A Commentary on New Theoretical Perspectives on Consumer Behaviour

David W. Stewart, University of Southern California
ABSTRACT - Research on consumer behavior has tended to focus on a narrow range of phenomena. Although the range of phenomena has expanded in recent years and new approaches for the study of consumer behavior have been developed, the full richness of consumer behavior has yet to be realized. Three papers suggest that consumer researchers broaden their agenda by focusing on aspects of consume behavior that have not been widely investigated. Although each paper suggests a different focal point, they are complementary. A model of consumer choice is used to show how the approaches suggested in three paper can be integrated with other research in the field. The model also helps identify other dimensions of consumer behavior that have not been widely researched.
[ to cite ]:
David W. Stewart (1990) ,"A Commentary on New Theoretical Perspectives on Consumer Behaviour", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 750-754.

Advances in Consumer Research Volume 17, 1990      Pages 750-754


David W. Stewart, University of Southern California


Research on consumer behavior has tended to focus on a narrow range of phenomena. Although the range of phenomena has expanded in recent years and new approaches for the study of consumer behavior have been developed, the full richness of consumer behavior has yet to be realized. Three papers suggest that consumer researchers broaden their agenda by focusing on aspects of consume behavior that have not been widely investigated. Although each paper suggests a different focal point, they are complementary. A model of consumer choice is used to show how the approaches suggested in three paper can be integrated with other research in the field. The model also helps identify other dimensions of consumer behavior that have not been widely researched.


Human decision making is the nexus of the study of consumer behavior. The factors that lead to the need for a decision, the processes by which decisions are made, and the development of decision strategies through experience are the elements from which we attempt to build an understanding of consumer behavior. Consumer researchers have also examined a wide range of other factors, life-style, attitudes, value systems, personality, information processing, cultural differences, and so forth. There is always at least an implicit recognition that these factors are related to decision making in some way, and we frequently find empirical demonstrations that differences in decision outcomes are related to these other factors. Despite many years of research on consumer behavior, the papers we have heard today (Cherian and Harris 1989, Duncan 1989, and Rassuli and Harrell 1989) remind us that much remains to be done before we have a full understanding of consumer behavior. Further, these papers remind us that there are dimensions of consumer behavior that have not been well explored and that our traditional views of the processes of consumer choice may not always be appropriate for all types of consumers and situations in which we find consumers. Let us consider what each of these papers suggests.


Duncan (1989) reminds us that consumers have a host of beliefs that influence consumer behavior. These beliefs have an impact on information search and shopping behavior, on decision rules, and even on problem recognition. Beliefs have not been ignored entirely by consumer researchers, but much of the research on beliefs has been in the context of models of attitudes where the focus of study has most often been beliefs about product attributes. Beliefs about other aspects of the market place such as sales personnel, ability as a shopper, usefulness of search activity, and so forth have received more limited attention. Yet, those studies that have examined the impact of more general consumer belief systems have consistently found them to have a very strong influence on consumer behavior.


Rassuli and Harrell (1989) remind us that the assumptions about the direction of causality in consumer behavior may be overly simplistic. The notion that the end point of consumer behavior is purchase overlooks the fact that purchase may, on many occasions, actually be more appropriately regarded as the beginning point of consumer behavior. In fact, the notions of beginning and end in an on-going system involving reciprocal causality are not particularly meaningful. In such a system the points of entry and exit are dictated more by the purpose of the researcher than the nature of the process itself. The frequent focus on purchase as the end of the process is the result the fact that most previous and current consumer research is carried out in a marketing context that is specifically oriented toward influencing purchase behavior. This does not mean that this is the best way to view consumer behavior, even when the purpose of the research is to inform marketing action. Starting the process at different points may yield some interesting and useful insights that could not be obtained by always looking at purchase as the end of the process. For example, one might look at the purchase process as the beginning of a process that gives rise to the identification of new beliefs or new problems for solution. In fact, if we assume some degree of purposefulness on the part of consumers, purchase is neither the beginning nor the end of this purposefulness. It is only a means to an end that falls somewhere in between the identification of a problem or need and its solution. This, of course, raises the question of whether consumers are always purposeful.


Cherian and Harris (1989) offer the intriguing notion that consumers are not fundamentally purposive. Rather, they suggest that the verbalizing component of the brain serves to impute purposefulness to all activities even though many activities are not, in actuality, purposeful. If by purposeful Cherian and Harris mean consciously purposeful, rational, and able to articulate purpose, then I would certainly agree. There are of course other meanings of purposefulness and I believe that all life forms are purposeful in at least some sense of the term. Rather than quibble with terminology, however, I would rather focus on the implications of the view of Cherian and Harris. First, I would note that their view is not new, nor is it necessarily irreconcilable with the broad view of consumer rationality. Nisbett and Wilson (1977) review evidence that makes a compelling case for the fact that people often do not know the reasons for their own behavior, though they also often feel obliged to make up reasons when pressed. More recently, Etzioni (1986) has suggested that people may make a rational decision to behave irrationally in situations in which the costs of rational behavior may be too high given the anticipated consequences In these types of situations an individual may still feel compelled to provide an explanation for behavior that is more rational rather than admit that they have decided not to be rational. It is also quite likely that in many such cases individuals are not even aware that they made a decision to be irrational because the decision to be irrational is so automated. In fact, we know that a great deal of decision making and behavior is automated and defies simple explanation.

Cherian and Harris (1989) suggest that because people impute purposiveness when none actually exists, self reports should not be trusted. I would modify their conclusion and suggest that self reports should not be relied upon exclusively. Neither should "experiments" that are constructed to produce outcomes consistent or inconsistent with the rational model (a number of researchers in consumer behavior appear to be sufficiently clever to produce all sorts of odd behavior through skillful manipulation of the decision environment, but this also tells us little about how consumers actually behave). I am equally uncomfortable with the suggestion of Cherian and Harris (1989) that because consumers do not know what they are actually doing, and why, it is better to have a researcher interpret the intent of a consumer. There is a logical fallacy in the assumption that someone who is incapable of understanding his or her own behavior is better able to understand someone else's behavior.

What I really think all this means is that we need multiple approaches to the study of consumer behavior. I also think these multiple approaches should be applied by the same researchers. Rather than one group of researchers doing experiments, another doing naturalist observation, and still another using self reports, the discipline would profit from having the same phenomenon investigated by several means. Thus, self reports might be complemented by observation and interpretation. "Interpretations" might be checked against self-reports or against the results of carefully designed experiments. Experimental results might be checked against the subjective perceptions of both the subjects of the experiment and other knowledgeable judges. This would make it somewhat more time consuming to do research, and require facility with numerous approaches to research, but it might lead us to a more complete understanding of consumer behavior.


Although I agree with many of the conclusions of the authors of the three papers we have heard today, I am concerned that each of the approaches we have heard is yet another opportunity for the fragmentation of the field of consumer research. Consumer researchers have borrowed quite liberally from other disciplines, but most of this borrowing has involved theories, concepts, or methods that are appropriate for a narrow class of phenomena. The more general phenomenon of consumer behavior gets lost as a result. While I do not think it is yet time for the development of new grand theories of consumer behavior, I do think it is time that we began to place specific studies, approaches, and theories in the more general context of consumer behavior.

At the risk of being accused of introducing yet another flow chart into a discipline that is already replete with them, I wish to try to illustrate what I mean by such integration using the three approaches suggested today. At the same time, I hope to point out some additional areas of neglect in the study of consumer behavior. Figure 1 is a simplified diagram of consumer decision making. Notice that there is no obvious beginning or ending.-- For purposes of illustration, let us enter the process in Box A-Product Environment Sampling. When a consumer recognizes a particular need, there is a search for a solution. Much of the consumer information processing literature has focussed on this aspect of the decision process and we have identified a wide range of factors that may influence the acquisition of information-the number of alternatives available, the number of attributes associated with each alternative, the spatial and temporal pattern of information, and so forth. The role of beliefs in determining the extent of this sampling has also received some attention, but far less than some of the characteristics of the information environment itself.

Social judgement theory and related theoretical perspectives have provided some understanding of this part of the process, particularly when the quality of information is not perfect and there is redundancy of information. Research on beliefs about the search process itself, sources of information, and the utility of particular types of information would also tell us a great deal about this type of consumer behavior. In fact, conceptualizing the process as a sampling process facilitates thinking about all of the determinants of the process. It also directs the focus away from the information environment itself and toward the sampling decisions being made by the consumer.

Ultimately information obtained in the search process may be used in a given decision rule (Box B). Decision rules have been the province of normative and behavioral decision theory, and more recently multiple criteria decision theory. Let me suggest that there have been few real efforts to link what goes on in Box A with what goes on in Box B. There must be a relationship and it is not so simple as more information leads to more complex decision rules. It is easy to find examples of extensive information search followed by the use of a simple heuristic, though it may be more difficult to find examples of little search followed by more complex rules. On the other hand, we do know that decisions can be automated with experience. More experienced consumers tend to make global, brand based decisions and do relatively little information search. This process of automation (represented in Box G) is not well understood though information integration theory provides a means for modeling the process.



Whatever the decision rule employed, the decision may or may not be followed by an action that implements the decision (Box C in Figure 1). The impact of situational factors which may prevent implementation of a particular decision is well documented. What is less well studied is the impact such factors may have on future search strategies and/or decision rules. For example, does a strategy or rule that frequently produces decisions that the consumer is unable to implement give way to other rules? In which situations are the decision rules actually used by a consumer known to the consumer and in what situations do the decision rules remain hidden from the consumer?.

Post-purchase evaluation is well documented and a number of factors that influence it have been identified. I would like to suggest, however, that there are really two distinct types of post-purchase evaluation. The first, represented in Box D, I have called outcome verification. In this process the consumer merely verifies that an outcome sought has been obtained-the aspirin relieved the headache, the soft drink quenched thirst, the life insurance policy provided a sense of security. There is no question about whether the outcome is desirable or not. The evaluation of the desirability of the outcome has been made at some previous point in time and the consumer needs only to verify that the expected outcome did in fact occur, or did not occur. In other situations it may not be so clear whether an outcome is desirable. This is particularly likely when the outcome has not been experience before. Consider the first time you tasted beer or a pickle. You probably had to decide whether you even liked the taste (and that may have required several trials). This facet of the decision process has largely been neglected by all theories of decision making, yet it is at the very heart of the process. Some of the factors that may influence this process are listed in Box H, but we still don't understand preference formation. Motivation has not been a hot issue in consumer psychology for a long time. While I am not suggesting a wholesale return to motivation research, this is a dimension of consumer decision making that has not received its due attention.

For some products this experiential dimension may be all that is important to the choice process, and this has important implications for information search and for the selection of a decision rule. Part of information search may be experiential, involving product trial and outcome verification (the chair "feels" comfortable, the tennis racket or golf club "feels" right in one's hand, the drink was "refreshing"). This type of evaluation is distinctly nonverbal and has received scant attention. Our theories of decision making have tended to be largely cognitive, and it is not clear how experiential outcomes or attributes fit. This is perfectly consistent with the perspectives articulated by Rassuli and Harrell (1989) and by Cherian and Harris (1989).

Attribution theory suggests that consumers seek a theory to guide their behavior in the future. Such theories represent belief systems. These belief systems have also received some attention from consumer researchers, but more needs to be done. In particular, the development of these belief systems over time would be a particularly fruitful avenue for research. For example, having experienced a given outcome associated with a product, the consumer may attempt to identify distinguishing characteristics of the product"Products with x produce y". This is depicted in Box F along with factors that have been shown to influence this process. The distinguishing characteristics identified by the consumer are likely to become more prominent in search activities and decision rules in the future. In effect these factors are believed to be important. Social judgment theory would also suggest that some situations provide better distinguishing information than others, though the lack of a product characteristic that clearly differentiates superior from inferior products does not preclude the consumer from inventing one. In time, these differentiating characteristics may be abbreviated or combined, making it difficult to sort out the original distinguishing attributes. Indeed surrogate attributes may be used that have no relationship to the original benefits sought or experienced.


I have now walked through the decision process, indicating where various approaches appear to make their contribution and where gaps seem to exist. I would note one other problem in studying consumer decision making. We almost never begin with a naive consumer. We study consumers who generally have a very long history, so we are looking at a well developed process rather than a developmental one. This tends to introduce some significant biases into our way of thinking about consumer decision. A little reflection will suggest the nature of those biases-the decision process will tend to be more abbreviated and more cognitive than would be the case for a truly naive consumer. This is because purchase decisions tend to become more automated with experience and because the experiential aspects of consumption (and the outcome evaluation process) tend to be subordinated to verbal surrogates with experience.


I believe we have much to learn about consumer decision making. We can advance our knowledge quickly by doing two things. First, we must integrate the multiple perspectives on the phenomenon and identify the linkages among them. This also means bring. multiple methodologies to bear on the study of the same events and processes. Second, we must look at the development of consumer behavior over time through well constructed longitudinal studies. This work is long overdue, and it is time to get on with the task.


Cherian, Joseph and Barbara Harris (1989), "Capricious Consumption and the Social Brain Theory: Why Consumers Seem Purposive Even in the Absence of Purpose," Advances in Consumer Research (Provo, Utah: Association for Consumer Research).

Duncan, Calvin P. (1989), "Consumer Market Beliefs: A Review of the Literature and an Agenda for Future Research," Advances in Consumer Research (Provo, Utah: Association for Consumer Research).

Etzioni, A. (1986), "Rationality is Anti-Entropic," Journal of Economic Psychology, 7, 17-36.

Nisbett, R. E. and T. D. Wilson (1977), "Telling More than We Can Know: Verbal Reports on Mental Processes," Psychological Review, 84, 231-259.

Rassuli, Kathleen M. and Gilbert D. Harrell (1989), "A New Perspective on Choice," Advances in Consumer Research (Provo, Utah: Association for Consumer Research).