Affect and Cognition: a Closer Look At Two Competing Theories

James A. Muncy, Texas Tech University
ABSTRACT - Two competing theories have developed which relate to the nature of affect. The present paper discusses some of the consumer behavior implications of the evidence presented by the proponents of these two competing theories. These implications relate to the storage of affect, the role of affect in memory, and the relationship between affect and physiology.
[ to cite ]:
James A. Muncy (1986) ,"Affect and Cognition: a Closer Look At Two Competing Theories", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 226-230.

Advances in Consumer Research Volume 13, 1986      Pages 226-230


James A. Muncy, Texas Tech University


Two competing theories have developed which relate to the nature of affect. The present paper discusses some of the consumer behavior implications of the evidence presented by the proponents of these two competing theories. These implications relate to the storage of affect, the role of affect in memory, and the relationship between affect and physiology.


One of the primary topics of interest to consumer researchers is that of consumer preferences. Much of the research in consumer behavior focuses on how preferences develop and how these preferences affect the consumer choice process. Almost by definition, these preferences have been assumed to be affectively based. Although few would deny that there exists a relationship between affect and preference, there has recently developed a major debate in the literature (particularly in psychology) as to the nature of affect. Historically, it has been assumed that affect is "post-cognitive." This means that affect occurs as a result of (and therefore after) cognition. In 1980, Zajonc proposed a "separate systems" view of affect which challenged this basic assumption. Thus, two competing theories have developed which differentially explain the nature of affect: the traditional or "postcognitive" theory of affect and the more recent "separate systems" view of affect (some key articles in this debate are Baars 1981; Birnbaum 1981; Lazarus 1981, 1982, 1984; Mellers 1981; Rachman 1981; Slife 1981; Watts 1983; Zajonc 1980. 1981. 1984; and Zajonc and Markus 1982).

ZaJonc's separate systems perspective of affect and cognition is based on three primary propositions. [Zajonc actually presents several propositions regarding affect. The three selected seem to be the ones which are central to the debate and germane to the field of consumer behavior. Thus, only these three are discussed.] The first is that affect and cognitions are controlled by two separate systems in the brain. a Though these two systems can affect one another, they are generally believed to operate independently. The second proposition is that affect is not necessarily post-cognitive. This does not imply that affect cannot result from cognitive processing (no one denies that affect can be post-cognitive). It simply states that affect can result from causes other than cognitive processing. > e third proposition (one that can be refuted without the whole theory being rejected) is that affect is primary. This implies that a person's primary reaction in a situation where affect and cognitions come into conflict is the affective reaction. E is can be seen in the dilemma often faced when one states "my thinking tells me do one thing but my feelings tell me to do something else. In such a situation, the person's typical problem 18 to stifle his or her primary reaction (i.e., do what he or she "feels") 80 that he or she can do the "rationale thing" (i.e., what he or she thinks).

As stated earlier, this separate systems view is opposed to the more traditional view which assumes that affect is always preceded by and occurs as a result of cognitive processing. Most consumer behavior models fall into this post-cognitive tradition. She only consumer behavior models which could be found that explicitly allow affect to either parallel or precede cognition were Ray's (1973) dissonance attribution hierarchy and Bagozzi's (1983) affect response, parallel response, and social response models.

The exact mechanics of these post-cognitive models of affect vary but they are typically decompositional. By being decompositional, they typically divide an idea or object into component attributes. By somehow utilizing a set of beliefs that pertain to the object or idea being evaluated (which are cognitive in nature), a person derives an affective reaction to that object or idea. [It must be noted that not all cognitive models fall into this decompositional mold. In fact, Lazarus' (1981, 1982, 1984) arguments against Zajonc's separate systems view of affect are based, to a large extent, on his position that these types models do not accurately represent cognitive processing.] The most researched decompositional models in consumer behavior are the multiattribute models of attitude (for a review of this class of models, see Lynch 1985).

It is unlikely that this debate will soon be resolved. First, there are definitional barriers to its resolution. As pointed out by Izard, Kagan, and Zajonc (1984) there is no generally accepted definition of either affect or cognition. One could solve this debate by simply defining affect and cognition in such a way that they either are or are not, by definition, related. However, as Izard, Kagan, and Zajonc point out, this does little to advance our understanding of either affect or cognition. If one does not simply define away this problem, there are many complex definitional issues which must be addressed before any viable solution can be proposed. Secondly, there are several complex operational issues that must be addressed in this area. Presently, one cannot unequivocally show the absence of cognitions. Without doing so, it is impossible to refute the traditional theory of affect. Thus, it does not appear as though an acceptable resolution of this debate is in the immediate horizon. Even without such a resolution, however, some issues have surfaced in this debate which should be of interest to consumer researchers. a e present paper discusses some of these issues. But before this is done, it is necessary to look in greater depth at what is meant by the term "affect."


As indicated earlier, there is no universal agreement as to a definition of affect. However, any discussion of the relation of affect to consumer behavior must be clearly rooted in an explicit definition of affect. Since the present discussion focuses on the evidence presented in the debate surrounding Zajonc's separate systems view of affect and cognition, it seems fruitful to define affect in a way consistent with his conceptualization. One problem in doing this is that Zajonc does not explicitly define affect. One must infer such a definition from what Zajonc does say.

Though Zajonc does not explicitly define affect, he does make a clear statement as to what types of "feelings" he wishes to explore and what types of "feelings" he wishes to ignore. He states:

The class of feelings considered here is that involved in the general quality of behavior that underlies the approach/avoidance distinction. Thus for the present purpose, other emotions such as surprise, anger, guilt, or shame which have been identified in the literature and extensively analyzed by Tomkins (1962, 1963), Izard (1977), and others are ignored. (p. 152)

From this brief quote, two implications can be drawn as to what Zajonc means by the term "affect." First, it appears as though Zajonc views affect as a type of human emotion. Secondly, it is clear that he does not wish to include all human emotions under his discussion of affect; he is interested only in those that explicitly (ant perhaps exclusively) relate to approaching or avoiding an object or idea.

Since Zajonc appears to view affect as a form of emotion, it seems reasonable to conclude that it would possess the characteristics of an emotion. Izard, Kagan, and Zajonc (1984) state that it is generally agreed that three components of emotion are "neurophysiological/biochemical, motor or behavior expressive, and subjective/experiential" (p. 3). Thus affect, as discussed by Zajonc should have some physiological basis, be expressed in behavior, and be individualistic and evaluative in nature.

To come up with a definition of affect with which to evaluate the separate systems view, these three components must then be incorporated with the approach/avoidance distinction. This approach/avoidance distinction adds two additional elements to a definition of affect: 1) it appears as though affect is object or idea specific (i.e., the subject or idea that is being approached or avoided) and 2) affect appears to have a valence component (i.e., the tendency to move" towards or away from an object or idea).

With all of this in mind, a working definition with which to evaluate the claims of the two theories of cognition and affect would be as follows:

Affect is an individualistic, physiological based predisposition to behave either positively or negatively towards an objective or idea.

It is interesting to note that this definition of affect does not deviate substantially from definitions of attitudes present in the literature (e.g., Fishbein and Ajzen 1975). The difference between the current definition of affect and previous definitions of attitudes is two-fold. First, the current definition does not require affect to be "learned" (though it does not say that affect is not "learned"). Secondly, it states that affect is physiologically based (thus, introducing the distinction between "hot" and "cold" cognitions). Since the study of attitudes has become a key topic in consumer behavior, it appears as though affect, as defined here, should also be of interest to consumer researchers.


With this working definition of affect, the present paper is now ready to investigate some of the implications of the affect/cognition debate for consumer behavior thought. The three particular issues addressed by the present paper are: 1) the role of memory in affect, 2) the role of familiarity in affect, and 3) the relationship between affect and physiology.


Much of the research in this area would indicate that affect and cognitions are stored separately in the brain (for an excellent presentation of this position, see Fiske and Taylor 1984, pp. 35-36). An example of how this might occur can be seen in the field of social cognition. When two people meet and interact, usually an affective based evaluation is formed (i.e., the people develop some feelings of liking or disliking for each other). At some later point in time, either person can probably state if he or she liked the other person (i.e., the affective reaction), regardless of whether or not he or she can remember the specifics of the interaction (i.e., the basis for the affective reaction).

If it is true that affect and cognitions are stored separately, then a corollary to this proposition is that affect can develop at the time of encoding and not just at the time of decoding. Most cognitive based models of consumer behavior seem to assume, at least implicitly, that affect occurs at the time of decoding. Figure 1 illustrates this difference. The first model presents the typical consumer behavior approach. It assumes that consumers search for information and store that information in memory. At the point of the decision, the consumer retrieves from memory (via internal information search) this information. This information is then used to form a preference. The second model presents an alternative interpretation of this process. It shows that affect is developed (perhaps as a result of cognitive processing) and then stored in memory (separate from the basis of this affect). At some later point, when the person is forced to behave based on the preference, the feelings are recovered from memory (not necessarily with the basis of these feelings) .

The differences between the models presented in Figure I are not trivial. The traditional model would imply that there should be a close correspondence between information in memory and evaluations that are presumably made based on this information. As Lichtenstein and Srull (1985) point out, most psychological the theories of persuasion make such an assumption. However, they also point out that the empirical evidence to date has not supported such an assumption. People do not always have access to the basis for their evaluative judgment. a 18 finding is problematic for the traditional model presented in Figure 1; however; it creates no problem for the alternative model.



Here it is interesting to note that models have developed which propose that presented information is stored separately from interpreted information (see Beattie and Mitchell 1985). Such models assume that the presented information is stored in episodic form and the interpreted information is stored by fitting it into the existing semantic structure. In such cases, the interpreted information (which is integrated) should be retained such longer than the original information (which is stored episodically). Both models presented in Figure 1 imply that the affective based evaluations occur as a result of interpreting incoming information. However, the second model would imply that the evaluations and the information on which the evaluations are made are both stored in memory. If the models of memory discussed by Beattie and Mitchell are accurate, then one would expect affective judgments (stored in memory through integration) to remain in memory longer than the information (stored episodically) which caused these judgments.

The above discussion implies that it is likely that a consumer may have strong affective based judgments stored in memory even when he or she is not able to remember the basis for the evaluation. This, of course, presents a major obstacle when studying such evaluations. There may be reasons for assuming that, even when consumers want to, they often cannot tell a researcher why they prefer one store or brand over another.

Another implication of the alternative model presented in Figure 1 relates specifically to the role of memory in the consumer choice process. The term used in consumer behavior for the process by which a person retrieves something from memory is "internal information search" (e.g., Bettman 1979). The implication is that the consumer is searching memory for "information" (which has definite cognitive overtones). In fact, the consumer may be searching his or her memory for feelings and not cognitions. The implication is that consumers engage in memory searches and these are not necessarily "information" searches.


One of the primary empirical arguments presented by Zajonc for his separate systems view of affect is that mere exposure can cause an affective reaction. This newly formed affect is usually positive and is present even when there is no detectable change in the person's ability to remember the exposure (Harrison 1977; Kunst-Wilson and Zajonc 1980; Moreland and Zajonc 1977, 1979; Wilson 1979; Zajonc 1968). [It should be noted that the mere exposure effect does not appear to be as simple or as universal as one night conclude from reading Zajonc (1980). There have been attempts that have failed to replicate the ere exposure effect under similar but not identical conditions (see Mandler and Shebo 1983 and Obermiller 1985).] Though this mere exposure effect may not prove that affect and cognitions are indeed separate systems (see Birnbaum 1981; Birnbaum and Mellers 1979a, 1979b; Gordon and Holyoak 1983; Mellers 1981; Mandler and Shebo 1983; and Obermiller 1985), it does have interesting implications for consumer behavior.

Many advertising campaigns appear to be very effective in altering consumers' preferences, even when they have little if any relevant information content. The research cited by Zajonc would indicate that people form preferences (e.g., a brand preference) simply on the basis of familiarity. If this does occur, then it makes sense for an advertiser, in the absence of real product differences, to affect preferences through exposure. This certainly does not explain all of advertising's effect, but it does deserve further research.

Familiarity has been investigated in consumer behavior. The type of familiarity typically discussed refers to the amount of prior experience with or knowledge about a product or product class (for a discussion of common definitions and operationalizations of familiarity, see Sirgy 1981). The present discussion centers on a different meaning of the word "familiarity." The familiarity discussed here is more akin to the idea of recognition. Copeland in his classic 1923 article hypothesized a relationship between this type of familiarity and consumer buying habits. Re hypothesized that, in the absence of perceived brand differences, the consumer will make a choice on the basis of "consumer recognition" which he defined as "an acquaintance with the general standing of the brand" (p. 287). The evidence presented by Zajonc seems to support the reasoning of Copeland.


The major difference between the definition of affect presented earlier and common definitions of attitudes (e.g., Fishbein and Ajzen 1975, p. 6) is that affect is "physiologically based." It is obvious that the simple statement that affective preferences are "physiologically based" is inadequate since even cold cognitions certainly have a physiological bases somewhere in the central nervous systems- If this definition is to be adequate, there must be some way of differentiating the physiological basis for cold cognitions' from the physiological basis for affective preferences. The present section briefly discusses the unique physiological aspects of emotion and a few implications of the physiological nature of emotion.

The most compelling arguments for a separate systems view of cognition and affect come from studies in physiological psychology. Izard ( 1984 ) notes that:

The case for considering emotions as a separate system seems fairly well established at the neurophysiological-biochemical level. At this level, it is well known that some brain structures, neural pathways, and neuro-transmitters are relatively more involved than others with emotion expression, emotion experience or feelings, and emotion related behavior. (p. 19)

It is well established that emotions (one of which would be affect as defined earlier) have a strong physiological component. One need only look at the synonym for emotions -- "feelings" -- to see this connection. The term feelings implies that something is occurring in the persons physiological system. Any adequate theory of affect must address this physiological basis of emotion"

Much of the work on the physiological aspects of b has focused on its interaction with the autonomic nervous system. The autonomic nervous system is that portion of the nervous system that controls survival functions. The autonomic nervous system has historically been divided into two parts: the sympathetic nervous system which is primarily responsible for heightened activity levels and the parasympathetic nervous system which is responsible for maintenance, repair, and resource conservation (Bruce 1977). Generally it has been the sympathetic division of the autonomic nervous system that has been implicated in emotional behavior (for easy to read reviews of this interrelationship between the autonomic nervous system and emotion, see Sigg 1975 or Lang, Rice, and Sternbach 1972).

It 18 interesting that the strongest physiological effects of affect seem to occur either in the autonomic nervous system or in the bodily organs which are controlled by the autonomic nervous system. This becomes important when one notes that most of the work on classical conditioning has focused on unconditioned and conditioned responses which are also controlled by the autonomic nervous system. Though classical conditioning is not exclusively tied to the autonomic nervous system (McSweeney and Bierley 1984), the majority of evidence in this area shows that classical conditioning also has its strongest and most significant effects on those aspects of the body which are controlled by the autonomic nervous system. Since the autonomic nervous system seems to be a co ,l,on ground for both affect and classical conditioning, then it is possible that affect sight often have its bases in classical conditioning.

The idea of emotions being classically conditioned is not totally new. In 1947, Mower described emotions (with particular emphasis on fear) in terms of conditioned responses. Recently, researchers have begun to further investigate the role of classical conditioning in emotion. For example, Grings (1983) argued that "the conditioned stimulus is not so much a signal of what is to come as it is a new stimulus with a modified affective tone" (p. 205). Though all of the evidence is not in, it is likely that classical conditioning can be at the root of a wide range of human feelings -including consumer preferences. At this point in time, there is evidence to suggest that affect may result from classical conditioning (which, it could be argued, is not cognitive in nature). Since this is only speculative at this point, the need definitely exists for more research in this area.

Another implication of the link between affect and physiology relates to the idea of consumer drives. Many consumer drives are physiologically based (e.g., the drives associated with food products). If affect is interrelated with physiological events, then it is likely that preferences may change as physiological changes occur in the consumer's system. It is well documented that nutrition deficiency can dramatically alter an animal's food preference (for reviews of this phenomenon know as "specific hungers", see Overman 1976 and Rozin and Kalat 1971). In consumer behavior, this type of effect can be seen in the advice that a person should never go grocery shopping when he or she is hungry. The danger is that the hunger drives could distort the consumer's preferences for certain foods. There may be a conflict between thinking and feeling and the feeling may win out. This is one example of pow consumer preferences for some products (e.g., food products) could possibly vacillate because of physiological changes. Empirical work is needed which would investigate changes in preferences as a function of physiological changes, both in the realm of primary needs (e.g., food) and in the realm of secondary needs (such as love or companionship).


One of the most significant debates in contemporary psychology centers around the relationship between affect and cognition. The present paper discussed some of the most obvious implications of this debate. This debate has raised some intriguing issues for consumer researchers. As a starting for research in this area, the present paper has pointed a set of specific hypotheses that can be derived from the evidence presented in his debate. In proper form, these hypotheses can be stated as follows:

H1: A stimulus will affect a consumer's preference at the time of encoding and at the time of decoding;

H2: Under certain conditions, the bases for consumer preference formation will be inaccessible even when the preferences themselves are readily accessible;

H3: Under certain conditions, consumer preferences can result from simple familiarity with an object;

H4: Under certain conditions, consumer, preferences can occur simply as a result of classical conditioning;

H5: Under some conditions, consumer preferences can change as a result of physiological changes.

These hypotheses provide specific direction for future research. It is clear that research is needed that will show that these hypotheses either possess or lack validity. In addition, the "under some conditions" clauses of these hypotheses need to be addressed. The key question here is "under what conditions?"

It is hoped that this paper will encourage researchers to take a closer look at the relationship between cognition, affect, and consumer preferences. Though it may be premature to conclude either that affect is always postcognitive or that affect can occur independently of cognitive processing, consumer researchers should address some of the issues which have recently emerged in the debate surround the relationship between affect and cognition. The role of affect in consumer preferences is probably much more significant and much more varied than contemporary consumer behavior theories would suggest.


Baars, Bernard J. (1981), "Cognitive Versus Inference," American Psychologist, 36 (February), 223-224.

Bagozzi, Richard P. (1983), -A Holistic Methodology For Modeling Consumer Response to Innovation," Operations Research. 31 (January-February), 128-176.

Beattie, Ann E. and Andrew A. Mitchell (1985), "The Relationship Between Advertising Recall and Persuasion: An Experimental Investigation," in Psychological Processes and Advertising Effects> Linda F. Alwitt and Andrew A. Mitchell (ed.). Hillsdale: Erlbaum, 129-156.

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

Birnbaum, Michael H. (1981), "Thinking and Feeling: A Skeptical Review, American Psychologist, 36 (January), 99-101.

Birnbaum, Michael H. and Barbara A. Mellers (1979a), "Stimulus Recognition May Mediate Exposure Effect, Journal of Personality and Social Psychology, 37 (March), 391-394.

Birnbaum, Michael H. and Barbara A. Mellers (1979b), "One-Mediator Model of Exposure Effect is Still Viable," Journal of Personality and Social Psychology, 37 (June), 1090-1096.

Bruce, Richard L. (1977), Fundamentals of Physiological Psychology, New York: Holt, Rinhart, and Winston.

Copeland, Melvin T (1923), "Relation of Consumer's Buying Marketing Methods, Harvard Business Review, 1, 282-289.

Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, and Behavior: An Introduction to Theory and Reading, MA: Addison-Wesley.

Fiske, Shelley E., and Susan T. Taylor (1984), Social Cognition, Reading, MA: Addison-Wesley.

Fiske, Shelley E., Gordon, Peter C. and Keith J. Holyoak (1983), "Implicit Learning and Generalization in the 'Mere Exposure' Effect," Journal of Personality and Social Psychology, 45 (March), 492-500.

Grings, William W. (1983), Comments On A. Levey and I. Martin's 'Cognitions, Evaluations, and Conditioning,'" Advances in Behavior Research and Therapy, 4 (4), 205-207.

Harrison, Albert A. (1977), He re Exposure," Advances in Experimental Social Psychology, 10, 39-83.

Izard, Carroll E. (1984), Emotion-cognition Relationships and Human Developments," in Emotions, Cognition and Behavior, Carroll E. Izard, Jerome Kagan, and Robert B. Zajonc (eds.), Cambridge: Cambridge University Press, 17-37.

Izard, Carroll E. (1977), Human Emotions, New York: Plenum Press.

Izard, Carroll E., Jerome Kagan and Robert B. Zajonc (1984), Introduction, in Emotions, Cognition" and Behavior, Carroll E. Izard, Jerome Kagan. and Robert Zajonc (eds.). Cambridge: Cambridge University Press, 1-16.

Kunst-Wilson, William R. and Robert B. Zajonc (1980), "Affective Discrimination of Stimuli That Cannot Be Recognized" Science, 208 (February) 557-558.

Lang, Peter J., David G. Rice, and Richard A. Sternbach (1972), The Psychophysiology of Emotion, in Handbook of Psychophysiology, Norman S. Greenfield and Richard A. Sternbach (ed . ), 623-644.

Lazarus, Richard S. (1981), A Cognitivist's Reply to Zajonc on Emotion and Cognition," American Psychologist, 36 (February). 222-223.

Lazarus, Richard S. (1982), Thoughts on the Relations Between Emotion and Cognition," American Psychologist, 37 (September), 1019-1024.

Lazarus, Richard S. (1984), "On the Primacy of Cognition, American Psychologist, 39 (February), 124-129.

Lictenstein, Meryl and Thomas R, Srull (1985), Conceptual and Methodological Issues in Examining the Relationship Between Consumer Memory and Judgment, in Psychological Processes and Advertising Effects, Linda P. Alwitt and Andrew A. Mitchell (ed.), Hillsdale: Erlbaum, 113-128.

Lynch, John G. Jr. (1985), "Uniqueness Issues in the Decompositional Modeling of Multi-attribute Overall Evaluations: An Information Integration Perspective, Journal of Marketing Research, 22 (February), 1-19.

Mandler, George and Billie Jo Shebo (1983), Knowing and Liking, Motivation and Emotion, 7 (June), 125-144.

McSweeney, Frances R. and Calvin Bierley (1984), "Recent Developments in Classical Conditioning, Journal of Consumer Research, 11 (September), 619-631.

Mellers, Barbara, A. (1981), -Feelings More Than Thinking," American Psychologist, 36 (July), 802-803.

Moreland, Richard L., and Robert B. Zajonc (1977), Is Stimulus Recognition a Necessary Condition For the Occurrence of Exposure Effects?" Journal of Personality and Social Psychology, 35 (June), 191-199.

Moreland, Richard L. and Robert B. Zajonc (1979), Exposure Effects May Not Depend On Stimulus Recognition, Journal of Personality and Social Psychology, 37 (June), 1085-1089.

Mower, 0.R. (1947), "On the Nature of Learning -A Reinterpretation of 'Conditioning' and Problem Solving," Harvard Educational Review, 17 (Spring), 102-148.

Obermiller, Carl (1985), "Varieties of Mere Exposure: The i>as Effects of Processing Style and Repetition on Affective Response," Journal of Consumer and Research, 2 (June), 17-30.

Overman, Stephen R. (1976), "Dietary Self-Selection by Animals," Psychological Bulletin, 83 (2), 218-235.

Rachman, S. (1981), -The Primacy of Affect: Some Theoretical Implications," Behavior Research and Therapy, 19 (4), 279-290.

Ray, Michael, L. (1973), "Marketing Communications and the Hierarchy-of-Effects, in New Models for Mass Communications Research, Volume 2, Peter Clarke (et.), Beverly Rills: Sage, 147-176.

Rozin, Paul and James W. Kalat (1971)-, Specific Hungers and Poison Avoidance as Adaptive Specializations of Learning, Psychological Review, 78 (November ), 459-486.

Sigg, E.B. (1975), The Organization and Function of the Central Sympathetic Nervous System, in Emotions: Their Parameters and Measurement, Lennart Levi (ed.), New York: Raven Press. 93-122.

Sirgy, H. Joseph (1981), Product Familiarity: Critical Comments on Selected Studies and Theoretical Extensions, Advances in Consumer Research, 8, 156-160.

Slife, Brent D. (1981), The Primacy of Affective Judgment From a Teleological Perspective, American Psychologist, 36 (February), 221-222.

Tomkins, S.S. (1962), Affect, Imagery, Consciousness: Volume I, The Positive Affect, New York: Springer.

Tomkins, S.S. (1962), Affect Imagery Consciousness: Volume 2, The Negative Affect, New York: Springer.

Watts, Fraser N. (1983), "Affective Cognition: A Sequel to Zajonc and Rachman," Behavior Research and Therapy, 21 (1), 89-90.

Wilson, William R. (1979), "Peeling More Than We Can know: Exposure Effect Without Learning," Journal of Personality and Social Psychology, 37 (June), 811-821.

Wilson, Timothy D. and Richard D. Nisbett (1978), -The Accuracy of Verbal Reports About the Affects of Stimuli on Evaluations and Behavior," Social Psychology, 41 (June), 118-131.

Zajonc, Robert B. (1968). "Attitudinal Effects of Here Exposure," Journal of Personality and Social Psychology: Monograph, 9 (2, Part 2), 1-27.

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

Zajonc, Robert B. (1981), "A One-Factor Mint About Mind and Emotion," American Psychologist, 36 (January), 102-103.

Zajonc, Robert B. (1984), "On the Primacy of Affect, American Psychologist, 39 (February), 117-123.

Zajonc, Robert B. and Hazel Markus (1982), "Affective and Cognitive Factors in Preference," Journal of Consumer Research, 9 (September), 123-131.