Message Reception and Cognitive Complexity

Jack J. Kasulis, University of Oklahoma
Gerald Zaltman, University of Pittsburgh
ABSTRACT - Message reception is a key element of communication effectiveness. This paper reports on two field experiments which examined whether advertising brochures formulated to be compatible with particular cognitive styles would result in greater message reception. Subjects for the study were teachers who had input in purchase decisions for their schools.
[ to cite ]:
Jack J. Kasulis and Gerald Zaltman (1977) ,"Message Reception and Cognitive Complexity", in NA - Advances in Consumer Research Volume 04, eds. William D. Perreault, Jr., Atlanta, GA : Association for Consumer Research, Pages: 93-97.

Advances in Consumer Research Volume 4, 1977   Pages 93-97

MESSAGE RECEPTION AND COGNITIVE COMPLEXITY

Jack J. Kasulis, University of Oklahoma

Gerald Zaltman, University of Pittsburgh

ABSTRACT -

Message reception is a key element of communication effectiveness. This paper reports on two field experiments which examined whether advertising brochures formulated to be compatible with particular cognitive styles would result in greater message reception. Subjects for the study were teachers who had input in purchase decisions for their schools.

INTRODUCTION

The effectiveness of advertising communication is a critical concern for advertisers. The discussion that follows focuses on communication effectiveness from the perspective of increasing receiver receptiveness to commercial messages. Generally accepted in the area of communication research is the notion of a process model' of persuasion (e.g., Lavidge and Steiner, 1961; McGuire, 1969; 1973). These models assert that in response to a message the receiver experiences a series of behavioral steps of persuasion. Each of the steps is necessary, but alone they are insufficient to allow the prescribed behavior to occur. The models differ but they commonly specify the behavioral elements of persuasion to include attention to the communication, comprehension of its content, persuasion to the message, and some overt behavior. This article is concerned with the attention and comprehension components of the models. When the attention and comprehension components are discussed concurrently, they will be referred to as reception.

It is accepted that message reception is an important prerequisite for inducing prescribed behavior. The value to advertisers and consumer educators of improving the receptiveness of their communications does not need elaboration. The literature has indicated that receptivity is a function of audience, message, source, situational, and channel characteristics. Each of these factors interact and determine how receptive a recipient is to a persuasive communication. Extensive research has focused on each of these factors (McGuire, 1973; Eagly and Chaiken, 1975; Sternthal and Craig, 1974; and Lin, 1973). However, in the area of audience characteristics, research devoted to cognitive structure, when operationalized as cognitive complexity, is noticeably absent.

COGNITIVE COMPLEXITY

Every day each of us is the receiver of numerous messages. It is widely accepted that humans are not passive recipients of the information, but are active participants in structuring this input. In other words, cognitive processes select or exclude, assimilate or distort, accept or reject, and retain or discard informational stimuli. Knowledge of these processes implies that predictions can be made of the way people cope with their environment and also may suggest opportunities for using these systematic tendencies. Thus, from an advertiser's and a consumer educator's point of view, the active nature of human information processing offers an opportunity for more effective utilization of resources available for communication programs.

Cognitive or conceptual complexity falls under the broad heading of cognitive styles. Witkin (1964) defined cognitive styles as the "characteristic, self consistent ways of functioning shown by the person in the cognitive sphere." (p. 172) The literature assumes that there are systematic patterns of response (styles) for handling informational stimuli. In a sense, it embodies the trait approach of personality theory, but at the same time does not exclude the possibility of situational or environmental factors altering these cognitive dispositions.

Implicit in the notion of cognitive styles is that, given objectively equivalent stimulus conditions, two people may manifest different response patterns depending on the way they function cognitively. One element said to affect an individual's response is one's cognitive complexity level. In perceiving one's environment, an extremely large number of attributes are available to distinguish items. However, only a small portion of these can be utilized at any one time. Consequently, with experience, individuals develop a set of dimensions which pertain to particular objects or events. The set of dimensions and the relationship between these attributes identifies the properties of an individual's cognitive structure. Put simply, the sophistication of one's cognitive structure is identified as one's conceptual complexity. Those who have a very extensive structure are categorized as being cognitively complex while those who have a fairly elementary structure are categorized as being cognitively simple. In actuality, cognitive structure is a continuum, believed to be normally distributed, with the extremes labeled as being cognitively complex or simple. In essence, conceptual structure mediates the input-output sequence. In this way, it may be regarded as an enabling mechanism. The greater the degree of conceptual complexity, the more information an individual utilizes. Gardiner (1972) has noted that an individual low in complexity "... uses little information in forming concepts, and has difficulty in developing alternative concepts of events," while one high in complexity "... moves away from this simplistic and rigid type of functioning and not only learns to use more information in forming concepts but also develops the capacity to conceptualize events in alternative ways." (p. 327)

Associated with the cognitive complexity levels are specific behavioral patterns. It was posited that these response modes would affect the receptivity of an advertisement. Consequently, attention to these dispositions when structuring a message was hypothesized to produce more effective advertisements for a particular cognitive structure than those developed without such consideration. This concern was believed to be especially critical for those low in conceptual complexity since it has been noted that they have a tendency to categorically exclude informational stimuli which does not specifically address their functional constructs (cf., Schroder et al., 1967; Gardiner, 1972). The manifestation of this is sometimes so apparent that such individuals have been referred to as being closed-minded. On the other hand, the possibility was considered that messages formulated for the cognitively simple would be less effective for the complex. Thus, the preceding suggests that a market should be segmented according to one's conceptual complexity with unique advertisements devised for each group.

METHODOLOGY

The research reported is viewed as being exploratory, with an initial attempt to determine whether communications formulated according to cognitive complexity criteria would result in greater message reception than- those designed-without Such criteria. The context for the study was that of educational products. The use of school teachers as consumers provided the advantages of greater cooperation in the study and facility of implementation. At the same time, it offered a field setting for the research which had authentic environmental conditions for purchase decisions to be made regarding existing educational products on the market by individuals who were responsible for having input into such decisions for their organizations.

Subjects

The subjects for the study consisted of primary and secondary-level school teachers obtained from a convenience sample of ninety-six designated schools in Illinois. The schools were primarily selected because of their availability (district approval was typically required), but an effort was made to ensure heterogeneity of community characteristics, ethnic backgrounds, and district wealth. Each teacher who, as indicated by the school principal, would influence the decision process for adopting one of the study's designated products was asked to voluntarily participate. By agreement with the principals, no attempt beyond the initial letter was made to obtain cooperation from the teachers. This procedure resulted in 540 usable subjects--a response rate of approximately sixty percent.

Products

Two products, representing different kinds of R & D outcomes and assessed by experts to be bona fide improvements over existing offerings, were selected for study. Both products had been on the market for a few years but knowledge about them was almost nonexistent.

Product A is a teacher development course which was designed to improve instructional skills without the need for the teacher to leave the school. The course consists of a self instructional process in which the teacher videotapes and then self evaluates sessions with a few pupils in which the teacher practiced skills explained in an instruction handbook.

Product B is a program which alters the traditional organization of schools by replacing one teacher autonomy in the classroom with team responsibility and replaces the typical grade system with large, nongraded groups of pupils who receive individualized instruction. Adoption of the program requires extensive in-school training.

Procedure

A questionnaire with an attached cover letter was placed in the teachers' mail boxes in the designated schools. The cover letter specified that permission had been granted to contact them to request their voluntary participation in the study. The letter further indicated that they were asked to complete the attached (pre-test) questionnaire and a subsequent one (post-test questionnaire) later in the school year. No mention was made that they would be receiving brochures in the mail pertaining to the products. The pre-test questionnaire was designed to measure product knowledge, adoption requirements, and cognitive complexity levels of the respondents.

Six weeks later, the respondents were mailed a brochure addressing one of the designated products using specified formats corresponding to their treatment group. All of the teachers within a particular school who were subjects for one of the products received the same pamphlets, thereby eliminating contamination of treatments because of teacher interaction. The brochures were of the caliber of those commonly used by commercial distributors of educational products.-

Two weeks after sending the first pamphlet, a duplicate brochure was sent to the teachers. However, in this instance, the pamphlet was mailed in an envelope and accompanied by a letter. The letters were formulated according to cognitive complexity criteria. In addition to indicating that the second brochure was a duplicate of the first and was only sent in case the original got mislaid, the letter briefly summarized the attributes of the specified product.

Two weeks later, the post-test questionnaire was administered. It was distributed in the same fashion as the earlier one. Among questions relating to other aspects of the study, it contained items designed to ascertain the extent of reception to the communications.

Cognitive Complexity Instruments

Upon completing the pre-test questionnaire, the respondents were categorized as being cognitively complex or simple. An abbreviated version of Tuckman's Interpersonal Topical Inventory (ITI) was used as the cognitive complexity measure (cf., Tuckman, 1966). The ITI instrument categorizes subjects into one of four cognitive systems from their answers to forced choice questions. Their responses are totaled for each system and compared to standardized scores for the appropriate population under study. Because of its length, it was necessary to use a subset of the instrument. Three of the test's six sections were selected as a result of a pilot test of school teachers in a graduate education class. Though in the pilot test the six sections generally correlated highly, the three sections which were the most consistent in categorizing the respondents were selected. However, the abbreviated version was no longer compatible with the standardization chart. Therefore in categorizing the respondents, the number of selections recorded for each system was multiplied by the respective system number and these figures were aggregated. A mean split was then done on these scores with those above the critical value listed as being cognitively complex and those below as cognitively simple. In this study, 295 respondents were categorized as being cognitively complex and 241 as cognitively simple.

Treatments

Subjects in the two complexity categories were randomly assigned to one of two treatments. The treatments consisted of the aforementioned brochures--each had unique formats which were designed according to cognitive complexity considerations. In other words, the analysis of the treatment will focus on the format or structure of the message and not on the specific appeals used. These structural differences were designed to be most effective for a given cognitive complexity level.

The literature indicates that the behavioral patterns of individuals who are low in cognitive complexity tend to stress similarities in stimuli, depend on external anchors, and prefer stimulus simplicity. On the other hand, their cognitive counterparts tend to emphasize dissimilarities, depend more on internal evaluations, and prefer a more complex environment. These differences were incorporated in the brochures and thereby were the basis for distinction between the two. Those theoretically designed to be appropriate for the cognitively simple are hereafter called concrete messages, and those for the cognitively complex, abstract messages.

Bieri et al. (1966) have concluded that complex individuals seem to be prepared for diversity in their environment, particularly for that which is conflictual or contradictory in nature. On the other hand, those low in complexity appear to be prone to perceive regularity. They prefer consistencies and recurring uniformity in stimuli. This is in keeping with studies on person perception where the cognitively simple tend to see greater similarities between themselves and others to the extent of making unwarranted assessments. In contrast, the complex appear to emphasize differences. They are more accurate in evaluating differences between themselves and others but not in perceiving similarities. (cf., Adams-Webber, 1973; Leventhal, 1957; Bieri, 1971).

The differences in behavior between the two complexity levels has been attributed to the discrepant abilities to discriminate stimuli. In other words, with a more sophisticated cognitive structure, the cognitively complex are more capable of noting relatively minor differences. However, conceptually simple people, because they possess a more limited structure, are less able to evaluate stimuli and therefore use known objects as referents. Since there is emphasis on similarities by those low in complexity, the concrete messages stressed the similarities with existing products, but one which contained additional improvements. On the other hand, the abstract brochures claimed the product to be totally new, different, and innovative.

Cognitively simple people have been characterized as anchoring their behavior to external conditions, while conversely, cognitively complex individuals anchor internally. At low complexity levels, individuals have more restrictive cognitive capabilities. Consequently, alternative resolutions or interpretations tend not to be as apparent to these people, thereby necessitating greater reliance on external evaluations. In contrast, complex people have more functional constructs and greater flexibility in examining the relationships between them. This gives the conceptually complex a potential for greater independence in the sense that they are less dependent on others in making evaluations. The literature indicates that those high in complexity are less compelled to conform to prevailing mores and authority figures than their cognitive counterparts. (cf., Gardiner, 1972; Streufert, 1966; Schroder et al., 1967). One study reported negative attitude change to information emanating from either peer or authority figures for the cognitively complex. (cf., Lundy and Berkowitz, 1957).

The treatments accommodated this by including references in the concrete messages to normative behavior, the opinions of their peers, and the involvement of prestigious institutions. The abstract brochures did not have this approach, but instead encouraged the utilization of internal evaluative mechanisms by recognizing the need for each to determine the personal value of the product.

Stimuli vary as to their degree of complexity. Bieri (1968) has defined stimulus structure as "... the number of events or alternatives present as well as the relation among these events." (p. 634) The literature indicates that complex individuals, by having a more complicated cognitive structure, tend to prefer stimuli with greater structure while simple people prefer less stimulus complexity. (cf., Tripodi and Bieri, 1964; Schroder, 1971; Scott, 1962; 1963).

Inferred from above is the value of using one-sided communications for the cognitively simple and two-sided messages for the complex. In other words, communications become more complex with an increase in the number of attributes presented and with the comprehensiveness to which each is addressed. Thus, a message containing inconsistent information is more complex than one which is purely one-sided. Moreover, Bieri et al. (1966) have reported that information totally congruent with existing beliefs may be considered "too pat" by the cognitively complex, and become intolerant of it. One could speculate that one-sided, but discrepant information might lead to similar results, thus indicating the need for two-sided communications for the complex.

Reception Measures

Four measures of reception were taken in this study. They were:

(1) the extent to which the respondent said that he was familiar with the product (Familiarity),

(2) the thoroughness with which the respondent indicated that he had read the brochure (Read),

(3) the degree to which the respondent was able to recall details of the brochure (Remember),  and

(4) the ability of the respondent to recognize specifics delineated in the brochure (Knowledge).

Familiarity is coded with four levels. The respondents were asked directly how familiar they were with the product and were given the options of not at all familiar, somewhat familiar, and very familiar. The not at all familiar response was further divided according to their answer to the previous question as to whether or not they recalled receiving the brochure, with the logic that a YES must mean that the subjects at least skimmed it to have known that they had received such material.

Read is also coded with four levels of response. The subjects were asked directly how completely he read the brochure. Response options included thoroughly, partially, just skimmed, and not at all.

The last of the four level measures is Remember. It was an unaided recall test in which the subjects were asked, in an open ended question, what they remembered most from the brochure. The responses were assigned to one of the four categories; not at all, vaguely, fairly clearly, and very clearly.

The final reception measure is Knowledge. This differs from above in that it is a test of aided recall. The respondents were presented with a number of statements about the product, some of which were false, and asked to indicate their opinion along a five-point Likert scale representing an agreement-disagreement continuum. The instructions were stated as mentioned because experience indicates among educational researchers that teachers are reluctant to answer questions that they perceive to be testing their ability to comprehend. In this study, the difference between opinion and knowledge was considered less critical because the subjects almost exclusively had no prior information about the products and the statements for these questions were almost purely informational. Additionally, the analysis used is of a relational nature and since there is no reason to assume that any subset of respondents should be more likely to respond in a particular fashion, the problem is likely to be moot.

The Hypothesis

Examined in this study was whether message reception would be increased if the structure of the brochures corresponded to the cognitive complexity of the respondent. Stated more formally, the basic hypothesis was: the greater the degree of message compatibility with cognitive style, the greater the degree of message reception. As indicated earlier, the particular operationalization of this hypothesis involving cognitive complexity and the consumer group used have not been previously reported on in the literature. The analysis which follows aggregates the two theoretically optimal treatments (a concrete message to a cognitively simple respondent and an abstract message to a cognitively complex respondent), and similarly, aggregates the two theoretically suboptimal treatments (a concrete message to a cognitively complex respondent and an abstract message to a. cognitively simple respondent). Inasmuch as there is ordinal data, the theoretically optimal and theoretically suboptimal treatments are analyzed with the Kruskal-Wallis Nonparametric Analysis of Variance (Siegel, pp. 184-193).

FINDINGS

The results of the field experiment are presented in Table 1. Included are the four dependent variables for both products and the corresponding number of usable responses, the H Statistic for the Kruskal-Wallis Non-parametric Analysis of Variance, and the probability designations. For each of the dependent variables in both experiments, the results were in the hypothesized direction though in some instances the findings were above generally accepted significance levels.

TABLE 1

KRUSKAL-WAILLS RANK ORDER ANALYSIS OF VARIANCE FOR THE FOUR RECEPTION VARIABLES

In the Product A experiment, acknowledged readership (Read), unaided recall of information in the brochures (Remember), and aided recall of the brochure information (Knowledge) were found to have statistically significant differences between the theoretically optimal and suboptimal treatments. In contrast, acknowledge acquaintance with the product (Familiarity) was not found to be statistically significant.

In the Product B experiment, Read and Knowledge were found to have statistically significant differences between the treatment groups. In contrast, Familiarity and Remember did not demonstrate statistically significant differences.

CONCLUSIONS AND IMPLICATIONS

The results of the field experiments reported in this paper are mixed. For Product A, three of the four criterion measures of reception were found to have significant differences between treatments. For Product B, two of the four dependent variables demonstrated significant differences in treatments (assuming a maximum significance level of .10). However, there are some points worth emphasizing which puts the research in its proper perspective in assessing its contribution.

The research reported was an exploratory effort. Previous to this study, the usage of cognitive complexity theory in the formulation of advertisements had not been reported in the literature. In fact, even in the non-advertising communication studies, there had been only one attempt to formulate communications according to cognitive complexity criteria (Hewitt, 1972). Instead, the literature has concentrated on recognizing the differences in conceptual styles and on factors which altered the information processing modes. Thus, this study was an initial attempt to operationalize this phenomenon managerially. Furthermore, the study was conducted in a real life purchase setting where external validity was a primary concern.

The mixed results could be attributed to a number of factors. In the first place, a general measure of cognitive complexity was used rather a specifically tailored instrument for a product setting. Development of such an instrument may prove productive.

Secondly, the multiple criterion measures may not have been sufficiently discriminating. Of the four measures used, Familiarity would seem the most suspect. Responses to the Familiarity question could have been confounded by past knowledge of the product on similar products. This initially was not thought to be a problem since both products were largely unknown in Illinois. However, it is possible that the products were sufficiently similar to other products for the respondent to feel familiar with these products. In fact, this is most likely to have occurred with Product B since some may have incorrectly equated it to be no more than the open classroom concept. Familiarity with Product B had the highest probability that its H statistic would have occurred by chance (.259).

Thirdly, the criterion measures may he viewed more as retention than reception. The post-test questionnaires were administered two weeks after the second brochure and four weeks after the first. During that time, there was opportunity for the message to decay in the memories of the respondents. Thus, the measures of effectiveness are more conservative in reporting the outcomes than liberal.

Lastly, the reader is reminded that this was a first attempt at formulating advertisements according to cognitive complexity criteria. With further experience, it is possible that the treatments could be made more effective.

In conclusion, it is emphasized that the results are mixed. However, there are factors which may have confounded these results. Furthermore, it is believed that the incidence of significant findings and the consistent pattern of the results being in the hypothesized direction are enough to warrant further exploration of this phenomenon.

REFERENCES

Jack R. Adams-Webber, "The Complexity of the Target as a Factor of Interpersonal Judgment," Social Behavior and Personality, 1 (1973), 35-8.

James Bieri, "Cognitive Complexity and Judgment of Inconsistent Information," in Robert F. Abelson, Elliot Aronson, William J. McGuire, Theodore N. Newcomb, Milton J. Rosenberg, and Percy T. Tannenbaum (eds.), Theories of Cognitive Consistency: A Source Book (Chicago: Rand McNally, 1968), 633-40.

James Bieri, "Cognitive Structures in Personality," in Harold M. Schroder and Paul Suedfeld (eds.), Personality Theory and Information Processing (New York: Ronald Press, 1971), 173-208.

James Bieri, Alvin L. Atkins, Scott Briar, Robert L. Leaman, Henry Miller, and Tony Tripodi, Clinical and Social Judgment: The Discrimination of Behavioral Information (New York: Wiley, 1966), 182-206.

Alice H. Eagley and Shelly Chaiken, "An Attribution Analysis of the Effect of Communicator Characteristics on Opinion Change: The Case of Communicator Attractiveness," Journal of Personality and Social Psychology, 32 (1975), 136-44.

Gareth S. Gardiner, "Complexity Training and Prejudice Reduction," Journal of Applied Social Psychology, 2 (1972), 326-42.

David Hewitt, "Conceptual Complexity, Environment Complexity, Communication Salience, and Attitude Change," European Journal of Social Psychology, 2 (1972), 285-305.

Robert J. Lavidge and Gary A. Steiner, "A Model for Predictive Measurements of Advertising Effectiveness," Journal of Marketing, 25 (1961), 59-62.

Howard Leventhal, "Cognitive Processes and Interpersonal Predictions," Journal of Abnormal Social Psychology, 55 (1957), 176-180.

Nan Lin, The Study of Human Communication (Chicago: Bobbs-Merrill, 1973).

Richard M. Lundy and Leonard Berkowitz, "Cognitive Complexity and Assimilative Projection in Attitude Change," Journal of Abnormal and Social Psychology, 55 (1957), 34-7.

William J. McGuire, "Persuasion, Resistance and Attitude Change," in Ithiel de Sola Pool and Wilbur Schramm (eds.), Handbook of Communication (Chicago: Rand McNally, 1973), 216-52.

William J. McGuire, "The Nature of Attitudes and Attitude Change," in Gardiner Lindzey and Elliot Aronson (eds.), Handbook of Social Psychology (Reading, Mass.: Addison-Wesley, 1969), 136-314.

Harold M. Schroder, "Conceptual Complexity and Personality Organization," in Harold M. Schroder and Paul Suedfeld (eds.), Personality Theory and Information Processing (New York: Ronald Press, 1971), 240-73.

Harold M. Schroder, Michael J. Driver, and Siegfried Streufert, Human Information Processing (New York: Holt, Rinehart and Winston, 1967).

William A. Scott, "Cognitive Complexity and Cognitive Balance," Sociometry, 25 (1962), 66-74.

William A. Scott, "Cognitive Complexity and Cognitive Flexibility," Sociometry, 26 (1963), 405-14.

Sidney Siegel, Nonparametric Statistics for the Behavioral Sciences (New York: McGraw-Hill, 1956), 184-193.

Brien Sternthal and C. Samuel Craig, "Fear Appeals: Revisited and Revisited," Journal of Consumer Research, 1 (1974), 22-34.

Siegfried Streufert, "Conceptual Structure, Communicator Importance and Interpersonal Attitudes toward Conforming and Deviant Group Members," Journal of Personality and Social Psychology, 4 (1966), 101-3.

Tony Tripodi and James Bieri, "Information Transmission in Clinical Judgments as a Function of Stimulus Dimensionality and Cognitive Complexity," Journal of Personality, 32 (1964), 119-37.

Bruce W. Tuckman, "Integrative Complexity: Its Measurement and Relation to Creativity," Educational and Psychological Measurement, 26 (1966), 369-82.

Herman A. Witkin, "Origins of Cognitive Style," in Constance Scheerer (ed.), Cognition Theory Research, Promise (New York: Harper and Row, 1964), 172-205.

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