Consumer Response to Marketing Stimuli: the Relationship Between Affect, Cognition, and Behavior

Carol Pluzinski, The University of Michigan [Doctoral Candidate, The University of Michigan, The graduate School of Business Administration, Ann Arbor, Michigan 48109.]
William J. Qualls, The University of Michigan [Assistant Professor of Marketing, The University of Michigan, The graduate School of Business Administration, Ann Arbor, Michigan 48109.]
ABSTRACT - This paper presents a dynamic framework for consumer response models that focuses on 1) the relationship between the constructs of interest (i.e., cognition, affect, behavior), and 2) the systemic properties of such an organization. The framework is an attempt to reconcile at the theoretical level the seemingly incompatible research positions currently held at the empirical level of analysis.
[ to cite ]:
Carol Pluzinski and William J. Qualls (1986) ,"Consumer Response to Marketing Stimuli: the Relationship Between Affect, Cognition, and Behavior", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 231-234.

Advances in Consumer Research Volume 13, 1986      Pages 231-234

CONSUMER RESPONSE TO MARKETING STIMULI: THE RELATIONSHIP BETWEEN AFFECT, COGNITION, AND BEHAVIOR

Carol Pluzinski, The University of Michigan [Doctoral Candidate, The University of Michigan, The graduate School of Business Administration, Ann Arbor, Michigan 48109.]

William J. Qualls, The University of Michigan [Assistant Professor of Marketing, The University of Michigan, The graduate School of Business Administration, Ann Arbor, Michigan 48109.]

ABSTRACT -

This paper presents a dynamic framework for consumer response models that focuses on 1) the relationship between the constructs of interest (i.e., cognition, affect, behavior), and 2) the systemic properties of such an organization. The framework is an attempt to reconcile at the theoretical level the seemingly incompatible research positions currently held at the empirical level of analysis.

INTRODUCTION

The tripartite construal of psychological experience has dominated consumer research for more than three decades. Based upon three components of cognition (thinking), affect (feeling), and conation (behavior) this conceptualization has formed the foundation of several consumer behavior and communication response models used in marketing (Colley 1961, Lavidge and Steiner 1961, Krugman 1962, Howard and Sheth 1969, Ray 1973, Smith and Swinyard 1982). While most researchers will agree with a tripartite division of consumer psychological experience, an on-going debate regarding the order in which the three components are arranged continues to surface in the research literature (Zajonc and Markus 1982, Van Raaij 1984, Sternthal 1984).

We propose that this research problem cannot be resolved by remaining at the empirical level of analysis. Rather we must now consider the theoretical meaning of the disparate empirical results by abstracting the essence of the problem to a higher level of analysis. a is requires framing the problem conceptually rather than empirically. Therefore a more dynamic framework will be presented under which previous motels can be subsumed, and which focuses on the nature of the relationship between the components rather than addressing the perfunctory question of temporal order.

This relationship will be described in terms of three general properties: iteration, parallelism, and complementarily. These properties are actually processes which will serve to define the boundaries of the framework. The present paper presents a framework which argues for a process orientation, rather than a outcome oriented model of consumer response.

First, a brief overview of consumer response models which addresses the difficulty in establishing their validity at the empirical level will be presented. Next an alternative, more global conceptualization of cognition, affect and conation will be developed, the rationale for which will be provided by the New Physics. Finally, the implications of such a framework for consumer research will be discussed.

CONSUMER RESPONSE MODELS

Traditionally, most models of consumer behavior have assumed some awareness or cognitive state prior to an affective or behavioral response (Engel and Blackwell 1982, Howard and Sheth 1969, Lavidge and Steiner 1961, Ray 1973, 1982). This cognitive dominated paradigm of consumer response holds that response order is based upon COGNITION--- >AFFECT - >CONATION. However, many researchers have begun to question the appropriateness of such an approach for understanding all types of consumer behavior (Kassarjian 1978, Olshavsky and Granbois 1983. Isen 1984).

Alternatively, a growing body of research argues and presents evidence suggesting that AFFECT, which is activated first, influences thought processes (COGNITION) which subsequently direct an individual's behavior (CONATION) (Zajonc and Markus 1982, Van Raaij 1984). Finally a third stream of research contends that behavior precedes both affect and cognition. This implies that individuals often exhibit behavior in response to marketing stimuli without actually thinking or considering their feelings (Nord and Peter 1980). Although empirical support exists for each of the alternative configurations, researchers continue to debate the "correct order of response (particularly with regard to the affect/cognition primacy question).

Yet further progress or resolution is inhibited because the proponents of each response sequence ordering subscribe to vastly different paradigms, or world views, which cannot be reconciled at the empirical level (Kuhn 1962). Perhaps the investigation of the relationship between these components would be more fruitful. Hence the study of the nature of the interaction process among the three components requires a higher level of analysis and a broader framework from which to understand the relationships.

Consumer response models suffer from the following problems which make them seem incompatible at the empirical level. First, the models conceptualized and tested in the literature are essentially variations of the same hierarchy-of-effects process (Colley 1961, Lavidge and Steiner 1961, Krugman 1965, Ray 1973). In this way they are unidirectional and static, reflecting a serial processing bias which implies that consumers are passive rather than active in response to their environment. Second, the temporal relationship of cognition, affect, and conation has never been fully explicated. As such, the dominance of the observed effect may be an artifact of the experimental design. Details regarding this point will be presented later in the paper.

Most models have operated under a cognitive activated paradigm; that is, cognition serves as the primary activator and moderator in the consumer response process. Recently, such cognitive models have come under attack and have been severely criticized for failing to recognize a more important role for both affect and behavior. Finally, an additional weakness with respect to consumer response models arises as a result of past research failing to develop standardized conceptualizations and operationalizations of the models' components. As a result researchers must still resolve such issues as whether or not there is a difference between recall and comprehension, and whether behavioral intentions or executed actions should be employed as the measure of behavior. Even more confusing has been the variation in the manner in which affect has been conceptualized, both as an attitude (which itself is comprised of affect, cognition, and conation, c.f. Fishbein and Azjen 1975) and as an evaluative judgment or a feeling state.

To help shed some light on these issues, we suggest a more global framework of consumer response incorporating systemic properties of how the components are related, as well as the theoretical boundaries which set the limits of the process. m us the framework attempts to explicate the components' boundaries, as well as to provide a more definitive conceptualization of the relationships between the model's components. This framework allows iterative steps as well as parallel processing, alleviating the dependence upon a sequential single component activated paradigm. After providing philosophic support, the remainder of this paper will elaborate on the characteristics and systemic properties of the proposed framework. Further justification for such a framework will be based upon evidence that has been presented in the literature.

In Support of Broader Framework

Since affect, cognition, and behavioral intention are unobservable variables, we may do well to follow the methodology of the New Physics (c. f . Zukav 1979). For example, subatomic particles are not only unobservable, they are not really particles at all in a material sense; and in this way they are much like the latent constructs we analyze as consumer researchers. The New Physics recognizes the subjective nature of our perceptual experience, as well as the notion that we cannot observe something without changing it. The problem of how to measure a particle is an illustrative example. A subatomic particle can be characterized by both position and momentum, but the two properties cannot be accurately measured at the same time. We can choose to measure both, but in doing so we sacrifice accuracy; or we can choose to measure one property precisely, at the expense of knowing nothing about the other (Heisenberg's Uncertainty Principle).

Zukav (1979) explains that because it is the nature of things that we can know either the momentum of a particle or its position, but not both, and because we must choose which of the properties we want to measure, we are actually creating certain properties by the very ace of choosing to measure them. Similarly, the measurement of a consumer's response is influenced by the point at which the researcher chooses to close the system and determine the initial point of response. If reality is defined by the nature of- the researcher's experimental design, then the component order which results becomes a mere experimental artifact. For example, if we define cognition as including the prerequisite processing stages of transducing sense data into information, then by definition cognition precedes an affective response and behavior. On the other hand, if affect is defined as a primary visceral response that we experience before we consciously know how to label it, then affect precedes cognition and behavior.

In order to execute an empirical investigation we must close the system, holding some variables constant while others are chosen to vary. us reality is dictated by methodology and our results have meaning only within the specific system for which they were constructed. If we accept the argument that temporal order is experimenter-induced, we are now left with the task of "making sense" of the data. Olson and Peter (1984) suggest that researchers must question the theoretical meaning of their results. In this way we move from a narrow empirical level to a broader conceptual level of analysis. Because concrete effects (at the empirical level) have very little intrinsic meaning, we should seek to develop abstract theoretical meanings so as to enhance the generality of those meanings and thus their usefulness (Olson and Peter 1984). Thus the framework presented here is consistent with contemporary philosophic views in that it accounts for the subjectivity of empirical findings, yet provides a unifying conceptual analysis in order to promote a consistent research program.

A DYNAMIC COMMUNICATION RESPONSE FRAMEWORK

The framework proposed here differs from earlier models in that the process of consumer response can be activated within the system at one or multiple points, and move throughout the tripartite process in either direction. We propose that three processes best describe the way in which the (cognitive affective and conative) response components are related to each other. In particular, the framework which is proposed postulates that iterative, parallel, and complementary processes function alternately in linking the response model's components. An individual does not necessarily have to complete connections with each of the components, but in fact may just cross between two components in an iterative process. For example, when deliberating about a product decision, a consumer may cycle between a thinking state (cognition) and a feeling state (affect). This may be the typical for consumers who cannot make a decision because they are torn between what they consider rational behavior and the way they actually feel (i.e., the household that needs a family car although the husband really wants a Corvette).

FIGURE 1

A DYNAMIC COMMUNICATION RESPONSE FRAMEWORK

Conversely, two components may simultaneously activate a third component as the result of parallel processing that has occurred. Such is the case of a purchaser whose thoughts (cognition) place him/her behind the wheel of the Corvette while he/she is physically in the process of signing the purchase contract (behavior), which brings a joyous smile to his/her face (affect).

Similarly, two components can complement each other in their impact upon a third component. For instance, the strength of our feelings (affect) may determine how we think (cognition); how we think may strengthen how we feel, thereby inducing a stronger behavioral response. These scenarios presented above represent only three of the possible ways in which affect, cognition, and conation may be related. Examples such as these clearly point out the need for a more robust framework if researchers are to understand and explain the dynamic phenomena of consumer response to marketing stimuli.

Theorizing does not take place in a vacuum. Rather our theories, methods, and results are based upon unstated assumptions about how we think the world works. As argued by Bristor (1984) these underlying assumptions are often inadvertently incommensurate, and so meaning becomes obscured. Thus, in order to help clarify the nature of the construct order debate, we will first discuss the theoretical boundaries of the three constructs.

Cognition, Affect, d Conation

Over the years, the components of the tripartite model have taken on varying definitions, with affect being the most ambiguously defined. Affect has been conceptualized variously as feeling, emotion, preference, and attitude. Interestingly, a dominant school of thought regarding attitude uses the entire tripartite, defining attitude as a combination of affect, cognition, and conation (Fishbein and Azjen 1975). In the present framework, affect is considered as evoking a form of experiential feeling or emotional response. Affect is intended to represent a more global measure of evaluation and is not meant to reflect specific affective responses. Recent research into the conceptualization of affect suggests that it may be composed of primarily perceptual reactions to external stimuli occurring without awareness or interpretation (Leventhal 1979, 1980, Zajonc 1980, Zajonc and Markus 1982). Advertisers have long held that, at the very least, advertising messages impact the way consumers feel rather than the way they think or behave.

Conversely, cognition has been construed as thoughts, beliefs, or any of a number of brain-related activities associated with perception, memory, and/or learning. While earlier communication response models have made a distinction between conscious (attentive) and preconscious (perceptual pre-attentive) cognitive activity, the present framework accommodates both conscious and preconscious cognitions.

The third component, conation or behavior, is possibly the easiest to understand as it is defined as purposive action. A classic example is operant conditioning which suggests that the probability that a behavioral response will occur which can be influenced by reinforcement directly following the behavior I c.f. Skinner 1966). With respect to consumer responses, the behavior of interest is typically purchase and decision behavior. While these constructs have been shown or hypothesized to be related via a consumer response model of behavior, a unifying framework has failed to emerge.

Systemic Properties

As noted above, three properties which influence the relationships in the consumer response process include: 1) iteration, 2) parallelism, and 3) complementarily. These properties serve as rules of correspondence and as the theoretical boundaries for the relationships between affect, cognition, and conation. Based on a more holistic approach, these systemic properties define the interaction between the consumer response components (interconstruct), as well as the processing that might occur within each of the constructs (intraconstruct). Hence, when a property occurs between affect, cognition, and/or conation, it represents an interconstruct process. When a property occurs within any of the constructs, it represents an intraconstruct process. Each of these properties is discussed in turn below.

Iterative Processing. We define an iterative property as one which allows for repetitive sequential processing. For example, a script (Abelson, 1976) may be thought of as a sequence of repetitive interaction between expectations (cognitions) of behavior as well as the actual execution of behavior (interconstruct). Processing within the cognitive component (cognitive intraconstruct) is evidenced by the very nature of consumer memory itself as it is based on the notion of hierarchical interaction among the various structural forms of knowledge (c.f. Anderson and Bower 1973; Klatzky 1980, Loftus and Loftus 1976). Research in the area of information processing and long-term memory provides an excellent example of how iterative properties of cognition might relate to consumer responses. Bettman (1979) contends that consumers use information stored in memory to aid in perceiving and translating incoming marketing stimuli. Described as a process by which semantic concepts, visual images, and auditory messages are associated by way of a series of interactions, it has been suggested that such a mapping process forms the basis of all cognitive activity (Holyoak and Gick 1980).

Parallel Processing. Parallel processing assumes that two or more constructs may be activated simultaneously. mat is, separate processes may occur at the same time. For example, Zajonc (1980) argues that one can have an initial emotional response to a stimulus which is paralleled by an associated behavioral (motor) response. With respect to consumer response behavior, affect and cognition may act parallel to each other without impacting either upon each other or upon the behavior. For example when a consumer hears a particular jingle or responds to a particular advertising stimulus (Mcdonald's Arches), the stimulus serves as a cue for memory retrieval, while at the same time evoking the particular feeling or mood which was present when the stimulus was originally stored. We have all remembered some event and the feelings we had which are triggered by a favorite song or melody. Thus parallel processing offers another solution to the continuing debate as to whether affect precedes cognition, i.e., they may occur simultaneously.

Perceptual processing (cognitive intraconstruct) of ten occurs in parallel. When executing a visual search during a pattern recognition task, an advertising stimulus could be compared to many internal perceptual codes at the same time. Bettman (1979) proposes that within a choice task, the decision maker works through several different goals, not necessarily one at a time. He states, "although goals may have some natural sequencing, progress can often be made on several goals simultaneously" (p. 21). If researchers accept the argument that affective as well as cognitive operations are performed when one acquires new information, it would appear that they both could occur simultaneously, through parallel processing.

Complementary Processing. Complementarily refers to the activation of two constructs as one means to the same end. With complementary processing, one phenomena serves to strengthen the other in its impact upon a third. For example affect and cognition may act as complements to each other to then influence behavior. Empirical evidence in a series of studies by Bower (1980) demonstrates that the affective mood state at the time of encoding may influence behavior only in conjunction with information retrieved from memory.

An example of the property of complementarily in action may be found in today's innovative advertising research. When consumers are asked to view television advertising commercials, multiple measurements are made of the perceptions one affective and the other cognitive. The cognitive measure may consist of a series of aided or unaided recall and recognition tests, while the affective measure may be based on a galvanic skin-response test or pupil dilation test. In response to the ad, cognitions serve to strengthen the affective response or vice-versa.

Intraconstruct complementarily may be illustrated with an example based upon the nature of memory. It is now generally accepted that there exists two conceptually distinct types of memory--episodic and semantic (Loftus and Loftus 1976, Klatzky 1980). Tulving (1972) was the first to propose a distinction between episodic and semantic memory. Loftus and Loftus summarize this difference as: "episodic memory refers to our memory for personal experiences and their temporal relations, whereas semantic memory includes the organized knowledge we have about wards and other verbal symbols, their meaning, and referents; about relationships among them; and about rules for manipulating them" (p. 120). m us even though they are distinct, it seems clear that episodic and semantic memory muse work in synergy under the auspices of the same overall memory structure and information processing system. Interestingly, this distinction between semantic and episodic memory may be of import for our understanding the relationship between cognitive and affective processes. Zajonc suggests that it may be possible to "store" affect in some way (Zajonc 1980, Zajonc and Markus 1982). This would require that affect possess some mnestic properties. Yet, rather than ascribing to affect the properties of a totally independent processing system as Zajonc suggests, perhaps it would be better to view affect and cognition as being associated with different types of memory. Isen (1984) provides justification for this view by arguing that "affect is an experience it feels like an experience, and its role in memory feels like that of place or time, which are also experiential" (p. 224).

SUMMARY

This paper has attempted to describe a broader, more dynamic, framework of consumer response to marketing stimuli. The framework focuses on the functional relationships that may exist between its constituent constructs of cognition, affect, and behavior. Three systemic properties are proposed as rules of correspondence between these constructs, and the implications of these relationships associated with both interconstruct and intraconstruct processing have been discussed. Moreover, the framework emphasizes the need to abstract the current empirical problem to the theoretical level of analysis.

Justified by epistemic assumptions from the New Physics, the present conceptualization provides an organizing framework which may serve to: 1) imbue more theoretical meaning to existing empirical observations; 2) provide a foundation for a consistent program of research.

REFERENCES

Abelson, R. P. (1976), "Script Processing in Attitude Formation and Decision Making," Cognition and Social Behavior, Hillsdale, NJ: L. Erlbaum, pp. 33-45.

Anderson, J. R. and Bower, G. H. (1973), Human Associative Memory, Washington, D.C.: V.H. Winston and Sons.

Bettman, J. R. (1979), An Information Processing Theory of Consumer Choice, Addison-Wesley.

Bower, G. H. (1981), "Mood and Memory," American Psychologist, 36, 129-148.

Bristor, J. (1984), "Organizing Consumer Behavior: A Paradigmatic Perspective," In Anderson, P. R. and Ryan, M. J. (Eds.), AMA Winter Educators' Conference: Scientific Method in Marketing, 177-196.

Colley, R. (1961), Defining Advertising Goals For Measured Advertising Results, New York: Association of National Advertisers.

Engel, J. and Blackwell, R. (1982), Consumer Behavior, New York: Dryden Press.

Fishbein, M. A. and Azjen, I. (1975), Belief, Attitude, Intention and Behavior: An Introduction to TheorY and Research, Reading, MA: Addison-Wesley.

Holyoak, R. J. and Gick, M. (1980), "Analogical Problems Solving," Cognitive Psychology, 12, 306-355.

Holyoak, R. J., and Gordon, P. D. (1984), "Information Processing in Social Cognition,- in R. S. Wyer, T. R. Srull, and J. Hartwick (Eds.) Handbook of Social Cognition, Hillsdale, NJ: Erlbaum.

Howard, J. and Sheth, J. (1969), The Theory of Buyer Behavior, New York: Wiley.

Isen, A. (1984), "Toward Understanding The Role of Affect in Cognition," in R. S. Wyer, T. R. Srull, and J. Hartwick (Eds.), Handbook of Social Cognition, Hillsdale, NJ: Erlbaum.

Kassarjian, H. H. (1981), "Low Involvement: A Second Look," in K. B. Monroe (Ed.), Advances in Consumer Research, Vol. 8, Ann Arbor: Association for Consumer Research. 31-34.

Klatzky, R. L. (1980), Human Memory (2nd ed.) San Francisco: W. H. Freeman & Company.

Krugman, H. E. (1965), "The Impact of Television Advertising: Learning Without Involvement," Public Opinion Quarterly, 29, 349-356.

Kuhn, Thomas S. (1970), The Structure of Scientific Revolutions, Chicago, IL: University of Chicago Press.

Lavidge, R. and Steiner, G. A. (1961), -A Model for Predictive Measures of Advertising Effectiveness," Journal of Marketing, 25, 59-62.

Leventhal, H. W. (1979), -A Perceptual Motor Processing Model of Emotion," in R. Pliner, K. R. Blankenstein, and I. M. Spigel (Eds.), Advances in the Study of Communication and Affect, 5, New York: Plenum.

Leventhal, H. W. (1980), " Toward a Comprehensive Theory of Emotion," in L. Berkowitz (Ed.), Advances in Experimental Social Psychology, 13, New York: Academic, 139-207.

Loftus, G. R. and Loftus, E. F. (1976), Human Memory: m e Processing of Information, Hillsdale, NJ: Lawrence Erlbaum Associates.

Nord, W. R. and Peter, J. P. (1980), -A Behavior Modification Perspective on Marketing, Journal of Marketing, 44, 36-47.

Olshavsky, R. W. and Granbois, Donald H. (1979), "Consumer Decision Making-Fact or Fiction?" Journal of Consumer Research, 6, ( September), 93-100.

Olson, J. C. and Peter, J. P. (1984), "External Validity?" in Anderson, P. F. and Ryan, M. J. (Eds.), AMA Winter Educators' Conference: Scientific Method in Marketing, 81-84.

Ray, Michael (1973), "Marketing Communication and the Hierarchy of Effects," in Clarke (Ed.), New Models for Mass Communication Research, Sage Publications, 147-176.

Ray, Michael (1982), Advertising and Communication Management, Englewood Cliffs: Prentice-Hall Inc.

Skinner, B. F. (1966), Behavior of Organisms: An Experimental Analysis, New York: Appleton.

Smith, R. E. and Swinyard, W. R. (1982), "Information Response Models: An Integrated Approach," Journal of Marketing, 46, 81-93.

Sternthal, B. (1984), "Comment on the Role of Affect and Cognition," Marketing Science Institute Working Paper, no. 84-111.

Tulving, E. (1972), -Episodic & Semantic Memory, in E. Tulving and W. Donaldson (eds.) Organization of Memory, New YorK: Academic Press.

Van Raaij, W. F. (1984), "Affective and Cognitive Reactions to Advertising," Marketing Science Institute Working Paper, no. 84-111.

Zajonc, R. B. (1968), "The Additudinal Effects of Mere Exposure," Journal of Personality and Social Psychology Monograph SupPlement, 9, (2 pt. 2), 1-27.

Zajonc, R. B. (1980), "Feeling and Thinking: Preferences Need No Inferences," American Psychologist, 35, 151-175.

Zajonc, R. B. and Markus, H. (1982), "Affective and Cognitive Factors in Preferences," Journal of Consumer Research, 9, 123-131.

Zukav, G. (1979), The Dancing Wu Li Masters: An Overview of the New Physics, New York: William Morrow and Company, Inc.

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