Feeling About Thinking: The Role(s) of Affect in Social Cognition

A review of

[book cover]

Cognition and Emotion
by Eric Eich, John F. Kihlstrom, Gordon W. Bower, Joseph P. Forgas, and Paula M. Niedenthal

New York: Oxford University Press, 2000. 259 pp. ISBN 0-19-511334-9. $24.00 paperback

Reviewed by
Michael D. Robinson
Paul D. Rokke
Maya Tamir

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #1 Cognition and Emotion is a concise and affordable text with chapters written by important figures within the mood and memory literature. To those familiar with this literature, the book is likely to be a very quick read. The predominant theoretical focus is on Bower's (1981) network model of mood and cognition. This model assumes that mood states influence the accessibility of memories because they are part of a common semantic network. For example, the emotion of sadness would be linked to memories of loss, loneliness, and rejection. As a consequence of these links in semantic memory, temporary feelings of sadness would activate or prime memories related to loss and rejection. In principle, such increased activation would influence cognitive processes related to encoding (i.e., perception, attention, and categorization) as well as those related to memory retrieval and judgment.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #2Chapters by Eich and Schooler, Bower and Forgas, and Niedenthal and Halberstadt are in part devoted to an assessment of Bower's network model some 20 years after it was proposed. However, these chapters go in a variety of directions. In addition, the chapter by Kihlstrom and colleagues is quite different from the others. For these reasons, it is useful to review each chapter separately.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #3The chapter by Eich and Schooler is puzzling. It is apparently the authors' intention to both (a) provide an overview of the upcoming chapters and (b) present their own contributions to the literature. However, perhaps because the chapter is relatively short, it did not succeed in either of these aims.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #4The chapter by Kihlstrom, Mul-vaney, Tobias, and Tobis is particularly wide-ranging, covering historical sources as well as modern ones and touching on topics like amnesia, hypnosis, repression, and pain perception. The central issue is whether a person could experience emotions without being aware of them. Neurologically, such a possibility is not far fetched. For example, Panksepp (1998) has proposed that emotions are the product of subcortical circuits, whereas awareness of emotions is the product of cortical brain areas. As such, dissociations between experience and awareness are quite possible. In clinical research, other authors have proposed that both repression (Shedler, Mayman, & Manis, 1993) and alexithymia (Lane, Ahern, Schwartz, & Kaszniak, 1997) could be associated with a disconnect between experience and awareness. After reviewing the literature, Kihlstrom and colleagues appropriately conclude that the case for unconscious emotion is “not yet convincing” (p. 67). Nevertheless, the chapter is stimulating and thought provoking.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #5The chapter by Bower and Forgas has two distinct halves. The first half reads like a retrospective assessment of the success of the Bower (1981) model of mood influences on cognition. Because the citations primarily concern papers in the 1980s and early 1990s, one gets the sense that Bower's (1981) model has been somewhat (but not wildly) successful and that active research in this area is dwindling a bit. The second half of the chapter, however, cites mostly more recent papers and presents an optimistic view of the ability of Forgas's (1995) affect infusion model, introduced as an extension of the Bower (1981) model, to explain nearly any pattern of data. The authors are relatively persuasive concerning one of the central predictions of the model, namely, that mood-congruent effects on memory and cognition are more likely to occur when the task involves elaboration than when the task involves direct retrieval from long-term memory. On the other hand, we feel that there are ambiguities related to Forgas's other two processing strategies labeled “heuristic” and “motivated.” The heuristic strategy is one that minimizes the scope of a serious alternative to Bower's (1981) model (Affect as Information, AIM) to hedonically unimportant decisions. However, it is not at all clear that this is justified (Clore & Tamir, 2002). The “motivated” strategy is one that can be invoked, post hoc, to account for failures to obtain mood-congruent memories or judgments. Although the integrative nature of Forgas's (1995) AIM model is admirable, we believe that many ambiguities remain.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #6Next, is a chapter by Niedenthal and Halberstadt. On the basis of numerous failures to obtain mood effects on perception, Bower (1987) suggested that he may have been wrong concerning this set of predictions. Niedenthal and colleagues, however, have been successful in obtaining such effects, as documented in this chapter. After an informative discussion of the challenges involved in manipulating mood states, the authors show that mood states can influence measures of lexical access (lexical decision, pronunciation). Nevertheless, the effects are rather circumscribed. For example, a pleasant (vs. unpleasant) mood state allows one to recognize the word joy quickly, but does not have implications for positive words that are unrelated to mood states (e.g., wisdom). Subsequent to describing these lexical access studies, Niedenthal and Halberstadt discuss a second program of research, culminating in a Psychological Review article in 1999, showing that mood states, whether pleasant or unpleasant (vs. neutral) result in an enhanced tendency to categorize objects in terms of hedonic (vs. cognitive) considerations. The latter results are not consistent with Bower's (1981) model because they are not mood-congruent tendencies. The chapter is relatively short, but important, pointing to some new directions in research on mood and cognition.

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-1 #7In the concluding discussion chapter, Eich asked his coeditors a series of 15 questions and documented their answers. The questions range from general ones related to the status of the field to specific ones related to individual chapters. As Eich (p. 204) notes, the responses tended to be honest and revealing. Bower and Kihlstrom offer balanced, historical reviews of the literature. By contrast, Forgas is especially optimistic about the state of the field, whereas Niedenthal is especially pessimistic. Space precludes a complete account of this chapter. Suffice it to say that (a) the editors disagree somewhat on the state of the field and (b) the chapter is a notable strength of the book, especially for those seeking an “insider's” perspective.

Strengths and Limitations

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-2 #1 Given the number of chapters and their coverage, Cognition and Emotion should not be viewed as a comprehensive summary of the literature on cognition and emotion. For example, the large literature on anxiety- and depression-related disorders is not covered extensively. For a more comprehensive volume, readers are encouraged to consider the Handbook of Cognition and Emotion edited by Dalgleish and Power (1999). Cognition and Emotion, however, is a concise and comprehensive survey of Bower's (1981) network model and how it has faired in the social cognition literature. Nevertheless, even within this more circumscribed literature, there are some gaps in coverage as well as theoretical issues that deserve commentary.

 Affect or Affective Concepts?

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-2 #2There are unresolved issues related to the nature of mood effects on cognition and judgment. When participants are encouraged to become “happy” or “sad” or think about positive or negative thoughts, the resulting effects on cognition may be the result of priming affective concepts rather than mood states. Indeed, Wyer, Clore, and Isbell (1999) have proposed that mood states, in and of themselves, do not influence the accessibility of positive versus negative memories. Rather, they claim that thoughts about mood states (e.g., “I seem to be happy”) do seem to play a role in activating other valence-congruent thoughts. Such a model is counter to Bower's (1981) network model. The book gives little attention to such important issues, which will likely receive increased attention in the future.

 Spreading Activation?

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-2 #3Bower's (1981) model likens mood states to nodes within a semantic network. Within this framework, mood states will automatically (i.e., without reflection) trigger mood-consistent thoughts. Such spreading activation assumptions should be tested with reaction times. Instead, researchers have used complex social judgments and behaviors as well as untimed autobiographical recall. In addition, they have encouraged elaborative, extended processing. It is not at all clear that spreading activation in semantic memory accounts for such mood effects (Clore & Tamir, 2002). Indeed, if Bower's model really does underlie the diversity of effects reported by Bower and Forgas, then it would seem desirable to focus directly on the activation process itself rather than on such “distal” phenomena such as social judgments. For a discussion of alternative theories of mood effects on social cognition, see Martin and Clore (2001).

 Personality Effects

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-2 #4Mood states do not produce the same cognitive effects among all people (Rusting, 1998). For example, mood-congruent effects are more likely among people who value (vs. do not value) their emotions (Gasper & Clore, 2000). In addition, it is becoming increasingly clear that emotional states interact with emotional traits in influencing cognition and judgment. Although we are just beginning to understand these interactions (Rusting, 1998), it seems likely that work will lead to new models of state effects on cognition (e.g., Tamir, Robinson, & Clore, 2002). Cognition and Emotion is relatively silent on personality by mood interactions.

 Emotion Regulation

Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-2 #5 Null (and mood-incongruent) findings are probably more common in this area of research than we would like to admit. Given such outcomes, it would be possible to describe them as occurring because of “motivated” processing or mood regulation. However, the processes underlying mood regulation have only recently begun to receive systematic treatment. If mood regulatory processes are really as common as some people claim (e.g., Davidson, 1999), then it may not be easy to disentangle the faciliatory effects of mood from regulatory processes. Cognition and Emotion does not meaningfully address mood regulation.


Robinson, Michael D.; Rokke, Paul D.; Tamir, Maya S-3 #1 We hope that our review has focused on both the strengths and weaknesses of Cognition and Emotion. It is a very useful summary of several programs of research that have followed Bower's (1981) influential network theory of mood effects. The authors are experts in the field. All of the chapters are accessible, yet scholarly. The question-and-answer chapter at the end of the book is particularly revealing. On the other hand, the chapter by Kihlstrom and colleagues, although terrific, addresses topics that are quite different from the other chapters. There is little discussion of personality or emotion regulation. Finally, some key issues in the field receive only cursory treatment. For example, there is little discussion of alternative models. Bower's (1981) model has been tremendously useful in generating research. However, we believe that the future will belong to more constrained models linking mood states to particular stages of cognitive processing.


Bower, G. H.. (1987). Commentary on mood and memory. Behaviour Research and Therapy, 25, 443-455.

Clore, G. L., & Tamir, M.. (2002). Affect as embodied information. Psychological Inquiry, 13, 37-45.

Dalgleish, T., & Power, M. J.. (1999). Handbook of cognition and emotion. New York: Wiley.

Davidson, R. J.. (1999). Neuropsychological perspectives on affective styles and their cognitive consequences. In T. Dalgleish & M. J. Power (Eds.), Handbook of cognition and emotion (pp. 103–123). New York: Wiley.

Forgas, J. P.. (1995). Mood and judgment: The Affect Infusion Model (AIM). Psychological Bulletin, 117, 39-66.

Gasper, K., & Clore, G. L.. (2000). Do you have to pay attention to your feelings to be influenced by them? Personality and Social Psychology Bulletin, 26, 698-711.

Lane, R. D., Ahern, G. L., Schwartz, G. E., & Kaszniak, A. W.. (1997). Is alexithymia the emotional equivalent of blindsight? Biological Psychiatry, 42, 834-844.

Martin, L. L., & Clore, G. L.. (2003). Theories of mood and cognition: A user's guidebook. Mahwah, NJ: Erlbaum.

Niedenthal, P. M., Halberstadt, J. B., & Innes-Kerr, A. H.. (1999). Emotional response categorization. Psychological Review, 106, 337-361.

Panksepp, J.. (1998). Affective neuroscience: The foundations of human and animal emotions. New York: Oxford University Press.

Rusting, C. L.. (1998). Personality, mood, and cognitive processing of emotional information: Three conceptual frameworks. Psychological Bulletin, 124, 165-196.

Shedler, J., Mayman, M., & Manis, M.. (1993). The illusion of mental health. American Psychologist, 48, 1117-1131.

ERIC EICH, Department of Psychology, University of British Columbia, Vancouver, British Columbia, Canada.

JOHN F. KIHLSTROM, Department of Psychology, University of California, Berkeley.

GORDON W. BOWER, Department of Psychology, Stanford University.

JOSEPH P. FORGAS, School of Psychology, University of New South Wales, Sydney, New South Wales, Australia.

PAULA M. NIEDENTHAL, Department of Psychology, Indiana University.

MICHAEL D. ROBINSON and PAUL D. ROKKE, Department of Psychology, North Dakota State University, Fargo, North Dakota 58105. E-mail: michael.robinson@ndsu.nodak.edu or paul.rokke@ndsu.nodak.edu

MAYA TAMIR, Department of Psychology, University of Illinois at Urbana-Champaign, 308 Psychology Building, mc716 61820. E-mail: tamir@s.psych.uiuc.edu

0010-7549© 2003 by the American Psychological Association
Previously published in Contemporary Psychology: APA Review of Books, June 2003, Vol. 48, No. 3, 356-358

For personal use only--not for distribution.