided by the experimenter.

Needless to say, this is an empirically-testable proposition. We propose a study to test our hypothesis. Subjects would read a persuasive communication and then list their thoughts about the issue. [If the communication is persuasive, it should elicit favorable thoughts from subjects (Petty, 1977; Cacioppo and Petty, 1979)]. Subjects would subsequently read an equally persuasive message on a very similar issue, and list the arguments in the message. [Order of presentation would be randomized such that half of the subjects list thoughts first, and half list message-arguments first.] Attitudes toward both issues would then be measured. If own thought processing is operative, subjects should remember more of their thoughts than the arguments in the communication. Since learning of persuasive material can facilitate attitude change (McGuire, 1968; Osterhouse and Brock, 1970), subjects should be more influenced by the first message -- about which they listed their thoughts -- than the second communication.


Greenwald (1968) reported that subjects regarded their own cognitive responses as more original than those improvised by another subject. Greenwald concluded that people appear to evaluate their own thoughts about a persuasive communication more positively than others' thoughts. Slamecka and Graf (1978) also demonstrated that subjects are more confident about their performance when they generate words themselves than when the same words are simply read. Why are self-generated responses so influential? We contend that people have a tendency to value products associated with the self and consequently attach value to their own cognitive responses to persuasive communications. Because these cognitive responses "belong to" or are associated with the self, they accrue more value and exert more influence than the arguments contained in the message.

The "ownness prejudice" stems from both territorial and nonterritorial motives. Aspects of one's personal space are valued even though these aspects may not be otherwise distinguishable. We prefer our own bed, even though it is the same as other beds in the barracks or dormitory. We prefer our own place or chair at a table. We prefer our towel, although identical to your towel. If external objects associated with the self can become so coveted, it is not surprising that internal extensions of the self, such as thoughts, dreams, fantasies, etc., can be accorded special value. Coveting and protecting personal space extends, we assume, to cognitive components of that space. A territorial perspective is sufficient to account for an "ownness prejudice.'' [There are undoubtedly individual differences in the extent to which people value their own cognitive responses to communications. Not everybody will attach more value to their own thoughts because they represent cognitive extensions of the self. Furthermore, the operation of self-oriented needs does not exclude the simultaneous operation of other processes. For example, people sometimes change their minds about an issue to escape the uncomfortable feeling they experience when they adopt an unpopular or unusual position. In these situations, individuals may attach more value to socially acceptable message-arguments than to their own deviant viewpoints.]

A number of nonterritorial motives could cause individuals to assign greater weight to mental products that they felt they generated themselves. The notion that people value their own products and efforts has been articulated by psychoanalytic theorists (e.g., Freud, 1938) who have long maintained that this egocentric tendency can be traced to basic inclinations to value extensions of the self, such as even feces, for example. More recently, Kohlberg (1966) has contended that children prefer same-sex objects and characteristics because they have a natural, built-in inclination to like objects and traits that are similar to themselves.

These explanations of the "ownness prejudice" are not meant to be exhaustive. Still other factors could be involved. For example, subjects may attribute persuasive intent to the communicator and be therefore suspicious of the message-arguments. By contrast, individuals do not ascribe any hidden motives to themselves when they list their own thoughts about an issue. Compared to message-arguments, subjects' thoughts may possess a freshness and originality that is not tainted by any attribution of persuasive intent. As a consequence, own thoughts may be more credible, valued --and influential -- than message-arguments.

By this point the reader may be wondering how anyone is ever persuaded about anything. After all, if people's own thoughts about a message are always more persuasive than communicators' arguments, then no one would ever change his or her attitudes about anything. Of course, people are persuaded to adopt different positions. It turns out that this is really not a problem for the cognitive response approach. We maintain that once a communicator has started to change people's minds about an issue, he or she can be most assured that this change will persist if audience members rehearse their own thoughts about the message rather than the speaker's arguments. If people begin to think more favorably about the issue, then attitude change, and persistence of this change, is aided and abetted by the production and rehearsal of self-generated favorable thoughts.


In this section we have suggested two explanations for the finding that own thoughts are more significantly associated with attitude change than recalled message-arguments. We first discussed research indicating that subjects remember self-generated cognitive responses better than those provided by someone else. An information-processing interpretation was invoked to explain these findings. We then suggested how more extensive processing of own cognitive responses could help account for their influence and persuasive power. A second explanation for the cognitive response findings was also proposed, based on people's tendency to prefer and value their own products. This tendency was called the "ownness prejudice."


In the preceding sections we have discussed several issues that have interesting implications for personality and persuasibility. Although there has been abundant research on individual differences in susceptibility to persuasion, the yield has not matched the expenditure of research energies. After over 20 years of research, we still are not certain of the impact of individual difference variables on persuasion; nor do we know which personality variables exert the greatest influence on persuasibility, and why.

We suggest that research attention be directed to the influence of a somewhat overlooked personality variable: the need to be unique and different from others (Fromkin, 1970, 1972, 1973). For example, individuals who have a strong need to be unique may be more sensitive to their own cognitive responses than individuals low in uniqueness needs. Subjects high in uniqueness needs should be especially motivated to regard their thoughts about a message as uniquely and distinctively their own. They should attach greater value to, and have more respect for, these thoughts and may even be more confident about their cogency and intellectual value than subjects low in uniqueness needs. As a result, cognitive responses should predict attitudes better for these subjects than for subjects low in uniqueness needs. Individual differences in the extent to which people are generally biased toward their own products may have interesting implications for research on personality and persuasibility.

For example, to the extent that a good self-image involves the favorable evaluation of one's own cognitive products, high self-esteem subjects should regard their own cognitive responses to a persuasive message as more original and of higher overall quality and importance than low self-esteem subjects. We predict that high self-esteem individuals will be more persuaded by a message when they rehearse their own proarguments than when they rehearse the favorable arguments contained in the communication. The opposite prediction is made for low self-esteem persons, who are likely to attach less value to their own arguments than to those generated by someone else.

A caveat needs to be added here. We believe that only a particular dimension of self-esteem may affect own cognitive responses and subsequent susceptibility to persuasion (Hedges, 1974; Perloff, 1978). It is possible, for example, that only individuals who have a high regard for their own intellect and cognitions will be most persuaded by a communication when they rehearse their own thoughts, and that only persons who have a low regard for their cognitive skills will be most influenced when they rehearse the arguments in the message. In essence, then, non-cognitive dimensions of the self-concept (such as self-esteem about social skills) should not affect the value people attach to their own cognitive responses to communications.

Note that our approach differs somewhat from McGuire's (1968) information-processing formulation of individual differences in persuasibility. McGuire contended that learning and comprehension of the content of a persuasive communication mediate the influence of personality factors on susceptibility to persuasion. McGuire asserted that the impact of a personality variable on persuasion will vary as a function of the complexity of the persuasive message and the individual's comprehension of the message. According to McGuire, comprehension and message learning are likely to mediate the impact of individual difference variables on persuasion when the message allows for variance in people's comprehension of the message, i.e., when the communication is complex and subtle.

McGuire's approach must be questioned, however, in view of the repeated evidence that message learning is only weakly correlated with attitudinal acceptance of the communication. The documented importance of cognitive responses in persuasion suggests an alternative proposition. Rather than solely emphasizing the mediating role of message comprehension and learning, we conjecture that the impact of individual difference variables on persuasion is also mediated by people's perceptions of the value and quality of their own cognitive responses to communications.

Individuals who are at high and low levels of a personality variable should place differential value on their arguments with a counterattitudinal communication. Thus, to the extent that a good self-image involves the favorable evaluation of one's accomplishments, products, and thoughts, high self-esteem persons should perceive that their counterarguments are of relatively high quality; believing that they effectively refuted the advocated position, they should be relatively immune to persuasion. On the other hand, low self-esteem subjects, who have a lower image of their overall competence, should perceive their counterarguments to be of relatively low quality; given their belief that they have not effectively criticized the speaker's position, they should be prime candidates for attitude change and persuasion.

Other individual difference variables may also exert predictable effects on resistance to persuasion to the extent that they influence people's perceptions of the value and quality of their own cognitive responses to communications. On the other hand, there may be dispositional variables that do not influence persuasibility because they do not affect the perceived value or quality of own mental reactions to communications.


Brock, Timothy C. (1962), "Cognitive Restructuring and Attitude Change," Journal of Abnormal and Social Psychology, 64, 264-271.

Cacioppo, John C. and Petty, Richard E. (1979), "Effects of Message Repetition and Position on Cognitive Responses, Recall and Persuasion," Journal of Personality and Social Psychology, 37, 97-109.

Calder, Bobby J. (1978), "Cognitive Response, Imagery and Scripts: What is the Cognitive Basis of Attitude?" in Advances in Consumer Research: Vol. 5, ed. H. Keith Hunt, Ann Arbor, Mich.: Association for Consumer Research, 630-634.

Freud, Sigmund (1938), The Basic Writings of Sigmund Freud, New York: Modern Library.

Fromkin, Howard L. (1970), "The Effects of Experimentally Aroused Feelings of Undistinctiveness upon Valuation of Scarce and Novel Experiences," Journal of Personality and Social Psychology, 16, 521-529.

Fromkin, Howard L., (1972), "Feelings of Interpersonal Undistinctiveness: An Unpleasant Affective State," Journal of Experimental Research in Personality, 6, 178-185.

Fromkin, Howard L., (1973), "The Psychology of Uniqueness: Avoidance of Similarity and Seeking of Differentness," Institute for Research in the Behavioral, Economic and Management Sciences, Paper No. 438, Purdue University.

Greenwald, Anthony G. (1968), "Cognitive Learning, Cognitive Response to Persuasion and Attitude Change," in Psychological Foundations of Attitudes, eds. Anthony Greenwald, Timothy Brock and Thomas Ostrom, New York: Academic Press, 147-170.

Hedges, Noel (1974), "Attitude Change as a Function of Confidence in Counterargumentation," Personality and Social Psychology Bulletin, 1, 141-143.

Hovland, Carl I., Lumsdaine, Arthur A. and Sheffield, Fred D. (1949), Experiments on Mass Communication. Princeton, NJ: Princeton University Press.

Kohlberg, Lawrence (1966), "A Cognitive-Developmental Analysis of Children's Sex-Role Concepts and Attitudes," in The Development of Sex Differences, ed. Eleanor Maccoby, Stanford, Calif.: Stanford University Press.

Markus, Hazel (1977), "Self-Schemata and Processing Information about the Self," Journal of Personality and Social Psychology, 35, 63-78.

McGuire, William J. (1968), "Personality and Attitude Change: An Information-Processing Theory," in Psychological Foundations of Attitudes, eds. Anthony Greenwald, Timothy Brock and Thomas Ostrom, New York: Academic Press, 171-196.

Osterhouse, Robert A. and Brock, Timothy C. (1970), "Distraction Increases Yielding to Propaganda by Inhibiting Counterarguing," Journal of Personality and Social Psychology, 15, 344-358.

Perloff, Richard M. (1978), An Attributional and Cognitive Response Approach to Personality and Persuasibility, Unpublished Doctoral Dissertation, University of Wisconsin.

Petty, Richard E. (1977), A Cognitive Response Analysis of the Temporal Persistence of Attitude Changes Induced by Persuasive Communications, Unpublished Doctoral Dissertation, Ohio State University.

Petty, Richard E., Ostrom, Thomas M. and Brock, Timothy C. (In press), eds., Cognitive Responses in Persuasion: A Text in Attitude Change, Hillsdale, NJ: Erlbaum Associates.

Roberts, Donald F. and Maccoby, Nathan (1973), "Information Processing and Persuasion: Counterarguing Behavior," in New Models for Communication Research, ed. Peter Clarke, Beverly Hills: Sage Publications.

Rogers, T. B., Kuiper, N. A., and Kirker, W. S. (1977), "Self-Reference and the Encoding of Personal Information" Journal of Personality and Social Psychology, 35, 677-688.

Slamecka, Norman J., and Graf, Peter (1978), "The Generation Effect: Delineation of a Phenomenon," Journal of Experimental Psychology: Human Learning and Memory, 4, 592-604.

Sternthal, Brian, Dholakia, Ruby and Leavitt, Clark (1978), "The Persuasive Effect of Source Credibility: Tests of Cognitive Response," Journal of Consumer Research, 4, 252-260.

Tybout, Alice M., Sternthal, Brian and Calder, Bobby J. (1978), "A Two-Stage Theory of Information Processing in Persuasion: An Integrative View of Cognitive Response and Self-Perception Theory," in Advances in Consumer Research: Vol. 5, ed. H. Keith Hunt, Ann Arbor, Mich.: Association for Consumer Research, 721-723.