An Introduction to Embodied Cognition: Implications For Consumer Research

Alan J. Malter, University of Wisconsin-Madison
ABSTRACT - This paper introduces the basic features of the newly emerging theory of embodied cognition and examines some possible implications for consumer research. The standard view of memory and information processing is critically reviewed and embodied cognition is proposed as a possible alternative theoretical basis for studying cognitive processes. Areas of consumer research which could potentially benefit from adopting an embodied view of cognition include decision making, information search behavior, and categorization. A brief concluding discussion examines embodied cognition as an alternative explanation for the phenomenon of impulse purchase behavior.
[ to cite ]:
Alan J. Malter (1996) ,"An Introduction to Embodied Cognition: Implications For Consumer Research", in NA - Advances in Consumer Research Volume 23, eds. Kim P. Corfman and John G. Lynch Jr., Provo, UT : Association for Consumer Research, Pages: 272-276.

Advances in Consumer Research Volume 23, 1996    Pages 272-276


Alan J. Malter, University of Wisconsin-Madison


This paper introduces the basic features of the newly emerging theory of embodied cognition and examines some possible implications for consumer research. The standard view of memory and information processing is critically reviewed and embodied cognition is proposed as a possible alternative theoretical basis for studying cognitive processes. Areas of consumer research which could potentially benefit from adopting an embodied view of cognition include decision making, information search behavior, and categorization. A brief concluding discussion examines embodied cognition as an alternative explanation for the phenomenon of impulse purchase behavior.


Memory has long been recognized as central to all phases of consumers' processing and interpretation of information (Bettman 1979). Though usually not the focus of consumer researchers, assumptions about the structure and operation of the human memory system underlie most of the core research areas of consumer behavior, including but not limited to: attention and perception, decision making, information search behavior, categorization and concept formation, and consumer expertise. Yet Alba, Hutchinson and Lynch (1991) note the unfortunate fact that "most traditional decision research has treated memory as an annoyance ... rather than as a target for investigation." Since our assumptions about how memory works shape our research designs, hypotheses and, ultimately, our understanding of consumer behavior, they are of critical importance and deserve both careful consideration and periodic reexamination.

Since memory and cognition issues were first introduced in the consumer research literature (Bettman 1975; McGuire 1976), a general consensus has emerged among consumer researchers in adopting the standard memory paradigms, which have often been referred to in the cognitive psychology literature as the "modal view." Most consumer researchers investigating memory and cognition consistently follow a core of authors regarding the fundamental issues of short and long-term memory (e.g., Waugh and Norman 1965; Atkinson and Shiffrin 1968), semantic and episodic memory (e.g., Tulving 1972), and the organization of memory in an associative spreading activation network (e.g., Anderson 1983; Collins and Loftus 1975). While a few consumer researchers (e.g., Alba and Hutchinson 1987; Cohen and Basu 1987) have reviewed some alternatives to these theories, they have either not pursued these ideas further or dismissed them due to various limitations. Though these latter authors exhibit a general sense that the standard memory paradigms are inadequate, they have been reluctant to abandon them for lack of a better theory.

Numerous consumer researchers have reported findings which cannot be easily accounted for by the standard paradigms, such as limited external search for information and non-choice "decisions" (Olshavsky and Granbois 1979), extremely rapid and apparently holistic purchase decisions (e.g., Hoyer 1984), failure to engage in an external search for information in cases of extreme incongruity in a categorization task (Ozanne, Brucks and Grewel 1992), the influence of positive affect on cognitive processes and decision making (Isen 1989); and the apparent suspension of normal cognitive processes in impulse purchase situations (Rook 1987; Thompson, Locander and Pollio 1990).

Meanwhile, following developments in cognitive linguistics (e.g., Lakoff 1987; Talmy 1988), a number of psychologists have recently proposed new theories of how memory works, which offer an alternative explanation for these and other perplexing types of consumer behaviors. While these new memory theories (e.g., Barsalou 1993; Barsalou et al. 1993; Glenberg, in press) are still in the early stages of development and testing and are called various names by their authors, they are part of a new paradigm which can generally be referred to as "embodied cognition." In general, the embodied view of cognition denies the core assumptions of the modal view; instead, it postulates a unitary cognitive system composed of "embodied" (i.e., perceptual, or non-linguistic) concepts. Systems of this type feature the necessary flexibility and sensitivity to context to account for complex analytical and non-analytical consumer behaviors.

The aim of this paper is to introduce some of the basic features of the new models of embodied cognition and to examine some possible implications for consumer research. Specifically, this paper will: (1) briefly review the principal features of the modal view of memory as used in consumer research; (2) describe some of the general limitations of the modal view and review some problematic findings in the consumer research literature; (3) briefly explain the central features of two recent models of embodied cognition (Barsalou 1993; Barsalou et al. 1993; Glenberg, in press); and (4) discuss some areas of consumer research which could potentially benefit from the adoption of an embodied view of cognition.


A Summary of Its Basic Features

The standard memory paradigms which have come to be known as the "modal view" in psychology were first discussed in the consumer research literature by Bettman (1975) and McGuire (1976). In a greatly expanded treatment of memory issues, Bettman (1979) further defined and solidified the modal view in consumer research. Central features of many models associated with this view include: (1) the multiple-store approach (e.g., Waugh and Norman 1965; Atkinson and Shiffrin 1968), proposing separate stores or, at least, separate functions and properties for short-term and long-term memory; (2) the multiple systems approach (e.g., Tulving 1972) characterized by distinct systems for semantic memory and episodic memory; and (3) the associative network approach (e.g., Anderson 1983; Collins and Loftus 1975), in which semantic knowledge is organized in a network of nodes linked by propositions and is retrieved through a process of spreading activation.

Subsequently, the modal view of memory in consumer research has been cited, with occasional expansion and elaboration, in countless journal articles and marketing textbooks. Many authors specifically note that Collins and colleagues' (e.g., Collins and Loftus 1975) or Anderson's (1983) spreading activation network serves as the basis for their models of such topics as consumer consideration sets (Nedungadi 1990), consumer learning (Huffman and Houston 1993), mood states (Gardner 1985) and memory for advertising (Burke and Srull 1988).

Limitations of the Modal View

Though the modal view has served as the dominant paradigm in consumer research as well as in cognitive psychology during the past few decades, many of its core elements have been questioned by subsequent research. For example, though the notion of a separate short-term memory store was initially supported in the 1960's and 1970's by findings of "recency effects" in free-recall tasks (in which the last few items on a list are remembered better than mid-list items), it was substantially weakened by later findings of such phenomena as long-term recency effects (Glenberg et al. 1983). Evidence showing rapid forgetting of simple information under conditions of distraction (Brown 1958; Peterson and Peterson 1959) also appeared to support the multiple-store approach, but the Brown-Peterson task was found to be critically flawed by proactive interference (see Glenberg, in press, and Greene 1992, for a review of this literature). Furthermore, the debate since the early 1970's over Tulving's (1972) proposal of distinct systems for semantic memory (i.e., general knowledge) and episodic memory (i.e., personal experience/autobiographical knowledge) was largely settled by the evidence presented by McKoon, Ratcliff and Dell (1986), which showed the inseparability of the two. Meanwhile, alternative models of cognition, including some based on exemplars (e.g., Hintzman 1986) and some based on embodied representations (e.g., Glenberg, in press), have shown that a single memory system with a single cognitive process may be capable of handling all types of information (i.e., both semantic and episodic information).

Given such shortcomings of the modal view, it is not surprising that the consumer research literature contains numerous reports of consumer behavior phenomena and experimental results for which the modal view does not provide a satisfactory explanation. Alba et al. (1991) argue that the nature of consumers' decision process is so complex and variable that traditional memory paradigms may be of limited value. They conclude that a "broader conceptualization of memory is necessary," as is a reconsideration of the paradigms we use to examine consumer decision making.

Other authors have directly addressed problems which might arise from the assumption that memory is based on an associative network with spreading activation. For instance, in her discussion of the influences of affect on cognitive processes, Isen (1989) points out that it would be problematic if a positive-affect advertisement for any specific brand of "soft drink" were to also serve as a retrieval cue for positive feelings toward Coke, Pepsi, and every other soft drink (or even other related products, such as beer)! Data on affect show that a person's ability to interpret incoming information, which is neglected in the modal view, is important in circumventing the likelihood that affect from one product will necessarily spread to all other associated products. Isen (1989, pp.111-112) states explicitly that spreading activation models of associative propositions do not account for all the data, noting that "the finding that positive affect influences (broadens) cognitive organization and interpretation of material seems unanticipated by associationist models."

One implication of the modal view of information processing as applied to categorization research is that a major determinant of product typicality should be attribute structure. Loken and Ward (1990) employed attribute structure as a key measure to study the degree to which the salient, goal-related attributes (as determined by an open-ended pretest of subjects) of a category member would be related to its perceived typicality. Though attribute structure was found to be one of six variables significantly related to typicality, it performed much more poorly than expected and yielded results which were inconsistent with Barsalou's (1983) data on ad-hoc categories. In another case, the design of categorization experiments by Ozanne et al. (1992) forced subjects to evaluate products on the basis of attribute information only, thus discouraging the use of holistic processing even in cases of high-discrepancy stimuli. These kinds of results seem to indicate that the standard attribute-based approaches to studying typicality judgments are not sufficiently flexible to reflect the true variability with which consumers classify products. Finally, in their review of alternative models of categorization, Cohen and Basu (1987) noted a growing disenchantment with the "overly cognitive approaches" of the modal view and its mechanistic and computer-like assumptions. Their article provides a rare example of questioning the fundamental assumptions of the modal view; they emphasize the flexible and context-dependent character of information processing and urge consumer researchers to consider alternative approaches.


So how might memory be structured and operate if it were not an associative semantic network? What other type of memory system might provide a better account of the broad range of consumer behaviors which the modal view has so much difficulty explaining? This paper introduces some of the basic features of two leading models of the newly emerging theory of embodied cognition and proposes that consumer researchers consider embodied cognition as a possible alternative to the modal view. [Though there have been many other alternative theories of cognition, including, among others, exemplar models and connectionist models, it is beyond the scope of the present paper to review them here or to present a detailed comparison between the modal view and these alternatives.]

Recently developed models of embodied cognition represent a whole new class of alternative memory theories which propose a completely different basis for the structure and operation of the cognitive system. The theorists who have created these models have been motivated by the most basic questions in cognitive science, which until very recently had been neglected by most scholars in the field. For example, Glenberg (in press) urges the reconsideration of the purpose of the memory system and Barsalou (1993) seeks an account of the core concepts in memory that would be sufficient to explain human task performance. Each has posed a slightly different question based on his own specific research area, leading to some variation in their theories and in the terminology employed in describing them.

Despite these differences, models of embodied cognition subscribe to a number of core principles which together constitute a clear departure from the modal view and establish a new paradigm for studying human cognition and related issues. First, embodied cognition theorists (e.g., Barsalou 1993; Barsalou et al. 1993; Glenberg, in press) explicitly reject a cognitive system based on associated amodal symbols (linguistic labels linked by propositions) in favor of a system based on a more fundamental cognitive unit - perceptual elements - whose meanings are nonpropositional and are derived from being directly grounded in the external environment (for a discussion of the symbol grounding problem, see Harnad 1990). Second, embodied cognition models criticize feature lists (commonly found in psychological models, including the modal view) as inadequate depictions of concepts and, instead, place great emphasis on the structural aspect of mental representations. Third, compositional mechanisms ensure that embodied cognition systems are productive and thus able to account for novel conceptual combinations and creative thought. Fourth, as a result of all of the above characteristics, embodied cognition systems are extremely flexible and highly sensitive to context. Finally, models of embodied cognition propose a single conceptual system with a single process for all types of information. Some theorists (e.g., Glenberg, in press) explicitly deny any structural or functional distinction between either short-term and long-term memory or between semantic and episodic memory.

Since the following two models of embodied cognition involve many complex issues and are still in the early stages of development, a brief introduction of their basic elements will necessarily appear somewhat dense. However, a closer examination of these two models [A review of other models of embodied cognition and a comparison of their features is beyond the scope of the present paper.] will illustrate their potential to contribute to our understanding of many perplexing consumer behavior phenomena.

Barsalou's Compositional System of Perceptual Symbols

Barsalou's (1993; Barsalou et al. 1993) theory of embodied cognition is based on a new view of concepts, conceptualizations and categories. Barsalou defines a conceptualization as a "temporary construction in working memory, derived from a larger body of knowledge in long-term memory to represent a category," which, in turn, is "a related set of entities from any ontological type (e.g., robins, sweaters, weddings, mountains, plans, anxieties)." A person's conceptualization of a particular category is extremely flexible and may change across contexts. Such categories may not be objectively correct or even accurate; they are simply a person's cognitive representation (perhaps largely unconscious) on a given occasion. In this view of cognition, Barsalou seeks to find a solution to the "ills of amodal symbol systems" by focusing on "the information in concepts that allows people to classify exemplars during perception, to process words semantically during language use, and to reason about categories in induction, problem solving, decision making and other forms of thought" (Barsalou 1993, p.30).

In order to develop a notion of the cognitive mechanisms which could produce such conceptualizations, Barsalou (1993) focused on three of their central properties: (1) Flexibility - conceptualizations can vary widely across individuals - due to individual differences in knowledge - and within individuals - due to the differential retrieval of this knowledge on any given occasion; (2) Structure - a conceptualization is a hierarchical relational structure, containing recursive properties, attribute-value sets, and structural invariants, and is constrained by context; and (3) Linguistic Vagary - verbal descriptions of conceptual content are inherently unprincipled, haphazard, and incomplete (rather than coherent, consistent and complete, as assumed in the modal view). The core concepts, or building blocks, in the cognitive architecture which Barsalou proposes and which can account for these three properties are perceptual symbols. This perceptual approach is a dramatic departure from the reliance on linguistic symbols and amodal propositions, which Barsalou notes is central to nearly every psychological theory (other than connectionism) for representing human knowledge.

A key aspect of Barsalou's (1993) system of embodied cognition is its constructive nature: "Selective attention extracts perceptual symbols from experience and ... compositional mechanisms integrate them productively during conceptual combination, imagery, and comprehension." Barsalou describes this mechanism in terms of frame theory, which accounts for the hierarchical relational structure of conceptualizations while allowing for the necessary degree of flexibility in cognition. In such a system, perceptual symbols constitute a "vocabulary of compositional elements" which can be combined (very rapidly) into novel representations in the form of "frames." Two kinds of frame structure represent knowledge of categories in this architecture: a perceptual frame and a linguistic frame. The perceptual frame "represents the perceptual symbols generally shared by the exemplars (or instances) of a category, as well as the spatial and temporal relations between them," while the linguistic frame consists of an organization of linguistic symbols grounded in the perceptual frame. The process of composing such frames into temporary conceptualizations allows a person to follow the unfolding of perceived events, produce imagined events, and comprehend language.

The structure of conceptualizations and their component frames is a stark contrast to the feature lists which represent knowledge in a wide variety of cognitive models, including most exemplar and prototype models, episodic memory theories, and connectionist models. Since there is no structure relating the various elements in a feature list, such lists contain only independent fragments of a concept's content and thus "provide grossly inadequate representations of concepts" (Barsalou 1993). Moreover, representations must be able to capture the recursive nature of conceptual knowledge, i.e., the "process by which the content of a concept can be continually decomposed." In Barsalou's view, concepts contain multiple levels of hierarchical structure. Knowledge, then, can be viewed as "frames all the way down, with every component of a frame potentially decomposing recursively into a more specific frame"; the content of a frame will depend on the level of analysis. Such a system results in mental representations which are necessarily "dense, complex, and messy," but Barsalou believes that any realistic assessment of the content of human knowledge will take this form. While some cognitive production systems based on propositional networks, e.g., Anderson's (1983) ACT*, can potentially offer a viable processing account of frames, Barsalou contends that ACT* and other such models have not been developed to produce the flexibility and structure of human conceptual knowledge.

Glenberg's Meshed Patterns of Action

Glenberg (in press) urges cognitive scientists to reconsider the basic purpose of the human memory system and to adopt a new approach in theorizing about its possible structure and operation. Glenberg proposes a compositional system of embodied cognition which is similar to Barsalou's (1993), but places greater emphasis on automatic processes and the continuous interaction between properties of the external environment and an individual's physical capabilities. In Glenberg's view, memory exists to serve perception and action in a three-dimensional world and, therefore, mental representations must "support real, physical actions involving your body and the environment." Given this purpose, he finds the associations used in standard memory paradigms (i.e., the modal view) to be "theoretically empty" and proposes that they be replaced with the new concept of mesh. More specifically, "mesh" refers to the automatic combination of the spatial-functional patterns of the projectable properties of the environment with encoded patterns of action and the non-projectable properties of previous experience, stored as embodied representations in memory. Such spatial-functional (i.e., embodied) representations of patterns of action are similar to the previously described notion of frames (Barsalou 1993).

In Glenberg's (in press) model of embodied cognition, a conceptualization is a meshed set of patterns which satisfies spatial-functional constraints (not propositional constraints) imposed by the structure of the external environment, the structure of our bodies, and memory. As the environment changes, one conceptualization flows into the next; a particular meshed conceptualization will necessarily be constrained by the previous conceptualization, as in coarticulation in speech production when the pronunciation of a particular vowel is constrained by the previous consonant. In other words, patterns of action mutually constrain and modify one another. Concepts can become associated only if their separate patterns of action can be combined, or "meshed," given the above constraints; related concepts will mesh easily. Such a cognitive system is extremely sensitive to context and many concepts may exhibit only "temporary compatibility" with one another. The property which allows diverse concepts to mesh together and is the common denominator of meaning is that all concepts are based on coherent patterns of possible bodily movement, represented internally by analogical structures that literally fit together.

In Glenberg's (in press) view, memory has two modes of operation: automatic and effortful. The primary mode involves the automatic meshing of patterns of action and is driven by the environment, which ensures that cognition remains largely reality-based. Alternatively, memory can guide conceptualization by the conscious and effortful suppression of the overriding contribution of the environment. Such suppression is necessary in order for the individual to engage in thinking beyond the immediate situation, e.g., to consciously recollect from memory, to predict and plan future actions, and to comprehend language. Anecdotal evidence supporting the notion of conscious suppression of the environment is the common phenomenon of a person literally blocking out the environment (e.g., by closing or covering their eyes, looking up or away, etc.) when engaging in a difficult mental task.

In contrast to the modal view of memory, Glenberg's (in press) model of embodied cognition proposes a single, embodied, conceptual system of spatial-functional patterns of action which specifically denies distinctions between short-term and long-term memory and semantic and episodic memory. Though Glenberg acknowledges that a type of working memory does exist in the form of the currently active conceptualization of mesh between the environment and embodied representations, he argues that this does not constitute a separate system of short-term memory and does not involve any different functions or processes. The "illusion" of short-term memory is created by constant changes in conceptualization in response to action and is enhanced by the limits on coherent conceptualization at any given moment.


This concluding section aims to show how embodied cognition principles may offer a viable explanation of many heretofore perplexing consumer behavior phenomena and will highlight some cases in which these principles have been discussed previously in the consumer research literature in the context of other theoretical paradigms. In particular, this section will focus on the potential contribution of embodied cognition to the following two areas of consumer research: (1) very rapid purchase decisions with limited external search for information; and (2) the case of impulse purchases.

Quick Decision Making with Limited Search for Information

Many consumer researchers (e.g., Hoyer 1984; Olshavsky and Granbois 1979) have observed that, contrary to the assumptions of economists and decision theorists, consumers typically make very fast purchase decisions, engage in very limited (if any) search for information and evaluate very few (if any) alternatives. It has been suggested (e.g., Alba et al. 1991) that such consumer behavior may be due to the use of memory-based strategies, which may give consumers the sense that they already have enough information to make a rational decision. The constructive aspect of embodied cognition extends this view by providing a more complete cognitive explanation for this type of behavior. In Glenberg's (in press) model, the constraints of a given situation would lead consumers to mesh limited contextual information with existing elements of knowledge stored in memory to construct a coherent conceptualization. If the current environment and conceptualization were to dominate thoughts of possible alternatives, consumers would feel they have sufficient information to make a rational decision. Only if the stakes were extremely high and time and resources allowed, might consumers feel a need to postpone the decision and pursue an external search for additional information.

Impulse Purchases

A related persistent riddle in consumer research has been the prevalence of what appears to be highly irrational impulse purchase behavior, in which consumers make spontaneous and seemingly choiceless decisions. Consumer researchers (Rook 1987; Thompson et al. 1990) have found that impulse situations are charged with affective feelings, produce an instant sense of rightness to consumers, and stimulate a desire to act immediately. The new models of embodied cognition may provide a reasonable explanation of this phenomenon and complement the many phenomenological descriptions of impulse purchase behavior reported in the consumer research literature. For example, a major theme in Thompson et al.'s (1990) analysis of shopping behavior is the notion of being captivated, which is described as "a perceptually oriented and embodied form of consumer experience... (in which) ... experiencing a product's charm takes priority over analytically evaluating its attributes" (p.356). In captivating situations, objects in the environment seem to dominate conceptualization; products are described by some consumers as not only "catching their eye," but also "jumping out," "striking," and "really hitting" them. In another study (Rook 1987, pp.193-194), consumers reported that impulse-purchase items "stood out from the rest" or formed a persistent visual image in their mind which would not go away until the item was purchased. Alba et al. (1991) have noted that the ease with which certain products will "catch one's eye" will be influenced by memory factors, which is consistent with models of embodied cognition.

A common impression (e.g., Thompson et al. 1990) is that in impulse situations consumers abandon their usually deliberate consideration of a purchase and ordinary behavioral constraints. However, the embodied view of cognition would argue that the usual and natural mode of processing is automatic, in which the current conceptualization is dominated by the external environment (especially by the target object). At that moment, projectable properties from the environment mesh perfectly with patterns of action from memory, producing an extremely coherent (i.e., seemingly "rational") conceptualization, strong positive affect for the product, and the feeling of captivation. The current environmentally dominated conceptualization creates constraints whereby the only compatible subsequent conceptualization will involve a decision to experience the item. In contrast, in order to deliberately evaluate an item, consumers must consciously suppress the contribution of the external environment by effortfully constructing counterarguments regarding such abstract concepts as the affordability and practicality of the item and the future consequences of purchasing it.


Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13 (March), 411-454.

Alba, Joseph W., J. Wesley Hutchinson, and John G. Lynch, Jr. (1991), "Memory and Decision Making," in Handbook of Consumer Behavior, eds. Thomas S. Robertson and Harold H. Kassarjian, Englewood Cliffs, NJ: Prentice Hall, 1-49.

Anderson, John R. (1983), The Architecture of Cognition, Cambridge, MA: Harvard University Press.

Atkinson, Richard C. and Richard M. Shiffrin (1968), "Human Memory: A Proposed System and Its Control Processes," in The Psychology of Learning and Motivation (Vol. 2), eds. K.W. Spence and J.T. Spence, New York: Academic Press, 89-105.

Barsalou, Lawrence W. (1983), "Ad-Hoc Categories," Memory and Cognition, 11 (3), 211-227.

Barsalou, Lawrence W. (1993), "Flexibility, Structure, and Linguistic Vagary in Concepts: Manifestations of a Compositional System of Perceptual Symbols," in Theories of Memory, eds. A.F. Collins, S.E. Gathercole, M.A. Conway, and P.E. Morris, Hillsdale, NJ: Erlbaum, 29-101.

Barsalou, Lawrence W., Wenchi Yeh, Barbara J. Luka, Karen L. Olseth, Kelly S. Mix, Ling-Ling Wu (1993), "Concepts and Meaning," in Chicago Linguistics Society 29: Papers from the Parasession on Conceptual Representations, eds. Katherine Beals, Gina Cooke, David Kathman, Sotaro Kita, Karl-Erik McCullough, and David Testen, Chicago: Chicago Linguistics Society, 23-61.

Bettman, James R. (1975), "Issues in Designing Consumer Information Environments," Journal of Consumer Research, 2 (December), 169-177.

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

Brown, J. (1958), "Some Tests of the Decay Theory of Immediate Memory," Quarterly Journal of Experimental Psychology, 10, 12-21.

Burke, Raymond R. and Thomas K. Srull (1988), "Competitive Interference and Consumer Memory for Advertising," Journal of Consumer Research, 15 (June), 55-68.

Cohen, Joel B. and Kunal Basu (1987), "Alternative Models of Categorization: Toward a Contingent Processing Framework," Journal of Consumer Research, 13 (March), 455-472.

Collins, Allan M. and Elizabeth A. Loftus (1975), "A Spreading-Activation Theory of Semantic Processing," Psychological Review, 82 (6), 407-428.

Gardner, Meryl Paula (1985), "Mood States and Consumer Behavior: A Critical Review," Journal of Consumer Research, 12 (December), 281-300.

Glenberg, Arthur M. (in press), "What is Memory For," Behavioral and Brain Sciences.

Glenberg, Arthur M., M.M. Bradley, T.A. Kraus, and G.T. Renzaglia (1983), "Studies of the Long-Term Recency Effect: Support for a Contextually Guided Retrieval Hypothesis," Journal of Experimental Psychology: Learning, Memory, and Cognition, 9, 231-255.

Greene, Robert L. (1992), Human Memory: Paradigms and Paradoxes, Hillsdale, NJ: Erlbaum.

Harnad, Stevan (1990), "The Symbol Grounding Problem," Physica D, 42, 335-346.

Hintzman, Douglas L. (1986), "'Schema Abstraction' in a Multiple-Trace Memory Model," Psychological Review, 93 (4), 411-428.

Hoyer, Wayne D. (1984), "An Examination of Consumer Decision Making for a Common Repeat Purchase Product," Journal of Consumer Research, 11 (December), 822-829.

Huffman, Cynthia and Michael J. Houston (1993), "Goal-oriented Experience and the Development of Knowledge," Journal of Consumer Research, 20 (September), 190-207.

Isen, Alice M. (1989), "Some Ways in Which Affect Influences Cognitive Processes: Implications for Advertising and Consumer Behavior," in Cognitive and Affective Responses to Advertising, eds. P. Cafferata and A. Tybout, New York: Lexington.

Lakoff, George (1987), Women, Fire, and Dangerous Things: What Categories Reveal About the Mind, Chicago: University of Chicago Press.

Loken, Barbara and James Ward (1990), "Alternative Approaches to Understanding the Determinants of Typicality," Journal of Consumer Research, 17 (September), 111-126.

McGuire, William J. (1976), "Some Internal Psychological Factors Influencing Consumer Choice," Journal of Consumer Research, 2 (March), 302-318.

McKoon, Gail, Roger Ratcliff, and Gary S. Dell (1986), "A Critical Evaluation of the Semantic-Episodic Distinction," Journal of Experimental Psychology: Learning, Memory, and Cognition, 12 (2), 295-306.

Nedungadi, Prakash (1990), "Recall and Consumer Consideration Sets: Influencing Choice without Altering Brand Evaluations," Journal of Consumer Research, 17 (December), 263-276.

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

Ozanne, Julie L., Merrie Brucks and Dhruv Grewel (1992), "A Study of Information Search Behavior During the Categorization of New Products," Journal of Consumer Research, 18 (March), 452-463.

Peterson, L.R. and M.R. Peterson (1959), "Short-Term Retention of Individual Verbal Items," Journal of Experimental Psychology, 58, 193-198.

Rook, Dennis W. (1987), "The Buying Impulse," Journal of Consumer Research, 14 (September), 189-199.

Talmy, Leonard (1988), "Force Dynamics in Language and Cognition," Cognitive Science, 12, 49-100.

Thompson, Craig J., William B. Locander, and Howard R. Pollio (1990), "The Lived Meaning of Free Choice: An Existential-Phenomenological Description of Everyday Consumer Experiences of Contemporary Married Women," Journal of Consumer Research, 17 (December), 346-361.

Tulving, Endel (1972), "Episodic and Semantic Memory," in Organization of Memory, eds. Endel Tulving and Wayne Donaldson, New York: Academic Press.

Waugh, N.C., and D.A. Norman (1965), "Primary Memory," Psychological Review, 72, 89-104.