Fellow's Award Speech the Decision Maker Who Came in From the Cold

James R. Bettman, Duke University
[ to cite ]:
James R. Bettman (1993) ,"Fellow's Award Speech the Decision Maker Who Came in From the Cold", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 7-11.

Advances in Consumer Research Volume 20, 1993      Pages 7-11

FELLOW'S AWARD SPEECH

THE DECISION MAKER WHO CAME IN FROM THE COLD

James R. Bettman, Duke University

I feel deeply privileged to have been elected a Fellow of the Association for Consumer Research. I would first like to express my profound thanks and gratitude to all those who have helped me over the years, especially my colleagues at UCLA and Duke, my extended set of colleagues from ACR, and my wife Joan and son David. Hal Kassarjian should be singled out for special recognition as an early mentor and source of encouragement. Finally, I would like to publicly thank a particularly important group to me, my doctoral students: Debbie Scammon, Jack Swasy, Hubert Gatignon, Mita Sujan, Chris Puto, Kevin Keller, Betsy Creyer, Itamar Simonson, Helen Anderson, Ron Goodstein, Pete Nye, Eloise Coupey, Carolyn Yoon, Ellen Garbarino, and Mary Frances Luce. Over the years, they have provided a constant source of new ideas and excitement.

In this paper, I will first provide a very brief historical perspective on my work in consumer information processing, then outline my current perspectives on consumer choice, and close by considering ways to raise the temperature of typically cold information processing models of consumer decision making.

HISTORICAL PERSPECTIVE

My path to an information processing viewpoint on consumer choice was quite convoluted and marked by many episodes of serendipity. The first bit of good fortune was my introduction as an undergraduate to the work of Herbert Simon. The piece which affected me the most was his paper "A Behavioral Model of Rational Choice" (Simon, 1955). Simon's pioneering idea of bounded rationality, that individuals have limited capabilities and must simplify the world in order to deal with it, played a major role in all of my later thinking on consumer choice.

A second piece of luck was my exposure as a graduate student to the edited book Computers and Thought (Feigenbaum and Feldman, 1963), which provided examples of research using the idea of bounded rationality to model the actual strategies used by problem solvers. The specific paper in the book which gave me the idea for my dissertation was Clarkson's model of the choice processes of a trust investor (Clarkson, 1963). His work gave me a method for implementing the ideas I was generating about consumer choice; I saw that I could use verbal protocols from actual consumer choice episodes to build models of those consumers' choice processes. In my dissertation research, I followed two consumers around the grocery store as they shopped over a six to eight week period. I asked them to think aloud while they were shopping, tape recorded these verbalizations, and then interpreted these data in developing decision net models of their choice processes.

After finishing my dissertation and going to UCLA in 1969, I continued my work on consumer information processing and published an article in the Journal of Marketing Research in 1971 which proposed a general model of decision and choice (Bettman, 1971). By the time I was due for my first sabbatical in 1975, I decided that I would try to write a book that would flesh out that model in more detail and review research relevant to consumer choice processes. While doing the research for the book I fortunately encountered three major ideas which provided the foundation for my future work and necessitated radically changing the simple 1971 framework: 1) the particular strategies people use depend upon the nature of the task (Payne, 1976); 2) a detailed task analysis of the choice environment can provide a great deal of insight into the nature of the strategies and heuristics which are likely to be used in that environment (Newell and Simon, 1972); and 3) decision strategies may be constructed or made up on the spot instead of being available in memory and simply implemented (Nakanishi, 1974).

At the time I began writing my book, it was becoming increasingly clear that the answer to the question "How do people make decisions?" should be "It depends." In 1975 I happened to see a manuscript version of John Payne's 1976 paper, which was clearly a tour de force. This paper began the transition from demonstrating that decision behavior was contingent to understanding why decision processes were contingent; it also provided a methodological blueprint for carrying out and analyzing process-tracing research on contingent processing (I had also been independently exposed to one aspect of that blueprint, monitoring information acquisitions, by Jack Jacoby (e.g., Jacoby, 1975; Jacoby, Chestnut, Weigl, and Fisher, 1976)).

Newell and Simon's (1972) concept of a task analysis was also a major factor in changing my thinking about choice processes. They argued that the structure of the task imposed constraints on the types of strategies that could be used; hence, understanding that structure provided insights into how individuals were likely to behave if they wished to be successful at that task. As I wrote the chapter on consumer decision processes for my book in the fall of 1975, I tried to do analyses of typical tasks faced by consumers. It quickly became apparent to me (and to others at about that time) that memory processes should play a major role in thinking about consumer decision making. Newell and Simon's task analysis idea, therefore, indirectly led to the focus on memory evident in my book.

The final idea, and one whose consequences I am still working out, was the notion that decision strategies may be constructed on the spot, made up on the fly, so to speak. Masao Nakanishi was the main instigator when, in typical fashion, he asked me a deceptively simple question one day. He asked me if I really believed that consumers had these complex decision nets that I pictured in my articles stored in memory. The implication, of course, was that he did not believe that. In thinking about it, I decided that I did not believe it either. This led me to develop ideas about constructive processing that played a major role in the chapter on decision processes in my book.

These ideas were central to the transformation in my thinking from the simple 1971 flowchart to the theory embodied in the 1979 book (Bettman, 1979). In the next section, I briefly discuss my current perspectives on consumer choice and then consider how to heat up consumer information processing research.

CURRENT PERSPECTIVES ON CONSUMER CHOICE

My current perspectives on choice processes in general and on consumer choice are described in detail elsewhere (Payne, Bettman, and Johnson, in press; Bettman, Johnson, and Payne, 1991), so I will outline these ideas only briefly here. Most of these ideas have been developed jointly with John Payne, Eric Johnson, and our graduate students, so I will usually say "we" in the following. Many of these perspectives represent further development of the three main ideas discussed in the previous section.

The focus of my current work is on strategy selection in decision making. One of the major empirical observations regarding decision behavior is that individuals use different strategies in different situations. We argue that this use of multiple strategies is generally an adaptive response by a limited capacity information processor. Two major goals of decision makers which help us to understand which strategy an individual will use in any given situation are the desire to make a good decision and the desire to conserve cognitive effort. We argue that individuals select strategies based upon tradeoffs between the accuracy a given strategy might attain in a particular choice environment and the cognitive effort required to execute that same strategy in that choice environment.

For any given choice environment, different strategies provide different levels of accuracy and require different amounts of cognitive effort. In addition, the same strategy may be characterized by differing accuracy and effort levels across different environments. By modeling various strategies as sequences of elementary information processes (e.g., comparisons, additions, eliminations), measuring cognitive effort using counts of such operations, and running computer simulations of these strategies, we are able to estimate accuracy and effort levels for each strategy and then make predictions about which strategies will be used in a particular choice task (e.g., Payne, Bettman, and Johnson, 1988). In general, this approach has been very successful in predicting the patterns of adaptive strategy selection individuals exhibit in a variety of situations (Payne, Bettman, and Johnson, in press). Hence, we believe that this accuracy/effort tradeoff approach provides a useful conceptual framework for understanding contingent decision processes.

Much of this work has assumed that the decison maker possesses a repertoire of strategies, evaluates the accuracy and effort levels characterizing those strategies for a particular choice, and chooses a strategy which represents a reasonable accuracy/effort tradeoff for that task. However, we have extended that view to take account of the fact that individuals often learn about the structure of a choice task as they go along and use information they have already extracted from the task to decide what to do next. As noted above, this is a constructive view of choice, where individuals can be opportunistic and change their processing to exploit what they have learned. For example, a consumer may start to compare alternatives on what is a priori the most important attribute and discover that there is little variation on that attribute across alternatives. He or she might then try to consider another attribute but find it too difficult to understand the information, and so on. Individuals sometimes make spur of the moment shifts in processing direction instead of merely executing some strategy determined beforehand. This constructive view implies that the resulting heuristics will be very sensitive to specific salient features of the choice situation.

Constructive processing is consistent with our accuracy/effort viewpoint in the sense that we postulate that the on the spot shifts in direction are based upon local, momentary accuracy/effort assessments. For example, if an individual notes that all values on an attribute are similar across alternatives and shifts to another attribute, that shift reflects a tradeoff of the low benefits from continued processing of that attribute versus the effort required to process it. This leads to a dynamic view of contingent processing, where the nature of the choice task changes as the individual progresses.

We have focused above on the notion that individuals build strategies on the fly. However, another possible approach to constructive processing was proposed by Eloise Coupey (1990). She argued that an individual can restructure the available data by transformation, elimination, reordering, and so on. Then the individual can make a choice by applying an existing heuristic to the restructured data. Whether individuals take the data as given and construct a heuristic, take heuristics as given and restructure the data, or use a combination of these two methods, however, it is clear that at least in some instances people make things up as the choice progresses. This viewpoint is becoming more widespread. As we have emphasized (Payne, Bettman, and Johnson, 1992), an underlying theme of much recent behavioral decision research is that preferences and beliefs regarding objects and events of any complexity are often constructed in generating a response to a judgment or choice task (e.g., Slovic, Griffin, and Tversky, 1990; Tversky, Sattath, and Slovic, 1988).

COMING IN FROM THE COLD

Much research on consumer information processing, including my 1979 book, is "cold" cognition. That is, such "hot" constructs as feelings and emotion are given little emphasis. The models are overbearingly cognitive; like the tin man in the Wizard of Oz, these models have no heart. Hal Kassarjian's initial reaction to draft chapters of my 1979 book was that no one could think that much about consumer choices (actually, he would often glance over at me and say that maybe one consumer thought that much). I believe that the time has come to begin to redress this imbalance, lest we, to paraphrase my favorite fictional private eye, Kinsey Millhone, come perilously close to boring ourselves insensible with mental processes (Grafton, 1987, p. 64). Some very nice work has already been done in some areas, such as the work of Edell and Burke (1987) on the role of feelings in the processing of advertisements. I would like to propose two additional areas for making consumer information processing "hot" rather than "cold": 1) considering the effects of invoking autobiographical memories via marketing stimuli and 2) examining decision making under stress.

Invoking Autobiographical Memories

Consumer information processing researchers have generally focussed on the more obvious attributes or benefits of products. However, one major critique from postmodern scholars is that we cannot understand consumer behavior without taking a much more global view about what consumers are trying to accomplish (Belk, 1987). If we take a functional perspective and ask what role consumer behavior plays in people's lives, we may need to consider a broader or deeper set of concerns that individuals have when they make their consumer choices. That is, we will have to consider not only what products do, but what they mean to consumers.

For example, Belk (1988) argues compellingly that possessions play a major role in shaping and reflecting our identities. In particular, one important set of meanings that possessions have for individuals is the events, people, or past experiences that they symbolize (e.g., Belk, 1991; however, see Wallendorf and Arnould (1988) for evidence that such meanings vary across cultures). The study of autobiographical memories considers similar phenomena (Brewer, 1986). An autobiographical memory is a recollection of a particular episode from one's past, and one of the general properties of such memories is that they are affectively charged. That is, feelings and emotions are often associated with such memories.

Hans Baumgartner, Mita Sujan, and I have proposed that one way that advertisements can evoke emotions and feelings is to cue the retrieval of product-related autobiographical memories. In a series of three studies, we demonstrated that cuing autobiographical memories led to reduced analysis of and memory for product information, led to stronger reported feelings, and influenced ad evaluations (Baumgartner, Sujan, and Bettman, 1992). In current research (Sujan, Bettman, and Baumgartner, 1992), we argue that the extent of transfer of autobiographical affect to brand evaluations depends upon the degree to which the ad forges a link between the brand and the autobiographical memory.

Autobiographical memory is also an intriguing approach for studying attachment to possessions, products, or even brand names. In particular, the centrality of the people and events in the autobiographical memories associated with a possession, product, or brand name should be related to the degree of felt attachment. Autobiographical memories involving important events and people significant to the individual should lead to greater attachment.

These ideas about autobiographical memory can be extended in several directions. First, consumers often anticipate how products might play a role in future events (e.g., a consumer may imagine the effect of wearing a particular outfit at a major social event). Thus, consumers may generate autobiographical anticipations when they envision possible outcomes for specific future events in which they themselves play a role. Again, the centrality of the events and individuals in these autobiographical anticipations may greatly influence how a consumer thinks about the product; if the anticipations include very important events/individuals, the consumer may become pre-attached to the product. That is, the product comes to have meaning to the individual prior to its use or perhaps even prior to its purchase. Such anticipations may also constrain the set of options an individual is willing to consider; for example, once the consumer has pictured him/herself in a particular type of outfit, he or she may find it difficult to consider other styles (I thank Julie Edell for this idea).

A second idea for generalizing these ideas about autobiographical memory involves emotional reactions to aesthetic events (Walters, 1989; Frijda, 1989). That is, a novel or a movie or an ad may make us teary-eyed even if no particular event from our past is evoked. How does this occur? As the papers by Walters and Frijda point out, it is very difficult to develop a theoretical framework for understanding such responses. However, I believe that they represent both an important category of emotional responses and an intriguing area for future research.

Decision Making Under Stress

In my dissertation and several studies following from it (e.g., Bettman, 1973), I attempted to examine the idea of perceived risk. My interest was sparked to a large extent by introspection; sometimes choosing involved feelings of great uncertainty and perhaps even worry or anxiety. In retrospect, my work at that time missed the point, since I tried to characterize risk cognitively rather than at a more visceral level. I have recently begun to consider the more general topic of individuals' reactions to decision making under stress.

In our first work involving stress, John Payne, Eric Johnson, and I examined how individuals respond to time pressure (Payne, Bettman, and Johnson, 1988). We found that individuals accelerated their processing, focussed on a subset of the available information, and changed processing strategies. In particular, under high time pressure individuals adapted by examining at least some information about all alternatives quickly rather than attempting to examine a limited number of options in more depth (see Eisenhardt (1989) for similar results from a case study of the computer industry).

Such externally-imposed stress is not the only possible source of stress, however. As in the case of perceived risk, individuals can generate stress and emotion while making a decision. In fact, there is evidence that generated stress has different effects than externally-imposed stress (Christianson, 1992). How might stress be generated during a decision? At least in part, stress and emotions are generated by autobiographical anticipations regarding the outcomes of a decision, the reactions of others, and so on (see Simonson, 1992 for an example of research considering anticipations of regret and responsibility). We hypothesize that the degree of stress/emotional response generated by such anticipations will be, as before, a function of the importance or centrality of the events and people involved. In addition, the degree of emotional response may be a function of the extent to which the anticipations contain the specific core relational themes characterizing different emotions (Lazarus, 1991). For example, anticipations of experiencing irrevocable loss would lead to sadness, anticipations of transgressing a moral imperative would lead to guilt, or anticipations of being slighted or demeaned to anger.

How might stress and emotion affect decision making? John Payne, Mary Frances Luce, and I have hypothesized in recent work in progress that stress generated during a decision makes any given decision process both more effortful and more error-prone. Given our accuracy/effort framework, that would imply greater use of simpler strategies. In addition, recent work on the effects of emotional stress on memory suggests that the level of emotion generated during a choice task will affect the type and amount of information recalled about the alternatives (i.e., higher emotion might lead to better recall of information central to the choice but less accurate memory for more peripheral items (Christianson, 1992)). Finally, we feel that introducing a broader range of dependent measures (e.g., feelings of conflict or the desire to procrastinate or avoid the choice) will provide more insights into some of the affective consequences of stress and emotion for decision making.

Autobiographical anticipations have emerged as a potentially important construct both in examining product meanings and in considering decision making under stress. Hence, in the next section I briefly consider the nature of autobiographical anticipations.

The Nature of Autobiographical Anticipations

How can we characterize autobiographical anticipations? We can probably say without much controversy that such anticipations, like autobiographical memories, are often affectively charged. However, what principles govern the elicitation and effects of these anticipations? For example, is their elicitation subject to principles similar to those of memory retrieval? The work of Marcia Johnson may be very relevant in answering such questions (e.g., Johnson and Raye, 1981; Johnson, Raye, Wang, and Taylor, 1979). Johnson has carefully studied the differences between and interactions among memory for real and imagined events. For example, she and her colleagues have shown that estimates of the frequency of actual events can be influenced by the degree to which similar events have been imagined (Johnson et al., 1979). In our context, if a consumer imagines certain sorts of uses involving a product before purchase (the notion of pre-attachment discussed above), this may later influence the consumer's recall of the frequency of that sort of product usage after purchase. Thus, pre-attachment can potentially influence both actual future usage patterns and the consumer's memory for those patterns.

However, actual events and autobiographical anticipations need not agree. As Kahneman and Snell (1990; 1992) have pointed out, individuals often may not have good insight into their future tastes and preferences. In addition, others involved in the event may not react to an individual's product choices as he or she had anticipated. It would be fascinating to study how such conflicts affect the resulting autobiographical memory for the event and anticipations for future events. In general, I believe that research on autobiographical anticipations represents a very fertile opportunity for consumer researchers.

CONCLUSIONS

The ideas above reflect my feeling that we should "come in from the cold" and thaw consumer information processing research by making it more "hot." Note that in several of the cases described above, insights from interpretive studies of consumer behavior have been used to generate ideas for studying consumer information processing in a different light, albeit still experimentally. In my view, such complementarity between research approaches, rather than competition, is both extremely desirable and potentially quite productive (see also McQuarrie and Mick, 1992).

Finally, as Stephen Jay Gould (1989) has emphasized, history is highly contingent. If we started the clock again and allowed evolution to unfold, the chances we would end up with humans and the other creatures we see today are probably vanishingly small. The same could be said for my journeys in consumer information processing, which have been subject to enormous serendipity. I am certainly glad they turned out the way they did and that I have had the great honor of being with you today.

REFERENCES

Baumgartner, Hans, Mita Sujan, and James R. Bettman (1992). Autobiographical Memories, Affect, and Consumer Information Processing. Journal of Consumer Psychology, 1 (1), 53-82.

Belk, Russell W. (1987). ACR Presidential Address: Happy Thought. In Melanie Wallendorf & Paul F.Anderson (Eds.), Advances in Consumer Research, Volume 14 (pp. 1-4). Provo, UT: Association for Consumer Research.

Belk, Russell W. (1988). Possessions and the Extended Self. Journal of Consumer Research, 15 (September), 139-168.

Belk, Russell W. (1991). Possessions and the Sense of Past. In Russell W. Belk (Ed.), Highways and Buyways: Naturalistic Research from the Consumer Behavior Odyssey (pp. 114-130). Provo, UT: Association for Consumer Research.

Bettman, James R. (1971). The Structure of Consumer Choice Processes. Journal of Marketing Research, 8 (November), 465-471.

Bettman, James R. (1973). Perceived Risk and Its Components: A Model and Empirical Test. Journal of Marketing Research, 10 (May), 184-190.

Bettman, James R. (1979). An Information Processing Theory of Consumer Choice. Reading, MA: Addison-Wesley.

Bettman, James R., Eric J. Johnson, and John W. Payne (1991). Consumer Decision Making. In Thomas S. Robertson & Harold H. Kassarjian (Eds.), Handbook of Consumer Behavior (pp. 50-84). Englewood Cliffs, NJ: Prentice-Hall.

Brewer, William F. (1986). What Is Autobiographical Memory? In David C. Rubin (Ed.), Autobiographical Memory (pp. 25-49). Cambridge, England: Cambridge University Press.

Christianson, Sven-+ke (1992). Emotional Stress and Eyewitness Memory: A Critical Review. Psychological Bulletin, 112 (September), 284-309.

Clarkson, Geoffrey P. E. (1963). A Model of the Trust Investment Process. In Edward A. Feigenbaum & Julian Feldman (Eds.), Computers and Thought (pp. 347-371). New York: McGraw-Hill.

Coupey, Eloise (1990). Decision Restructuring in Consumer Choice. Doctoral dissertation, Fuqua School of Business, Duke University.

Edell, Julie A. and Marian Chapman Burke (1987). The Power of Feelings in Understanding Advertising Effects. Journal of Consumer Research, 14 (December), 421-433.

Eisenhardt, Kathleen M. (1989). Making Fast Strategic Decisions in High Velocity Environments. Academy of Management Journal, 32 (September), 543-575.

Feigenbaum, Edward A. and Julian Feldman (Eds.). (1963). Computers and Thought. New York: McGraw-Hill.

Frijda, Nico H. (1989). Aesthetic Emotions and Reality. American Psychologist, 44 (December), 1546-1547.

Gould, Stephen Jay (1989). Wonderful Life. New York: W. W. Norton.

Grafton, Sue (1987). 'D' Is for Deadbeat. New York: Bantam Books.

Jacoby, Jacob (1975). Perspectives on a Consumer Information Processing Research Program. Communication Research, 2 (July), 203-215.

Jacoby, Jacob, Robert W. Chestnut, Karl C. Weigl, and William Fisher (1976). Pre-purchase Information Acquisition: Description of a Process Methodology, Research Paradigm, and Pilot Investigation. In Beverlee B. Anderson (Ed.), Advances in Consumer Research, Volume 3 (pp. 306-314). Chicago: Association for Consumer Research.

Johnson, Marcia K. and Carol L. Raye (1981). Reality Monitoring. Psychological Review, 88 (January), 67-85.

Johnson, Marcia K., Carol L. Raye, Alvin Y. Wang, and Thomas H. Taylor (1979). Fact and Fantasy: The Roles of Accuracy and Variability in Confusing Imaginations with Perceptual Experiences. Journal of Experimental Psychology: Human Learning and Memory, 5 (May), 229-240.

Kahneman, Daniel and Jackie Snell (1990). Predicting Utility. In Robin M. Hogarth (Ed.), Insights in Decision Making: A Tribute to Hillel J. Einhorn (pp. 295-310). Chicago: University of Chicago Press.

Kahneman, Daniel and Jackie Snell (1992). Predicting a Changing Taste: Do People Know What They Will Like? Journal of Behavioral Decision Making, 5 (July-September), 187-200.

Lazarus, Richard S. (1991). Progress on a Cognitive-Motivational-Relational Theory of Emotion. American Psychologist, 46 (August), 819-834.

McQuarrie, Edward F. and David G. Mick (1992). On Resonance: A Critical Pluralistic Inquiry into Advertising Rhetoric. Journal of Consumer Research, 19 (September), 180-197.

Nakanishi, Masao (1974). Decision-net Models and Human Information Processing. In G. David Hughes & Michael L. Ray (Eds.), Buyer/Consumer Information Processing (pp. 75-88). Chapel Hill, NC: University of North Carolina Press.

Newell, Allen and Herbert A. Simon (1972). Human Problem Solving. Englewood Cliffs, NJ: Prentice-Hall.

Payne, John W. (1976). Task Complexity and Contingent Processing in Decision Making: An Information Search and Protocol Analysis. Organizational Behavior and Human Performance, 16 (August), 366-387.

Payne, John W., James R. Bettman, and Eric J. Johnson (1988). Adaptive Strategy Selection in Decision Making. Journal of Experimental Psychology: Learning, Memory, and Cognition, 14 (July), 534-552.

Payne, John W., James R. Bettman, and Eric J. Johnson (1992). Behavioral Decision Research: A Constructive Processing Perspective. Annual Review of Psychology, 43, 87-131.

Payne, John W., James R. Bettman, and Eric J. Johnson (In press). The Adaptive Decision Maker. Cambridge, England: Cambridge University Press.

Simon, Herbert A. (1955). A Behavioral Model of Rational Choice. Quarterly Journal of Economics, 69, 99-118.

Simonson, Itamar (1992). The Influence of Anticipating Regret and Responsibility on Purchase Decisions. Journal of Consumer Research, 19 (June), 105-118.

Slovic, Paul, Dale Griffin, and Amos Tversky (1990). Compatibility Effects in Judgment and Choice. In Robin M. Hogarth (Ed.), Insights in Decision Making: A Tribute to Hillel J. Einhorn (pp. 5-27). Chicago: University of Chicago Press.

Sujan, Mita, James R. Bettman, and Hans Baumgartner (1992). Autobiographical Memories and Consumer Judgments. Working paper, Pennsylvania State University.

Tversky, Amos, Shmuel Sattath, and Paul Slovic (1988). Contingent Weighting in Judgment and Choice. Psychological Review, 95 (July), 371-384.

Wallendorf, Melanie and Eric J. Arnould, E. J. (1988). "My Favorite Things": A Cross-cultural Inquiry into Object Attachment, Possessiveness, and Social Linkage. Journal of Consumer Research, 14 (March), 531-547.

Walters, Kerry S. (1989). The Law of Apparent Reality and Aesthetic Emotions. American Psychologist, 44 (December), 1545-1546.

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