An Examination of Information Processing Traits: General Social Confidence and Information Processing Confidence

ABSTRACT - This paper reviews Wright's (1975) findings concerning the role of the information processing traits General Social Confidence (GSC) and Information Processing Confidence (IPC). This review concludes that Wright's findings were an artifact of the experimental procedure. Also this paper presents suggestions for future research concerning the role of information processing traits as factors influencing cognitive resistance to persuasion.

Citation:

James M. Munch and John L. Swasy (1981) ,"An Examination of Information Processing Traits: General Social Confidence and Information Processing Confidence", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 349-354.

Advances in Consumer Research Volume 8, 1981      Pages 349-354

AN EXAMINATION OF INFORMATION PROCESSING TRAITS: GENERAL SOCIAL CONFIDENCE AND INFORMATION PROCESSING CONFIDENCE

James M. Munch, The Pennsylvania State University

John L. Swasy, The Pennsylvania State University

[James Munch is a doctoral student in Marketing and John Swasy is an Assistant Professor of Marketing at The Pennsylvania State University, University Park, Pennsylvania 16802. This research was supported by a grant to John Swasy from the Center for Research, College of Business Administration, The Pennsylvania State University. The authors gratefully acknowledge the helpful comments from the ACR reviewers. All correspondence should be addressed to John Swasy.]

ABSTRACT -

This paper reviews Wright's (1975) findings concerning the role of the information processing traits General Social Confidence (GSC) and Information Processing Confidence (IPC). This review concludes that Wright's findings were an artifact of the experimental procedure. Also this paper presents suggestions for future research concerning the role of information processing traits as factors influencing cognitive resistance to persuasion.

INTRODUCTION

The systematic examination of persuasive communication is of great importance to both the public and private sectors of our economy. According to 1980 projections, U.S. total advertising expenditures will surpass $55 billion. The ability to increase the effectiveness of such expenditures rests upon understanding audience responses to persuasive communications.

Early research in persuasion typically focused upon the direct measurement of outcome variables (e.g., attention, recall, beliefs, attitudes). More recent efforts have taken a process-oriented focus in an attempt to achieve a greater understanding of persuasion. One common approach is based upon evidence that receivers of a persuasive message actively process and subvocally react to the message content (Greenwald 1968). Reports of these reactions have been termed cognitive responses and are believed to reflect the results of the receiver's information processing. One typical categorization of such responses distinguishes counterarguments and support arguments to the message claims and also derogations or attacks against the message source (Wright 1973). Furthermore, these reported thoughts have also been categorized according to their origin (i.e., thoughts which were restatements or reflections of a message statement, or thoughts which were "original" or developed by the receiver, independent of message claims) (Greenwald 1968).

Marketing and psychology studies have shown that frequency counts of these response categories, taken from receivers' reports of their thoughts during a message, are good predictors of subsequent message acceptance and attitude change. This predictive ability, as well as physiological evidence supporting the mediating role of subvocal responses in attitude change (Cacioppo, Sandman and Walker 1978), provides strong evidence that cognitive responses such as counterarguing may underlie persuasion outcomes.

Much theorizing and research have been directed toward understanding the antecedents of message-evoked thoughts.

Such differential response to persuasive message arguments has been attributed to: prior experiences (e.g., individual differences), differences in communications (e.g., source effects), and the social context (e.g., situational factors), (Eagley 1980). Research has been undertaken to examine such factors as the accessibility of various responses from long-term memory (Edell and Mitchell 1978), prior commitment on the message topic (Osterhouse and Brock 1970), response opportunity (Petty, Wells and Brock 1976) and message processing goals (Petty and Cacioppo 1979).

To date, however, there has been quite limited interest in research attempting to systematically examine these cognitive responses in relation to receivers' information processing traits. The only published article to address the relationship between receivers' information processing traits and cognitive responses is Wright (1975). The purposes of this paper are to position Wright's research within the framework of the existing personality research paradigms, to review Wright's findings and to suggest an alternative interpretation of his results. Additionally several suggestions for future research dealing with the study of information processing traits will be presented.

Fundamental Paradigms in Personality Research

The premise of the relative stability of personality variables represents one of the classic issues for debate in personality research. The purpose of this section of the paper is not to resolve the issue of stability in personality research. Rather, our attempt is to briefly examine three fundamental paradigms pertinent to this issue and to position Wright's conceptualization of General Social Confidence (GSC) and Information Processing Confidence (IPC) within this framework. The three paradigms to be reviewed are the situationalist, trait, and interactionist positions. For a more comprehensive discussion of these personality/ stability issues see Epstein (1979).

Situationalist Position.  The situationalists argue that behavior is determined almost exclusively by situational variables. This premise is based on three fundamental sources of evidence:

(1)  When behavior in one situation is correlated with behavior in another situation the results are extremely low--usually less than .30. Thus, Mischel (1968) notes, "A correlation of .30 leaves us understanding less than 10% of the relevant variance."

(2)  Analysis of variance results have demonstrated that the variance attributable to individual differences is usually much smaller than the variance attributable to the interaction of individuals and situations, Endler and Hunt 1968, 1969; Endler, Hunt and Rosenstein 1962).

(3)  Ratings of stability of personality by others have consistently overstated the stability of individuals across situations (Bem and Allen 1974; Jones and Nisbett 1971; Mischel 1968; Shweder 1975).

Trait Position.  Trait theorists contend that the failure to find stable traits across situations is primarily due to inappropriate research procedures, a host of arguments are advanced to support such a contention, the more compelling being:

(1)  The unit of analysis has not been adequately taken into account. What appears to be instability at a behavioral level of analysis may be stability at the trait level of analysis (Alker 1972; Bowers 1973). Stated differently, apparently unstable behaviors may be based upon relatively stable traits.

(2)  The use of moderate variables would consistently increase reliability coefficients (Alker 1972).

(3)  A failure to recognize that some people are more variable than others results in reporting generally low stability coefficients.

(4)  Stability in personality is mediated by an individual's cognitions; therefore, stability will only be found when ideographic (i.e., symbolic or graphic) procedures are used that take into account the subjective nature of perception (Alker 1972; Bem and Allen 1974; Buyers 1973; Mischel 1973).

Interactionist Position.  Although the interactionist position does not resolve the issue of stability of personality, proponents of the interactionist position contend that the controversy between individual (i.e., trait) versus situation is meaningless, (i.e., behavior is always a function of the joint interaction of these variables.) Clearly, in any psychological situation the interaction of individuals and situations should account for more variance than either source of variance by itself. Thus it has been argued that an interactionist position should supplant both the trait and situationalist positions (Bowers 1973; Ekehammer 1974; Endler 1966; Endler and Hunt 1968).

As Epstein (1979) suggests, the situationalist position, trait position and the interactionist position are in fact all "correct" approaches to different problems. The situationalist focuses upon the general effects of situations over a sample of individuals. The trait theorist examines the consistency of behavioral tendencies in individuals over a sample of situations. The interactionist studies the behavior of people with certain attributes in situations with certain attributes.

The researcher's decision regarding which personality paradigm or conceptual framework to choose is thus dictated by the research purpose. Since Wright conceptualized General Social Confidence (GSC) and Information Processing Confidence (IPC) as "chronic" traits and examined these traits in four different reception environments, an interactionist position is presumed. Additional comments regarding the compatibility of Wright's conceptual framework with the measurement of that conceptual framework are presented in later discussion.

WRIGHT (1975)

This section of the paper will briefly review Wright's hypotheses and methodology. The discussion will be limited to only those aspects most germane to the role of information processing traits. Following this overview, Wright's results will be presented and examined.

In his article, "Factors Affecting Cognitive Resistance to Advertising," Wright discusses:

(1)  the role of information processing traits, and

(2)  decision task involvement in audio and print message presentations.

The information processing traits under study were Information Processing Confidence (IPC) and General Social Confidence (GSC). Both GSC and IPC were conceptualized as "chronic" personal characteristics related to transitional abilities and habits. More specifically, GSC was regarded as one's degree of skill or self-confidence in assessing personal ability in coping with everyday problems (i.e., a characteristic which exhibits both social and cognitive dimensions). IPC was described as one's confidence regarding the ability to process information or think fast (i.e., a characteristic which is evidenced by quickness of mental reflexes or cognitive agility). It should be noted that both were self perceptions of one's ability and not necessarily equivalent to actual knowledge structures or processing performance levels.

The decision task involvement factor addressed a motivational determinant of a receiver's cognitive resistance to persuasion. Subjects in the "high processing involvement" treatment were informed they could expect to make a short-run decision about the product appearing in the impending advertisement. The relevance of this decision was emphasized in terms of their families, their own time and effort, and their personal finances. Subjects in the "low content involvement" treatment received no such information.

Wright's persuasive message treatment consisted of the presentation of either an audio or print advertisement. The audio message was defined to be a strainful information processing situation since subjects receiving the treatment had no control over the speed of message delivery or inflections of the speaker. Those subjects who received the print ad were by definition, engaged in a relatively leisurely information processing task.

Wright's Hypotheses

The focus of Wright's research was to address the form of the GSC-message acceptance relationship. Cognitive responses were also examined in order to test the hypothesis that people low in GSC might find it difficult to cope with both message encoding and critical message scrutinizing under strained processing conditions (i.e., audio mode) and therefore opt for source evaluation responses (e.g., source derogations) rather than message oriented responses (e.g., counterargumentation). Additionally, Wright introduced the information processing confidence (IPC) construct as an indicator of a person's mental reflexes.

Wright's hypothesized arguments regarding GSC and IPC are summarized as follows:

(1)  In a strained information processing environment (i.e., audio mode) both GSC and IPC will shape the nature of the responses. This can be contrasted with the unstrained (i.e., print) condition in which no strong relationship between GSC and IPC and message acceptance was expected.

(2)  In the audio presentation to a highly involved audience, a sense of social confrontation will be created. Therefore as in highly social situations, both GSC and IPC will be important determinants of a receiver's responses, however GSC will be relatively more important.

(3)  For audio presentations which reflect a less social confrontation (i.e., low involvement), both the GSC and IPC factors will again emerge, but those information processing capabilities (i.e., IPC) are posited to contribute more to individual differences in responses.

Figure 1 illustrates Wright's 2 x 2 crossed factorial design and highlights these hypothesized relationships.

Wright's Methodology

Subjects in Wright's study were 160 adult housewives. Each woman was randomly assigned to one of the four communication conditions in the 2 x 2 crossed-factorial design. A new type of food product (i.e., a soybean derivative) was chosen for the topic of the message to minimize product knowledge differences. The message arguments contained six favorable

FIGURE 1

WRIGHT'S 2 X 2 CROSS FACTORIAL DESIGN AND HYPOTHESIZED RELATIONSHIPS

Reasons for possible product adoption. In order to simulate as far as possible the natural conditions in which people encounter mass media advertising messages, Wright presented the ad following a short "editorial" statement. Immediately after exposure, cognitive responses were collected, followed by measures of message acceptance, content recall, and a check of the involvement manipulation. Subjects then rated each thought as "extremely important," "moderately important," or "slightly important." Also they rated each thought according to whether they felt it had originated directly in the message, had been a reaction on her part to an idea directly stated in the message test, or was an "original" thought in the sense that it was self-initiated rather than a response to a direct statement in the message text (Wright 1975, p.5). This thought judging task was then followed by a peer evaluation task [The peer-designated opinion leadership factor is not central to our analysis and will not be addressed in this review] and the self report GSC and IPC measures. These are reported in Table 1.

Wright's Results

Wright's findings were generally consistent with the hypothesis presented above. The GSC and IPC scores were more highly correlated with counterarguing (CA) in the strained (i.e., audio) than in the less strained (i.e., print) conditions. As expected, for the audio presentation when involvement was low, the IPC-CA correlation was stronger than the GSC-CA correlation, while the GSC-CA correlation was stronger in the high-involvement audio condition. (See the selected results reproduced in Table 2). Wright also suggests that the same pattern (i.e., the flip-flop in importance for GSC and IPC) was observed for the trait message acceptance correlations.

CRITIQUE OF WRIGHT (1975)

A close examination of Wright's findings suggests that several conceptual and methodological issues need to be reconsidered prior to the validation of GSC and IPC as information processing traits. These issues are:

(1)  Why did only the number of counterarguments classified by subjects as "original" evidence the hypothesized interactions with task involvement?

(2)  Is it appropriate to conceptualize GSC and IPC as "chronic" or enduring information processing traits and measure them in the manner used in this study?

TABLE 1

GENERAL SOCIAL CONFIDENCE AND INFORMATION PROCESSING CONFIDENCE MEASURES (WRIGHT 1975)

General Social Confidence

1. I do not spend much time worrying about what people think of me.

2. It doesn't bother me to have to enter a room where people are gathered and already talking.

3. When confronted by a group of strangers, my reaction is always one of shyness and inferiority.

4. I am sure someday my companions will look up to me and respect me.

5. I have frequently given up doing something because I thought too little of my ability.

6. I feel capable of handling myself in most social situations.

7. I don't make a very favorable first impression on people.

8. I often fear my actions will cause other people to have a poor opinion of me.

9. In group discussions I usually fear my opinions are inferior.

10. I am not at all lacking in self confidence.

Information Processing Confidence

1. I have more trouble concentrating than most people.

2. I am able to solve riddles and puzzles rapidly.

3. My mind seems to work slowly compared to those around me.

4. I am totally confident about my ability to judge messages coming from the mass media.

5. I am certainly able to think quickly.

6. When I hear an argument being presented, I am quick to spot the weaknesses in it.

7. I usually have to stop and think for awhile before making up my mind even in unimportant matters.

8. My thoughts frequently race ahead faster than I can speak them.

9. I am never at a loss for words.

10. I don't seem to be very quick-witted.

-----------------------------

Consistency of Results and Theory

The question of why only the number of counterarguments coded by subjects as being "original" counterarguments were related to personality traits can initially be discussed at a conceptual level. That is, one might have reasonably expected that the total number of counterarguments generated, irrespective of origin should have been related to these information processing traits. Indeed, as Wright suggests, prior theoretical discussions (Cohen 1958; McGuire 1969; Bither and Wright 1973) all imply that a negative linear relationship with counterargumentation and GSC and IPC should be expected. However these discussions do not address expected differences between "original" and other counterarguments' relationship to processing traits. Regarding this issue, the question also arises as to why the total number of thoughts irrespective of nature (i.e., counter or supportive) and origin should not be related to these information processing traits. Although Wright's premise suggests that individuals might hold a range of attitudes toward the content of a persuasive communication, existing theoretical and empirical evidence appears weak as to why the number of "original" counterarguments should be correlated more strongly than other cognitive response types with these personality traits. One plausible explanation for the resultant relationship between GSC, IPC and the number of "original" counterarguments may be the experimental procedure employed by Wright.

Methodological Considerations

This section of the paper attempts to address three important methodological issues in Wright's research:

(1)  The appropriateness of the data collection order.

(2)  The problem of interpretational confounding.

(3)  The measurement issue concerning the validity of the GSC and IPC scales.

TABLE 2

SELECTED WITHIN CELL CORRELATIONS OF GSC AND IPC WITH RESPECT TO MEASURES (WRIGHT 1975)

Order of Data Collection.  In Wright's study the subjects reported their thoughts directly following their viewing of the ad. They were then given the task of indicating the "origin" of each thought (e.g., "Was this an idea or reaction which reflected a point presented in the ad message or was this one of your own?") and then given the IPC and GSC measures. Since one normally conceives of personality traits as being CAUSAL determinants of cognitive responses, the order of measurement for these variables should have been consistent with this conceptualization.

In all likelihood the experimental treatment which subjects had just experienced created a mental framework or set which influenced their responses on the IPC and GSC measures. It appears evident that subjects who have just experienced a task which called upon them to assess whether or not their "thoughts" were their "own" (rather than a reflection of the speaker's) should respond consistently when asked questions regarding their skills in a social confrontation situation (i.e., How well they can spot a weakness in an argument?). That is, subjects who have indicated that each counterargument was "original" (and therefore have a large number of original counterarguments), should express a high degree of "social confidence."

It should also be recognized that despite Wright's claim that the GSC/IPC-message acceptance measures exhibit the same pattern of results, a close examination of these correlations shows that only the "acceptance (index)" measure shows the hypothesized reversal in importance which parallels the GSC/IPC-"original" counterargument pattern. (See Table 2). It is apparent that this index should be closely related to the number of original counterargument cognitive response measure, since these responses contribute to the calculation of this index (i.e., S" SIZE="2w_iSA_i - Sw_jCA_j - Sw_kSD_k). Note also that this index includes an additional perceptual influence stemming from the thought rating task (i.e., the subjective weighted importance ratings) which might also affect the subsequent responses-on IPC and GSC (e.g., "I rated my original thoughts as very important and therefore I guess I am "not at all lacking in self confidence.''). The consistent relationships between both of these dependent variables (and not the others) with the GSC and IPC measures strongly suggests that the thoughts judging tasks affected the subsequent responses on the GSC and IPC measures. In summary, the experimental procedure employed by Wright offers an alternative explanation for the consistent relationships he observed.

So far this critique has argued that the likelihood of a carry-over effect (between the thought judging, i.e., classification of "originality" and "importance" and trait measures) was quite high. One question still remains as to why this effect only occurred in those particular conditions. The next portion of this paper will address this issue.

If one accepts Wright's argument that the high involvement-audio message is representative of a strainful and overt social influence situation (i.e., the uncontrolled message delivered by another person only once, after which the subject expects to make an important decision), then it seems clear that subjects would be experiencing a task which creates strong arousal potential. In this social comparison processing environment it appears likely that subjects' ratings of the number of "original" counterarguments and GSC would necessarily be correlated. On the other hand, consistent with Wright's hypotheses, IPC should not strongly re-lace to the number of "original" CA as the information-processing task orientation is subsumed by the pervasive social factors.

Compared with the high involvement treatment, the low involvement conditions exemplifies a less social and more task-oriented situation. Message receivers therefore place less emphasis on cognitive resistance activities inherent in the high involvement decision mode. The lack of personal decision making permits the information processing task environment to become more salient. Hence, the subjects' self-reports of IPC and number of "original" counterarguments are related in the low involvement condition. General Social Confidence is not a strongly related factor in this processing environment since the strong task orientation and lack of the personal decision mode diffuse the strength of a social orientation.

In summary, it appears that Wright's manipulations were effective in creating the desired processing environments. However, because the thought coding and importance weighting tasks immediately preceded the trait measurement, it appears that the conclusions regarding a "trait-affecting-cognitive response" causal relationship are unfounded.

The Issue of Interpretational Confounding. Figure 2 represents Wright's nomological network for the proposed General Social Confidence--"original" counterargument (i.e., GSC_"original" CA) and General Social Confidence--Weighted message acceptance index (i.e., GSC _ Wt'd Acceptance Index) relationships. A similar nomological network exists and could be replicated for the IPC construct.

The model illustrates why one would necessarily expect significant correlations between the weighted message "acceptance (index)" and the GSC--"original" counterargument constructs. Because the number of "original" counterarguments has been utilized to operationalize both CA and the message acceptance index we have encountered a problem of interpretational confounding (Bagozzi 1980, p. 153). This general problem of ascertaining the meaning of theoretical variables i.e., interpretational confounding, "occurs as the assignment of empirical meaning to an unobservable variable which is other than the meaning assigned to it by an individual a priori to estimating unknown parameters"(Burt 1976, p.,4).

In Wright's model it appears that the g₂ relationships is strongly influenced by the g₁ relationship. Evidence for this premise can be obtained by examining the expected covariances i₁ and i₂ and by noting the temporal order of measurement.

FIGURE 2

WRIGHT'S NOMOLOGICAL NETWORK

The expected covariation between total number of counter-arguments (i.e., #CA) and the number of "original" counter-arguments (i.e., # original CA), (Noted as i₁), implies that the CA construct and weighted message acceptance construct are not conceptually distinct. The fact that other measures of message acceptance did not yield significant relationships with GSC/IPC provides additional credence for this premise.

The logic for the expected i₂ covariance is quite similar to that of i₁. Additionally, because the # original CA and the importance weights observables represent self-reported measures subject to the earlier suggested biases of carryover effects and arousal potential, these measures may share a considerable amount of covariance (i.e., i₂). A stronger test of theory would be to demonstrate a significant relationship between GSC and the weighted acceptance index while controlling for # original CA.

One additional rationale for the case of interpretational confounding involves the temporal order of measurement. As previously discussed, this ordering suggests that the causal relations g₁ and g₂ were actually reversed in the experimental setting. A final methodological issue concerns the validity of the GSC and IPC scales.

Factor Analysis of GSC/IPC

Measures of GSC and IPC were collected in two separate investigations by these researchers. In the first study (Swasy 1980) measures of GSC and IPC were collected for comparison with cognitive responses. Several aspects of this study closely resembled Wright's approach. These aspects included the use of an uncontrolled persuasive message (i.e., a videotaped sales presentation), similar sample of subjects (i.e., mothers, who were volunteer subjects and church group affiliates, N = 249), and who had a relatively high level of involvement with the topic (i.e., children's books). One important difference between this investigation and Wright's approach however, is that the GSC and IPC measures were collected one week prior to the experiment. Earlier arguments regarding possible carryover effects in Wright's design are eliminated with these pre-measures of GSC and IPC. No relationships between GSC and IPC with message acceptance or frequency of any cognitive responses were found. Subsequent orthogonal and oblique factor analyses of the GSC and IPC scales revealed that they were not unidimensional.

To reassess these findings a second investigation was conducted. A random sample (N = 135) of undergraduate college students completed a questionnaire which contained a number of attitude scales and the GSC and IPC measures. Factor analyses again revealed non-unidimensionality.

At a conceptual level, the oblique factor structure illustrates some interesting dimensions underlying the IPC and GSC scales. (See Tables 3 and 4). An assessment of these factors suggests that GSC contains both social and information processing subfactors. Although this is consistent with Wright's notion of the GSC construct, an examination of the IPC scales introduces some confusion. The IPC scale also appears to contain aspects of a "social" comparison (e.g., "I have more trouble concentrating than most people"). Although these factor analysis results are not a conclusive test of the scales' validity, they do suggest the need for a more systematic approach toward measuring individual differences.

One aspect of the scales' validity which should be addressed is the social desirability of the responses to these measures. In the survey of undergraduate college students, mentioned above, the Crowne/Marlowe Social Desirability Scale (1964) was also administered. For this sample, the GSC-SD correlation was .4449 (p=.001) and the IPC-SD correlation was .2094 (p=.007). These findings provide additional evidence of the weakness of the GSC and IPC measures and again suggest the possibility that a repertoire of subjective tasks such as judging the originality and importance of verbalized thoughts may also contain a high degree of social desirability bias which may be shared with the GSC/IPC scales.

TABLE 3

OBLIQUE FACTOR ANALYSIS - GENERAL SOCIAL CONFIDENCE

TABLE 4

OBLIQUE FACTOR ANALYSIS - INFORMATION PROCESSING CONFIDENCE

DISCUSSION

This paper has reviewed Wright's (1975) findings concerning the role of information processing traits influencing cognitive resistance to advertising. At a conceptual level, an examination of the methodological procedures suggests that the experimental design affected the GSC and IPC measures. A thorough examination of the pattern of results in the study also suggests that the thought coding and judgement tasks created a response set which carried over to the GSC and IPC measures. Finally, examination of the GSC and IPC scales in two separate investigations showed that the scales may not be unidimensional.

Although "chronic" traits as conceptualized by Wright represent an important determinant of cognitive responses and message acceptance, work by Epstein (1979) suggests that multiple measures and multiple observations yield more meaningful and consistent relationships between trait and behavioral variables. This approach is necessarily more demanding of the research, however an interactionist perspective appears to necessitate such rigor.

In summary, for the study of GSC/IPC future research should:

(1)  reassess the dimensionality of the GSC/IPC scales via factor analysis;

(2)  address the GSC/IPC-social desirability issue for evidence of possible format changes in GSC/IPC scales (e.g., forced choice responses);

(3)  investigate the social desirability aspects of thought reporting and coding tasks; and

(4)  following an interactionist's position:

(a)  measure traits over several occasions to the experimental situation, and

(b)  observe cognitive responses and attitude change in several stressful information processing situations. These multiple measurements and observations over several similar situations will provide a stronger test of the cognitive mediators of persuasion.

REFERENCES

Alker, H. A. (1972), "Is Personality Situationally Specific or Intrapsychically Consistent," Journal of Personality, 40, 1-16.

Bagozzi, Richard P. (1980), Causal Models in Marketing, New York: John Wiley and Sons, Inc.

Bern, Daryl J. and Allen, Andrea (1974), "On Predicting Some of the People Some of the Time: The Search for Cross-Situational Consistencies in Behavior," Psychological Review, 81, 506-20.

Bither, Stewart W. and Wright, Peter (1973), "The Self Confidence - Advertising Response Relationship: A Function of Situational Distraction," Journal of Marketing Research, 10, 146-52.

Bowers, K. S. (1973), "Situationism in Psychology: An Analysis and A Critique," Psychological Review, 80, 307-36.

Burt, R. S. (1976), "Interpretational Confounding of Unobserved Variables in Structural Equation Models," Sociological Methods end Research, 5, 3-52.

Cacioppo, John T., Sandman, C. A. and Walker, B. B. (1978), "The Effect of Operant Heart Rate Conditioning on Cognitive Elaboration and Attitude Change," Psychophysiology, 15, 330-8.

Cohen, Arthur R. (1959), "Some Implications of Self-Esteem for Social Influence," in Personality and Persuasibllity, eds. Carl I. Hovland and Irving L. Janis, New Haven: Yale University Press, 102-20.

Eagley, Alice H. (1980), "Recipient Characteristics as Determinants of Responses to Persuasion," in Cognitive Responses to Persuasion," ed. T. C. Brock, New York: McGraw-Hill.

Edell, Julie and Mitchell, Andrew (1978), "An Information Processing Approach to Cognitive Responses," in Research Frontiers in Marketing, ed. S. Jain, Chicago: American Marketing Association, 178-83.

Endler, N. S. (1966), "Estimating Variance Components from Mean Squares for Random and Mixed Effects Analysis of Variance Models," Perceptual and Motor Skills, 22, 559-70.

Endler, N. S. and Hunt, J. McV. (1968), "S-R Inventories of Hostility and Comparisons of the Proportion of Variance from Persons, Responses, and Situations for Hostility and Anxiousness," Journal of Personality and Social Psychology, 9, 309-15.

Endler, N. S. and Hunt, J. McV. (1969), "Generalizability of Contributions from Sources of Variance in the S-R Inventories of Anxiousness," Journal of Personality, 37, 1-24.

Endler, N. S. and Hunt, J. McV. and Rossenstein, A. J. (1962), "An S-R Inventory of Anxiousness," Psychological Monographs, 76, (17, Whole No. 526).

Ekehammer, B. (1974), "Interactionism in Personality from a Historical Perspective," Psychological Bulletin, 81, 1026-48.

Epstein, S. (1979), "The Stability of Behavior: I. On Predicting Most of the People Much of the Time," Journal of Personality and Social Psychology, 37, 1097-1126.

Greenwald, Anthony G. (196g), "Cognitive Learning, Cognitive Response to Persuasion, and Attitude Change," in Psychological Foundations of Attitudes, eds. A. G. Greenwald, T. C. Brock, and T. M. Ostrom, New York: Academic Press.

Jones, E. E. and Nisbett, R. E. (1971), The Actor and Observer: Divergent Perceptions of the Cases of Behavior, New York: General Learning Press.

McGuire, William J. (1969), "Personality and Susceptibility to Social Influence," in Handbook of Personality Theory and Research, eds. E. F. Borgatta and W. W. Lambert, Chicago: Rand McNally.

Mischel, W. (1968), "Toward a Cognitive Social Learning Reconceptualization of Personality," Psychological Review, 80, 252-83.

Osterhouse, Robert A. and Brock, Timothy C. (1970), "Distraction Increases Yielding to Propaganda by Inhibiting Counterarguing," Journal of Personality and Social Psychology, 15, 344-58.

Petty, Richard E., Wells, Gary and Brock, Timothy (1976), "Distraction Can Enhance or Reduce Yielding to Propaganda: Thought Disruption vs. Effort Justification," Journal of Personality and Social Psychology, 34, 874-84.

Petty, Richard E. and Cacioppo, John (1979), "Issue Involvement Can Increase or Decrease Persuasion by Enhancing Message-Relevant Cognitive Responses," Journal of Personality and Social Psychology, 37, 1915-26.

Shweder, R. A. (1975), "How Relevant is An Individual Difference Theory of Personality?" Journal of Personality, 43, 455-84.

Swasy, J. L. (1980), "Conceptual and Modeling Issues in a Cognitive Response Perspective on Persuasion Processes," unpublished Ph.D. dissertation, University of California, Los Angeles.

Wright, Peter (1973), "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, 10, 53-67.

Wright, Peter (1975), "Factors Affecting Cognitive Resistance to Advertising," Journal of Consumer Research, 2, 60-7.

----------------------------------------