• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Developing and validating a measure of cognitive complexity: The role of cognitive complexity in processing of health messages

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Zhanna Bagdasarov
Abstract:
The present three-part study investigated the links between cognitive complexity, message processing, and extremity of attitudes. The focus of the first two studies was to create and validate a self-report measure of cognitive complexity. The development of a reliable and easy to administer instrument will further clarify the investigation in the area of cognitive complexity and message processing. In Study 3, social judgment theory was utilized to provide a theoretical framework for examining the relationship between cognitive complexity, message processing and outcome variables. Study 3 used a repeated measure design with two weeks between pre- and posttests. Participants were randomly assigned to one of the eight different experimental groups, varying by topic (alcohol vs. sleep) and message evidence (narrative vs. statistical; two message replication each), with a control group. The project concluded with analysis of the performance of the new measure and the potential role of cognitive complexity in health message design. The results confirmed previous findings that cognitive complexity is not a static trait variable, but rather a function of the interaction of psychological, contextual, and environmental variables. In addition, results from this study indicated that although message evidence types do not have a direct impact on attitudes and behaviors, some do affect message perception variables. Implications and directions for future research are discussed. Keywords : alcohol, cognitive complexity, lack of sleep, message evidence type, message perceptions, narrative messages, range of opinions, social judgment theory, statistical messages, transportation theory.

vii TABLE OF CONTENTS ABSTRACT

ii ACKNOWLEDGMENTS

iv DEDICATION

vi LIST OF APPENDICES

viii LIST OF TABLES

xii LIST OF FIGURES

xvi Chapter 1 INTRODUCTION

1 Chapter 2 STUDY1

14 Chapter 3 STUDY 2

85 Chapter 4 STUDY 3

96 Chapter 5 OVERALL DISCUSSION

191 REFERENCES

201 APPENDICES

218 TABLES

278 FIGURES

336 CURRICULUM VITAE 342

viii LIST OF APPENDICES

Page Appendix A Cognitive Complexity Instrument, CCI (Study 1)

218 Appendix B Instructions for RCQ

219

Appendix C Coding RCQ

220

Appendix D Paragraph Completion Inventory (PCI)

225

Appendix E

Cognitive Development Measure

226 Appendix F

Need for Cognition Measure

229 Appendix G

Need to Evaluate Measure 231 Appendix H Measure of Openness

233 Appendix I Measure of Creativity

234 Appendix J Measure of Intellectual Complexity

235 Appendix K Measure of Complexity

236 Appendix L

Measure of Intellect

237 Appendix M Measure of Intellectual Breadth

238 Appendix N Measure of Ingenuity

239

ix Appendix O Measure of Smartness

240 Appendix P Measure of Argumentativeness

241 Appendix Q Measure of Intolerance to Ambiguity

243 Appendix R Measure of Cognitive Failures

244 Appendix S Measure of Quickness

246 Appendix T Measure of Sensation Seeking

247 Appendix U Measure of Self-Esteem

248 Appendix V Measure of Cognitive Complexity, CCI (Study 2)

249 Appendix W Narrative Message about Negative Consequences of Drinking (Version 1)

251 Appendix X Narrative Message about Negative Consequences of Drinking (Version 2 )

252 Appendix Y Statistical Message about Negative Consequences of Drinking (Version 1)

253 Appendix Z Statistical Message about Negative Consequences of Drinking (Version 2)

254 Appendix AA Narrative Message about Negative Consequences of Lack of Sleep (Version 1)

255

Appendix AB Narrative Message about Negative Consequences of Lack of Sleep (Version 2)

256

x Appendix AC Statistical Message about Negative Consequences of Lack of Sleep (Version 1)

257

Appendix AD Statistical Message about Negative Consequences of Lack of Sleep ( Version 2)

258

Appendix AE Control Message about Bridge Collapse

259

Appendix AF Measure of Cognitive Complexity (Study 3)

260

Appendix AE Measure of Attitudes toward Drinking

262

Appendix AF Measure of Attitudes toward Sleep

263

Appendix AI Measure of Opinions about Drinking

264

Appendix AJ Measure of Opinions about Sleep

265

Appendix AK Measure of Personal Involvement with Drinking (Self- and Others’ Experience)

266

Appendix AL Measure of Personal Involvement with Sleep (Self- and Others’ Experience)

267

Appendix AM Measure of Drinking Behavior (Study 3, Time 1)

268

Appendix AN Measure of Drinking Behavior (Study 3, Time 2)

269

Appendix AO Measure of Sleep Behavior (Study 3, Time 1)

270

Appendix AP Measure of Sleep Behavior (Study 3, Time 2) 271

xi Appendix AQ Measure of Behavioral Intention to Drink

272

Appendix AR Measure of Behavioral Intention to Sleep

273

Appendix AS Measure of Message Perception

274

Appendix AT Measure of Message Processing Effort

275

Appendix AU Measure of Message Recall

276

xii LIST OF TABLES

Page

Study 1

Table 1.1 Independent Sample t tests Results for Order Effect

278 Table 1.2 Item Loadings for Cognitive Complexity Instrument (CCI) by Dimensions

279 Table 1.3 Bivariate Zero Order Correlation Matrix for All Variables

280 Table 1.4 Zero Order Correlation Matrix between Measures of Cognitive Complexity by Dimensions and Other Variables

283 Table 1.5 Pattern of Associations between Measures of Cognitive Complexity an d Other Variables

286 Table 1.6 Bivariate Zero Order Correlation Matrix for Cognitive Complexity Measures, by Dimensions

288 Table 1.7 Independent Sample t tests Results for Education

289

Study 2

Table 2.1 Results of one-way ANOVA for Order Effects for CCI

290 Table 2.2 Cognitive Complexity Instrument Items by Dimensions

291 Table 2.3 Zero Order Correlation Matrix for All Variables

293

Study 3

Table 3.1 List of All Variables Measured at Time 1 and Time 2

294

xiii

Table 3.2 Cognitive Complexity Instrument Items by Dimensions

295 Table 3.3 Bivariate Zero Order Correlation Matrix for All Variables

297 Table 3.4 Results of Independent Sample t tests for Message Replication

300 Table 3.5 Results of Independent Sample t tests for Overall Attrition

302 Table 3.6 Bivariate Zero Order Correlation Matrix for Cognitive Complexity Measures by Dimensions

304 Table 3.7 ANCOVA for Attitudinal Change toward Drinking and Personal Involvement (self -experience)

305 Table 3.8 ANCOVA for Attitudinal Change toward Drinking and Personal Involvement (others’ experience)

306 Table 3.9 ANCOVA for Attitudinal Change toward Sleep and Personal Involvement (self -experience)

307 Table 3.10 ANCOVA for Attitudinal Change toward Sleep and Personal Involvement (others’ experience)

308 Table 3.11 ANCOVA for Drinking Behavior Change and Personal Involvement (self -experience)

309 Table 3.12 ANCOVA for Drinking Behavior Change and Personal Involvement (others’ experience)

310 Table 3.13 ANCOVA for Sleep Behavior Change and Personal Involvement (self -experience)

311

xiv Table 3.14 ANCOVA for Sleep Behavior Change and Personal Involvement (others’ experience)

312 Table 3.15 Results of Independent Sample t tests for Message Evidence and Attitudina l and Behavioral Changes

313 Table 3.16 Between-Subjects ANOVA for Attitudinal Change toward Drinking

314 Table 3.17 Cell Means for Attitudinal Change toward Drinking

315 Table 3.18 Between-Subjects ANOVA for Attitudinal Change toward Sleep

316 Table 3.19 Cell Means for Attitudinal Change toward Sleep

317 Table 3.20 Between-Subjects ANOVA for Drinking Behavior Change

318 Table 3.21 Cell Means for Drinking Behavior Change

319 Table 3.22 Between-Subjects ANOVA for Sleep Behavior Change

320 Table 3.23 Cell Means for Sleep Behavior Change

321 Table 3.24 Between-Subjects ANOVA for Message Realism

322 Table 3.25 Cell Means for Message Realism

323 Table 3.26 Between-Subjects ANOVA for Message Reflectiveness

324 Table 3.27 Cell Means for Message Reflectiveness

325

xv Table 3.28 Between-Subjects ANOVA for Message Believabiltiy

326 Table 3.29 Cell Means for Message Believability

327 Table 3.30 Between-Subjects ANOVA for Message Processing Effort

328 Table 3.31 Cell Means for Message Processing Effort

329 Table 3.32 Between-Subjects ANOVA for Message Recall

330 Table 3.33 Cell Means for Message Recall

331 Table 3.34 Paired-samples t tests Results for Control Group

332 Table 3.35 Independent sample t tests Results for Education

333 Table 3.36 Summary of Findings for All Hypotheses and Research Questions

334

xvi LIST OF FIGURES

Page Figure 1 Second-Order Confirmatory Factor Analysis for CCI

336 Figure 2 Interaction between Cognitive Complexity and Message Evidenc e for Sleep Attitudinal Change

337 Figure 3 Interaction between Topic and Message Evidence for Message Realism

338 Figure 4 Interaction between Topic and Message Evidence for Message Reflectiveness

339 Figure 5 Interaction between Topic and Message Evidence for Message Believability

340 Figure 6 Interaction between Topic and Message Evidence for Message Processing Effort

341

1

CHAPTER 1 INTRODUCTION The present series of studies investigates the link between people’s cognitive complexity and their attitudes. Although scholars acknowledge that most people vary in cognitive complexity, it is relatively unknown in what ways cognitive complexity affects perceptions and attitudes. Investigating cognitive complexity and attitudes toward particular issues, for example, attitudes toward lack of sleep and excessive drinking, will help shed light on these issues and allow message designers to create more effective persuasive campaigns. The present research consists of three separate studies. The first two studies involve the creation and validation of a self-report measure of cognitive complexity. Although there are several existing measures of cognitive complexity, each of them is either difficult to administer or involves time consuming coding, or both. In addition, these measures are often modified from one researcher to another, making any comparisons and generalizations across the studies difficult. A new measure that is reliable, valid and relatively easy to administer and analyze will have substantial theoretical and practical value. Such a measure will further clarify and advance investigations in the area of cognitive complexity and eliminate inconsistencies in the existing research. Developing a valid and easily administered measure of cognitive complexity has a substantial value for the communication field, as cognitive complexity underlies a diverse array of communication-related activities, including skills in social perception, message production, message reception, and social interaction (Burleson & Caplan, 1998).

2

The third study investigates the impact of cognitive complexity on message processing. It also examines if and how cognitive complexity moderates attitudes and behaviors in the persuasion process. The study seeks to understand if people with different levels of cognitive complexity process messages differently. In addition, it answers the questions if people with various levels of cognitive complexity are easier/harder to persuade and are more/less likely to change their attitudes toward some specific issues. In this study, issues of lack of sleep and excessive drinking among college students are utilized to provide domains for these tests. The knowledge gained from this research will help to create more effective messages by tailoring the information contained in persuasive messages to a person’s level of cognitive complexity, and by better predicting individuals’ response to the same persuasive message. Cognitive Complexity: An Overview Cognition is the mental activity that guides and underlines human action. Glass (2004), for example, defines human cognition as information processing, with the goal to perform effective actions in the world. One of the most common approaches to the study of human cognition is through the framework of information processing models (Lachman, Lachman, & Butterfield, 1979). The information-processing model has two major assumptions: first, it assumes that people are information processors, who are constantly engaged in processing perceived stimuli. The second assumption is that people’s ability to process information is limited because of the limited amount of mental resources needed to process information. Thus, although people can think of several things simultaneously, eventually all the mental resources will be used, and in order to think about one more issue the previous thought should be let go (Lang, 2000). Currently,

3

this information-processing framework is considered to be the most influential in the study of human cognition (e.g., Ericsson & Simon, 1981; Resnick & Resnick, 1992). During the past several decades the focus of research on human cognition has been on identifying variables that affect information processing. Communication researchers examined basic information processing differences between people and tried to identify variables predicting these differences (e.g., Petty & Cacioppo, 1986; Schroder, Driver, & Streufert, 1967) and develop theories explaining variations in message processing. In particular, they were trying to answer why individuals differ in how they perceive and assign meaning to messages. Starting with 1960s, one potential variable that received particular attention is cognitive complexity, which initially was introduced by Bieri (1955) as a personality trait and has subsequently been studied within the framework of personal construct theories (Kelly, 1955). Schroder et al. (1967) redefined cognitive complexity as a characteristic of information processing in the cognitive system. Over the years, many different approaches to the study of cognitive complexity have been explored. However, as yet, there is no agreed-upon definition of cognitive complexity. For some researchers, cognitive complexity is related to knowledge structures in a cognitive system and refers to the sophistication of those cognitive structures that are used for organizing and storing cognitive content (e.g., Curseu, Schruijer & Boros, 2007; Kelly, 1955). For others, cognitive complexity reflects the ability to be flexible and adaptive in information processing (e.g., Schroder et al., 1967). Other researchers, however, define cognitive complexity through concepts of simplicity and conceptualize individual differences in need for structure or closure as a desire for

4

simplicity (Neuberg & Newsom, 1993). The perspective of this research is that cognitive complexity is a variable describing the degree of elaboration of a social-cognitive system. Cognitive complexity elucidates how a person “filters and processes stimuli so that the environment takes on psychological meaning” (Goldstein & Blackman, 1978, p. 463). This approach is related to Kelly’s (1955) notion that people should be viewed as information processors who actively engage in organizing the stimuli impinging on them. The common thread in all these approaches is that cognitive complexity is concerned with the structure or organization of thinking rather than specific content (Suedfeld, 1971) or, in other words, how thinking is organized rather than information that is available. Cognitive complexity is a “close relative” of conceptual/integrative complexity construct (e.g., Suedfeld, Tetlock, & Streufert, 1992). According to this view, a relatively complex cognitive system consists of a comparatively large number of finely articulated, abstract, and well-integrated elements. Thus, cognitive complexity is a multidimensional phenomenon that “indexes the degree of differentiation, articulation, and integration within a cognitive system” (Burleson & Caplan, 1998, p. 233) and refers to the way a person conceptually organizes the environment. In other words, cognitive complexity is the ability to view people, objects, ideas and other entities in a multidimensional way. Individuals with more “developed” systems of interpersonal constructs have more differentiated (i.e., numerically larger), more abstract (i.e., more refined or specialized elements), and more integrated (i.e., more organized) construct systems. Such individuals are characterized as more complex (Burleson, Waltman & Samter, 1987).

5

Different levels of cognitive complexity are linked to different styles of thinking, reasoning, and mentally organizing the world and lead to different perspectives and approaches on a particular issue. That is, the same issue could be perceived as more complex or less complex for people at different ends of the cognitive complexity continuum. Conceptualization of Cognitive Complexity: State or Trait Variable? Despite active interest in cognitive complexity and research on how it affects communicative behavior, the nature of variation in cognitive complexity has not been apprehended. One of the important shortcomings of previous work on cognitive complexity is the failure of past researchers to fully conceptualize cognitive complexity. This limitation clearly hampered the development of valid and reliable measures of the cognitive complexity construct and contributed to the inability to distinguish it, conceptually and empirically, from other constructs. Overall, there are three important conceptualizations of cognitive complexity, and the key distinction between them is the presumed source, or cause, of variation of cognitive complexity among people. Some scholars conceptualize cognitive complexity as a trait variable (something endemic, that people are born with), others as a contextual variable (situation-dependent, that is activated only when people are motivated to do so), and still others view cognitive complexity as an environmental variable (a skill that could be acquired or taught; that is, a function of people’s education, professional experience, and other socio-demographic variables). However, none of these approaches separately is appropriate when trying to address such a complex phenomenon. A successful conceptualization of cognitive

6

complexity embodies and utilizes different parts of each of these approaches. A brief review of findings for each of these approaches will be presented next. Different Approaches to Cognitive Complexity Early conceptual complexity theory and conceptual systems theory viewed cognitive complexity as a stable personality variable (for a more complete overview, refer to Suedfeld et al., 1992). Research in that tradition demonstrated that systematic individual differences in cognitive complexity do, indeed, exist. For example, Burleson and Caplan (1998) suggested that cognitive complexity reflects individual differences in social information-processing capacity, but not a motivational orientation or predisposition (p. 240). However, other findings demonstrated that cognitive complexity could manifest itself in one situation and not in another (e.g., Schroder et al., 1967), and this could be explained by the absence of motivation for central/systematic processing. Thus, despite some evidence that individual traits could be best predictors of various cognitive styles, it was later acknowledged that situational and environmental factors could have some major impact on cognitive complexity (see Suedfeld et al., 1992). Streufert and Streufert (1978) viewed complexity as context-specific and found that most people are relatively simple in some situations and relatively complex in others. This assumption was based on the findings that level of complexity could be changed as a result of crisis, stress, and certain personality characteristics (e.g., Hunsberger, Lea, Pancer, Pratt, & McKenzie, 1992; Suedfeld & Bluck, 1988). Such findings indicated that a static trait model of cognitive complexity is inadequate. Subsequently, Suedfeld et al. (1992) proposed that although complexity may be a trait variable, it is not necessarily an unchangeable one. The underlying assumption here is that cognitive complexity is

7

activated when people are personally motivated to process information systematically and on a deeper level. Thus, most people (with some exceptions, such as people with developmental or learning disabilities) are more or less equally capable of engaging in cognitively complex ways. Some variations in cognitive complexity (both intra- and inter-personal) could be explained by varying degrees of personal motivation or vested interest rather than by the presence or absence of a certain trait or by the level of relevant cognitive skills. Findings that skill-based training could improve cognitive complexity (Little, Packman, Smaby, & Maddax, 2005) made the conceptualization of complexity even more complicated. Some studies demonstrated that exposure to different types of messages could influence social-cognitive structure and processes, including levels of cognitive complexity (e.g., Roloff & Berger, 1982; Samter, Burleson, & Basden-Murphy, 1989). These findings, however, could be explained from the conceptualization of complexity as a trait-like variable; it could be argued that certain personality traits may influence the extent to which a given level of cognitive complexity is sustained or enhanced. For example, Schroder et al. (1967) and later Streufert and Streufert (1978) proposed that different levels of threat, time pressure, and information load could affect levels of cognitive complexity, and that individual differences in cognitive complexity determine how people react to these changes in environmental variables. The present study is conceptually rooted in the notion that while the “hereditary” approach (when cognitive complexity is viewed as a genetic, or trait, predisposition) provides some insight on nature of cognitive complexity, conceptualizing cognitive complexity as a function of the interaction of psychological, contextual, and

8

environmental variables offers a better and more consistent explanation of the nature of cognitive complexity. This socio-psychological theoretical approach is in line with the interactive approach to cognitive complexity presented by Schroder et al. (1967). They suggested that organisms either inherit or develop certain modes of thinking and information processing, and acknowledged that level of information processing in a given area can vary across content areas. In other words, complexity could vary as a function of different forms of environmental stress (Schroder et al., 1967, p. 11). It could be said, then, that interaction of genes and environment affect people’s levels of cognitive complexity. To summarize, the present study employs an “interactive approach” in which cognitive complexity will be viewed as interplay of many factors and events, including genetics, predisposition, and environment. Cognitive Complexity: Content vs. Structure Rokeach (1960) warned that when studying the organization of belief systems, which are closely related to attitudes, it is important to investigate the structure rather than the content. He noted: “It is not so much what you believe that counts, but how you believe” (p. 6). Based on that notion, the present study measures cognitive complexity not by the type or amount of information known but rather as anchored in a socio- psychological theoretical framework. Typically, research on cognitive complexity focuses on the structure of thinking rather than on the content. Following that trend, Schroder et al. (1967) in their study of complexity “tried to divorce” measurement of cognitive complexity from content (Goldstein & Blackman, 1978, p. 473). They stated that cognitive complexity (or what they called low- and high- levels of information processing) should not be measured by

9

the amount of information known, but rather by the ability to generate new rule combinations. According to Schroder et al. (1967), “higher-level structures can function more effectively in situations that are undergoing change and in which new perspectives and solutions are required” (p. 10). They also proposed that people vary in their cognitive complexity and based on that criterion could be placed along a continuum ranging from cognitively simple to cognitively complex. On the one end of this continuum are people with lower levels of cognitive complexity who tend to think in a simple way about other people, issues, and events. They often form dichotomous (bad vs. good or right vs. wrong) impressions (Tetlock, Peterson, & Berry, 1993). On the other end of the continuum are people with higher levels of cognitive complexity who tend to recognize the multidimensionality and avoid categorization of the issues and events. People with higher cognitive complexity are able to generate and apply a variety of alternative interpretations and perspectives on a situation (Sotirovic, 2001). They perceive other people, events, and issues in relatively complex and personalized terms. They also tend to use many dimensions for the judging of events, make finer discriminations along these dimensions, and integrate the dimensions into meaningful conceptual wholes (Harvey, Hunt, & Schroeder, 1961). In considering the properties of cognitive complexity, it is important to keep this continuum in mind. It should be acknowledged that people cannot be classified simply as cognitively complex or cognitively simple, and that having these extreme categories is just a convenience often employed in communication, social psychology, and related research.

10

Cognitive Complexity, Perceptions, and Attitudes In one of the earliest studies investigating the concept of cognitive complexity, Schroder et al. (1967) proposed that how a person thinks about a particular issue often determines what the person thinks. They also suggested that people who score higher on measures of cognitive complexity hold less extreme attitudes and judgments, and those people who score lower on cognitive complexity hold more extreme attitudes and judgments. Similarly, Witkin, Oltman, Raskin, and Karp (1971) postulated that individual differences in perception are reflections of their cognitive complexity that, in turn, could be correlated to their behavior. Research demonstrated that cognitively complex people consider and take into account others’ perspectives (e.g., Hale & Delia, 1976). Cognitively complex people tend to be more open to new information, rely on their own integrative efforts, seek more novel information, search across more categories of information, and are less externally information bound. They tend to take in more information and form more well-rounded impressions than less complex individuals. There is an association between cognitive complexity and person-centered communication skills (Burleson & Caplan, 1998). The person-centered communicator forms more complex impressions of others that enable a more adaptive interaction (Woods, 1998). Cognitively complex people are more likely to understand the perspectives of others, generate more possible explanations for others’ behavior, and feel more empathy toward others. In addition, higher cognitive complexity is related to one’s ability to tolerate inconsistency (Goldstein & Blackman, 1978) and greater cognitive flexibility (Scott, 1962). Generally, it is asserted that those who have higher levels of cognitive complexity tend to generate messages that reflect perspectives

11

of others (Applegate, 1982) and produce more sophisticated messages in a variety of situations, including persuasive situations (Applegate & Woods, 1991). Linville and Jones (1980) investigated if there is a link between cognitive complexity and extremity of attitudes. They found that cognitively complex people tend to perceive more complexity in others and judge them more moderately. Earlier, White and Harvey (1965) found that cognitively simple people are more likely to respond with more extreme scores on a judgment scale than cognitively complex people. There is also some evidence that while people with lower complexity are usually attracted only to each other (based on having similar attitudes), people with higher cognitive complexity could be attracted to both people with higher cognitive complexity as well as to people with lower cognitive complexity (Streufert & Swezey, 1986). It should be acknowledged here that while people with lower cognitive complexity could be taught certain attitudes or a complex set of distinctions for a specific context, people with higher levels of cognitive complexity are very flexible in developing new attitudes and new distinctions in novel and unusual situations. People higher in cognitive complexity are able to analyze a situation into many constituent elements and then explore connections and potential relationships among the elements; they are multidimensional in their thinking. Summary From the review of literature on cognitive complexity, it could be concluded that examining cognitive complexity is important to communication research and persuasive communication in particular. There are several contexts in which the importance of this variable could be especially apparent, and some of these contexts are reviewed next. One

12

critical question relates to the nature of the cognitive complexity construct and the mechanisms which both cause and are affected by it. For example, cognitive complexity could be important in the context of information seeking because it causes people to selectively expose themselves to certain information and certain types of media consumption but not others (Sotirovich, 2001, 2003). Next, through its effect on how people process information, cognitive complexity could serve as a moderator of the relationship between exposure to persuasive information and attitude change (or other attitudinal-based outcomes). Cognitive complexity could also mediate the influence of exposure on attitude change, wherein exposure influences cognitive complexity (through learning and/or activation), and the change in cognitive complexity results in attitudinal response (for example, the extremity of the position held). In addition, cognitive complexity could influence attitudes directly, independent of exposure, by leading people to assimilate information that is consistent with their current attitude, and reject, or contrast attitude-incongruent information. Finally, cognitive complexity could be viewed as a cause of both information processing and attitudinal outcomes, that is, people selectively expose themselves to different types of information, and this information affects their attitudes. The effects of cognitive complexity could be even more complex if there are some indirect effects on information processing (for example, if cognitive complexity influences information seeking behavior and/or motivates people to process a particular type of information). Although this study was not designed to test all of these propositions regarding cognitive complexity, some inferences will be drawn and discussed based on conclusions.

Full document contains 359 pages
Abstract: The present three-part study investigated the links between cognitive complexity, message processing, and extremity of attitudes. The focus of the first two studies was to create and validate a self-report measure of cognitive complexity. The development of a reliable and easy to administer instrument will further clarify the investigation in the area of cognitive complexity and message processing. In Study 3, social judgment theory was utilized to provide a theoretical framework for examining the relationship between cognitive complexity, message processing and outcome variables. Study 3 used a repeated measure design with two weeks between pre- and posttests. Participants were randomly assigned to one of the eight different experimental groups, varying by topic (alcohol vs. sleep) and message evidence (narrative vs. statistical; two message replication each), with a control group. The project concluded with analysis of the performance of the new measure and the potential role of cognitive complexity in health message design. The results confirmed previous findings that cognitive complexity is not a static trait variable, but rather a function of the interaction of psychological, contextual, and environmental variables. In addition, results from this study indicated that although message evidence types do not have a direct impact on attitudes and behaviors, some do affect message perception variables. Implications and directions for future research are discussed. Keywords : alcohol, cognitive complexity, lack of sleep, message evidence type, message perceptions, narrative messages, range of opinions, social judgment theory, statistical messages, transportation theory.