ATTITUDES AND SOCIAL COGNITION Happiness

ATTITUDES AND SOCIAL COGNITION Happiness and Stereotypic Thinking in Social Judgment Galen V. Bodenhausen, Geoffrey P. Kramer, and Karin Siisser Four experiments examined the effects of happiness on the tendency to use stereotypes in social judgment. In each experiment, individuals who had been induced to feel happy rendered more stereotypic judgments than did those in a neutral mood. Experiment 1 demonstrated this phenomenon with a mood induction procedure that involved recalling life experiences. Experiments 2 and 3 suggested that the greater reliance on stereotypes evident in the judgments of happy individuals was not attributable to cognitive capacity deficits created by intrusive happy thoughts or by cognitively disruptive excitement or energetic arousal that may accompany the experience of happiness. In Experiment 4, happy individuals again were found to render more stereotypic judgments, except under conditions in which they had been told that they would be held accountable for their judgments. These results suggest that although happy people's tendency to engage in stereotypic thinking may be pervasive, they are quite capable of avoiding the influence of stereotypes in their judgments when situational factors provide a motivational impetus for such effort.

Discovering the conditions under which group stereotypes are likely to be applied in forming impressions of and making judgments about individuals has been an issue of perennial interest in social psychology. Factors such as information overload (Pratto & Bargh, 1991; Stangor & Duan, 1991) and task difficulty (Bodenhausen & Lichtenstein, 1987), for example, have been shown to increase the social perceiver's reliance on stereotypic preconceptions (for a review, see Hamilton & Sherman, in press). In the present research, we investigated the role of emotion, specifically happiness, in the application of stereotypes during social information processing. Does being happy have any impact on the likelihood of stereotyping others? If so, what is the mechanism involved? It was these questions that we sought to address. Interest in the relationship between emotion and stereotyping is certainly not new. However, previous attempts to understand the role of affective experience in prejudice and stereotyping have focused almost exclusively on the impact of negative emotions. Conventional wisdom indicates that it is during times of stress, anxiety, or hostility that prejudice and stereotypes are most likely to emerge and exert their influence on social perception. Psychological research lends credence to the idea that anger, conflict, frustration, and anxiety are indeed associated with

the tendency to see the world through a stereotypic lens (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950; Dollard, Miller, Doob, Mowrer, & Sears, 1939; Greenberg et al., 1990; Greenberg, Simon, Pyszczynski, Solomon, & Chatel, 1992; Peak, Muney, & Clay, 1960; Sherif & Sherif, 1953; Stephan & Stephan, 1985; Wills, 1981). From the standpoint of the various theoretical ideas that are represented by the array of research addressing the link between various negative emotions and the phenomena of prejudice and stereotyping, negative affect often serves as a motivational impetus for these processes. Whether by displacement, projection, rationalization, or some other defensive process, individuals in negative states may derive some measure of relief or gratification by expressing their prejudices and stereotypes toward outgroups. According to these approaches, then, negative affect is the fuel for the fire of prejudice and stereotyping. By implication, the experience of happiness or contentment should provide little motivational impetus to view others in stereotypic terms. Perhaps it is during times of happiness that we are most likely to look beyond our gross generalizations about outgroups. Contemporary research on stereotyping has often deemphasized the motivational aspects of the phenomenon that were so central to many earlier conceptualizations. Instead, the focus has been on stereotyping as a mundane cognitive function that provides simplification and structure to our subjective experience of the complex social milieu in which we live (Fiske & Neuberg, 1990; Hamilton, 1979;Rothbart, 1981). Several studies suggest that stereotypes operate as heuristic cues in social information processing, providing a basis for a quick response to members of outgroups that may suffice whenever social perceivers cannot, or prefer not to, engage in a more thoughtful analysis of the unique personal qualities of specific outgroup members (Bodenhausen, 1988, 1990; Bodenhausen & Lichtenstein, 1987; Bodenhausen & Wyer, 1985; see also Chaiken,

Galen V. Bodenhausen, Department of Psychology, Michigan State University; Geoffrey P. Kramer, Department of Social and Behavioral Sciences, Indiana University-Kokomo; Karin Siisser, Department of Psychology, Vanderbilt University. We are grateful to Linda Jackson and the anonymous reviewers for their helpful comments on a previous version of this report. Correspondence concerning this article should be addressed to Galen V. Bodenhausen, Department of Psychology, Michigan State University, East Lansing, Michigan 48824-1117. Electronic mail may be sent to [email protected].

Journal of Personality and Social Psychology, 1994, Vol. 66, No. 4, 621-632 Copyright 1994 by the American Psychological Association, Inc. 0022-3514/94/S3.00

621

622

G. BODENHAUSEN, G. KRAMER, AND K. SUSSER

Liberman, & Eagly, 1989; Fiske & Neuberg, 1990; Gilbert & Hixon, 1991). From this perspective, stereotyping is not motivated by a need to cope with negative affective experience so much as a need or desire to avoid extensive cognitive work. In essence, stereotyping represents a functional shortcut strategy for social information-processing tasks, a strategy that in fact frees up cognitive resources for use on other tasks (Macrae, Milne, & Bodenhausen, 1994). What are the implications of happiness for stereotyping, when viewed from the framework of contemporary social-cognitive research? Although some recent research has begun to investigate the impact of happiness on memory for and judgments about social groups (e.g., Hamilton, Stroessner, & Mackie, 1993; Stroessner, Hamilton, & Mackie, 1992; Stroessner & Mackie, 1992, 1993; see Hamilton & Sherman, in press, for a review), there has been very little direct evidence about how happiness might affect the use of stereotypes in judgments of individual group members. Two lines of research can be identified that have some relevance to an understanding of how happiness and stereotyping might be related in this context. One of these is research on the mood congruency effect in social judgment (e.g., Bower, 1991; Forgas & Moylan, 1987; Isen, 1987). Essentially, this effect involves rendering judgments that are biased in the direction of one's prevailing mood. Happy people thus would be expected to make more positive, favorable judgments. Various models have been proposed to account for why this would happen. The mood-as-information approach taken by Schwarz and Clore (1983, 1988) proposes that when people are asked to make a judgment about some object (X), they often use their gut reaction, asking themselves "How do I feel about X?" As long as it is plausible to them that their current feelings (whatever their actual source) are at least partly a reaction to X, they will tend to make judgments that are colored by their momentary mood. At least insofar as negative stereotypes are concerned, this implies that happy people may set aside their unfavorable stereotypes of outgroups and reach more positive assessments of members of these groups than do those in more neutral moods. This approach is similar to the historical motivational approaches in predicting less use of negative stereotypes among those who are feeling happy. A second line of contemporary research has different implications. A growing body of evidence addressing the relationship between emotional experience and social information processing suggests that the type of strategy people use in performing a social judgment task may be determined in part by their momentary emotional state (for reviews, see Bodenhausen, 1993; Clark & Williamson, 1989; Clore, Schwarz, & Conway, in press; Forgas, 1992a, 1992b; Schwarz, 1990; Sinclair & Mark, 1992). Interestingly, the state of happiness has been associated repeatedly with the use of more superficial or cursory styles of thinking. Some of the most compelling evidence supporting this proposition has emerged from research on mood and persuasion (Mackie & Worth, 1989, 1991; Schwarz, Bless, & Bohner, 1991; Worth & Mackie, 1987). These studies document that when people are made happy prior to the presentation of a persuasive message, they are less affected by variations in argument quality. Presumably, systematic analysis of a persuasive message would result in the detection and rejection of specious or weak

arguments, but happy people appear to accept weak messages just as readily as those founded on stronger, more valid arguments. Moreover, happy people seem to be more attuned than those in a neutral or sad mood to simple heuristic cues present in the persuasion situation; such cues permit a relatively quick and easy response to the persuasive appeal (cf. Chaiken, 1980, 1987; Eagly & Chaiken, 1993). Credibility cues such as expertise and trustworthiness are classic examples of cues that form one basis for responding to a persuasive communication that may largely circumvent cognitive elaboration of the content of the message (Petty & Cacioppo, 1986). Thus, it appears that happy people often prefer to base their reactions to a persuasive message more on simple cues such as the communicator's putative expertise rather than engaging in close scrutiny of the message content (see also Petty, Gleicher, & Baker, 1991). The implications of this research for predictions about happiness and stereotyping are provocative. It may be that happy people, who show a preference for simple heuristic cues in persuasion situations, may also tend to rely more on simplistic stereotypes in other social judgment situations. This prediction is at odds with both conventional wisdom about the relationship between mood and stereotyping and the implications of the traditional motivational theories of prejudice and stereotyping. It also runs against the implications of the mood congruency effect. According to previous theory and research, then, it is somewhat unclear whether happiness will promote or inhibit stereotypic thinking in the social perception of individuals who happen to be members of stereotyped social groups. There are precedents for expecting both outcomes. The initial experiment reported in this article was intended to provide further evidence on this issue. After documenting whether stereotyping was augmented or reduced among happy people, a series of follow-up studies was planned to examine mechanisms that might account for the observed effects.

Experiment 1 Concern about the impact of social stereotypes on judgment processes has often focused on particularly consequential kinds of judgment situations, such as decision making in organizational personnel contexts (e.g., Terborg & Ilgen, 1975) and criminal justice contexts (e.g., Bodenhausen & Kramer, in press). It is, of course, a matter of considerable importance to determine when and how members of stereotyped social groups are disadvantaged when it comes to perceptions of their guilt, their deservingness of parole, their likelihood of recidivism, their credibility as witnesses, and countless other perceptions relevant to decision making in justice contexts. In the research reported in this article, we focused on stereotyping in a situation that was analogous to a criminal justice setting. The experiments we undertook examined the role of happiness and stereotyping in students' perceptions of and judgments about their fellow students' alleged misconduct. Method Overview and Design We asked participants to engage in two ostensibly unrelated tasks. The first task, which involved recalling prior experiences, was designed

623

HAPPINESS AND STEREOTYPIC THINKING to induce a positive mood in approximately one half of the participants. Subjects then completed a social judgment task in which they read a synopsis of a disciplinary hearing and made judgments about the case. There were two different scenarios, one involving a case of alleged assault and one involving a case of alleged cheating. The identity of the student accused of each offense was varied so that he was either a member of a group stereotypically associated with the offense or was not. Thus, the experiment was a 2 (affect: happy vs. neutral) X 2 (stereotype: present vs. absent) between-subjects design that was tested within two different scenarios and stereotypes.

Subjects Participants were 94 undergraduates (21 men and 73 women) who received course credit in return for their participation.

Materials and Procedures Subjects participated in groups of approximately 8. Upon arrival at the laboratory, participants were greeted by an experimenter who told them that they would be involved in two different experiments, one entitled "Mood and Memories" and one entitled "Students' Court." They were seated at individual cubicles, the basic nature of thefirsttask was explained, and subjects signed a consent form. Mood induction. One half of the experimental sessions were randomly assigned to the happy condition. Subjects in this condition were told that the researchers were interested in the relationship between emotion and memory, and that today they wanted to understand the "psychological structure" of happy memories. To do this, subjects were asked to recall, reexperience, and write about an event that had made them particularly happy. Following Strack, Schwarz, and Gschneidinger (1985), the instructions emphasized focusing on concrete, vivid, experiential aspects of the event rather than on an abstract or objective assessment of it. Strack et al. found this procedure to be most successful in inducing a state of happiness. Subjects in the other condition were told that the research involved the psychological structure of everyday memory and were asked to recall and describe the mundane events of the previous day. Approximately 12 min were allotted for the completion of this task. After subjects had finished writing, the experimenter thanked them, collected their forms, and left. Social judgment task. A new experimenter introduced the second task, which was described as a study of legal socialization. The experimenter explained that some institutions of higher learning had adopted adjudicatory systems in which students took responsibility for disciplinary proceedings. To examine how this kind of system might work on their own campus, the subjects were asked to take on the role of a student member of a peer disciplinary review panel. They were asked to read a case allegedly taken from another college campus and make decisions about the appropriate response to the case. After signing a consent form for the "Students' Court" study, all subjects received a booklet containing (a) a participants characteristics questionnaire, (b) a case description, (c) a brief questionnaire about the case, and (d) a probe for suspicions about the purpose of the study. The demographic questionnaire requested information such as subjects' age and sex, and asked them to rate themselves on several characteristics. Each rating involved circling a number from 0 (not at alt) to 7 (extremely) that best reflected how applicable the characteristic was to them at the moment. Embedded within fillers (e.g., conscientious, bored) was an item designed to assess the effectiveness of the mood manipulation (happy). Subjects received one of two cases, involving either an allegation of assault (beating up a roommate) or cheating (on a mathematics examination). The case summary consisted of approximatelyfiveor six sentences detailing the nature of the accusation and providing a mixture of

evidence, some implying innocence and some implying guilt. The identity of the student accused in each case was manipulated so that, for half the subjects, he was identified as a member of a group stereotypically associated with the alleged offense (stereotype condition), whereas for the remainder he was not (no stereotype condition). Specifically, for the case of assault, the student-defendant was given either an obviously Hispanic name (Juan Garcia) or an ethnically nondescript name (John Garner); for the cheating case, in one half of the booklets the target's name was followed by the phrase "a well-known track-and-field athlete on campus," whereas in the others, this phrase was omitted. A high prevalence of endorsement of these stereotypes within the targeted subject population was established in previous research (see Bodenhausen, 1990). Within sessions, subjects were randomly assigned to one of the four possible combinations of offense type and stereotype activation. After reading the case, subjects were asked to report the likelihood of the accused student's guilt on an 11-point scale ranging from 0 (not at all likely) to 10 (extremely likely). They were also asked to complete somefillerquestions designed to match the alleged purpose of the experiment (e.g., To what extent could a peer discipline system work on our campus?). Upon completion of the case questionnaire, subjects were probed for suspicions about the research, then they were given an educational debriefing and dismissed.

Results and Discussion Manipulation Check Effectiveness of the mood manipulation was assessed by comparing subjects' self-ratings of happiness in the happy and neutral mood conditions. As intended, subjects in the happy condition rated themselves as significantly happier than their neutral mood counterparts (Ms = 4.29 and 3.72, respectively), t(92) = 2.]4,p< .025, one-tailed. Perceived Guilt If traditional motivational approaches to stereotyping and prejudice are correct in asserting that these phenomena are driven by a need to cope with negative states, then we might expect those who are feeling happy to be particularly unlikely to show evidence of stereotypic thinking. On the other hand, if happy people are more inclined to use mental shortcuts, it may be that happy people stereotype more than those in a neutral mood. Data relevant to this issue are depicted in Figure 1. The

HAPPY

NEUTRAL AFFECT CONDITION

I STEREOTYPE H NO STEREOTYPE

Figure 1. Mean perceived guilt as a function of stereotype activation and emotional state (Experiment 1).

624


data were collapsed across offense type, as this variable exhibited no interactions with other independent variables, ps > .25. Inspection of Figure 1 reveals that, among neutral mood subjects, stereotyped targets were treated no differently than nonstereotyped targets, suggesting that these subjects based their judgments on the implications of the specific case evidence provided. However, happy mood subjects'judgments of the stereotyped target were significantly more negative than their judgments of the nonstereotyped target, p < .01. Overall, the interaction of mood and stereotype activation was marginally significant, F(l, 90) = 2.88, p < .09. Consistent with the persuasion literature described previously, these data show a greater reliance on stereotypes among happy subjects. Conventional wisdom notwithstanding, it seems that members of stereotyped social groups can be disadvantaged as a result of the positive moods of social perceivers. It is unclear, however, exactly why this is so. The remaining experiments were undertaken with the goal of better understanding the process producing more stereotypic judgments among those who are feeling happy. Experiment 2 Why might happiness induce a greater tendency to stereotype, or to use any other kind of cognitive shortcut for that matter? Explanations for social perceivers' use of heuristic processing strategies typically focus on limitations of momentary cognitive capacity or motivation (Chaiken et al., 1989; Fiske & Neuberg, 1990; Petty & Cacioppo, 1986). One possibility that seems particularly plausible in accounting for the results of Experiment 1 might be called the cognitive distraction hypothesis. According to this view (cf. Mackie & Worth, 1989), happy people have less cognitive capacity for social information processing tasks because a portion of their cognitive resources is eaten up by rumination on happy thoughts and the life events that have evoked one's current happiness. As a result of diminished capacity for more extensive processing, perceivers rely to a greater extent on mental shortcuts, when available, to guide their reactions to social stimulus input. This view seems particularly plausible in accounting for the findings reported in Experiment 1 because the mood induction required subjects to fill their minds with happy thoughts. It is quite likely that these happy thoughts are retained longer in short-term memory (the "workspace") than the neutral, mundane thoughts elicited among control group subjects, and these persisting thoughts may well distract the social perceiver from devoting full attention to other social perception tasks at hand. In ourfirstconceptual replication of Experiment 1, we sought to determine the validity of this capacity-based argument as an explanation for happy subjects' greater reliance on social stereotypes. Experiment 2 sought specifically to address the importance of distracting thoughts in producing the tendency among happy people to exhibit more stereotypic judgments. In this experiment, we used a mood induction procedure that does not require the imposition of any cognitive load. Specifically, we used a facial feedback procedure to directly produce subjective happiness. Previous research shows that contraction of facial muscles into poses associated with the expression of an emotion can also induce the subjective experience of the particular emo-

tion (Adelman & Zajonc, 1989; Duclos et al., 1989; Ekman, Levenson, & Friesen, 1983), and this has been found even when using an elaborate and clever ruse to eliminate demand characteristics (Strack, Martin, & Stepper, 1988). If this procedure results in levels of subjective happiness similar to those reported by subjects who were made happy with the memory elicitation procedure (Experiment 1), but it is not associated with any greater level of stereotyping, this would provide support for the notion that it is distracting happy thoughts that produce a greater reliance on stereotypes. This pattern would be consistent with the claim that only when happiness is accompanied by rumination about specific happy circumstances does it promote a tendency to stereotype others. Method Subjects and Design We randomly assigned 51 participants (37 women and 14 men) to a 2 (Affect: happy vs. neutral) X 2 (Stereotype: present vs. absent) betweensubjects design. All subjects were undergraduates recruited from introductory psychology classes. They received course credit in return for their participation.

Materials and Procedures Upon arrival at the laboratory, participants were greeted by an experimenter who described the purpose of the study, entitled "CognitiveMotor Coordination." It was explained that the researchers were interested in the extent to which the part of the brain that controls motor behavior is independent of the part of the brain that controls higher thought processes. To study this, subjects would be asked to engage in a cognitive and a motor task simultaneously. The "motor" task was actually the mood induction procedure. Under happy conditions, subjects were given the task of contracting the zygomaticus muscle (producing a smile). The term smile was never used; instead, subjects were individually instructed in terms of specific muscle contractions until the pose successfully emulated a smile. In the control condition, subjects were simply instructed to loosely contract their nondominant hand into a fist. Once they had the proper pose, subjects were instructed to maintain the pose while working on the "cognitive" task, which was the same Students' Court task used in Experiment 1. In this case, only the cheating scenario was used. Subjects completed the same questionnaires as in Experiment 1, including a mood manipulation check and case judgments, focusing primarily on assessments of the guilt of the student accused in the case. In addition, in line with the experimental cover story, they completed some questions assessing how difficult it was for them to engage in both tasks simultaneously. We were particularly interested to know whether the facial pose (happy condition) was any more difficult or distracting than the fist pose (neutral condition). Following completion of a probe for suspicions about the purpose of the study, participants received an educational debriefing and were dismissed.

Results and Discussion Manipulation Checks Analysis of subjects' self-ratings of happiness, collected as part of the participant characteristics questionnaire, revealed that participants in the happy smile condition were indeed significantly happier than those in the neutralfistcondition (Ms = 4.62 vs. 3.76), r(49) = 1.96, p < .025, one-tailed.

625

HAPPINESS AND STEREOTYPIC THINKING

Subjects were also asked to report how difficult it was to coordinate the cognitive and motor tasks. This provides one basis for determining whether the two muscular contraction tasks were differentially distracting or taxing. Comparison of this rating between subjects in the facial versus hand contraction task revealed that both tasks were rated as equally distracting, (F < 1). Thus, the two tasks differed in how happy they made subjects feel, but they did not differ in how distracting they were.1 Perceived Guilt If distracting mental content is the reason that happy people showed greater reliance on stereotypes in Experiment 1, then it was expected that people who are made happy by a means that does not require filling the mind with happy ideas would not show the same stereotyping tendency. Data relevant to this prediction are presented in Figure 2. Contrary to this hypothesis, people made happy by a facial feedback process showed the same proclivity for stereotyping in their perceptions of the accused student's guilt as did people made happy by a reminiscing procedure. The interaction of mood and stereotype activation was significant, F(l, 47) = 4.90, p < .05, with significantly greater perceptions of guilt among the happy subjects who considered a stereotyped target. Thesefindingscast doubt upon the cognitive-distraction interpretation, which holds that happy thoughts are distracting people in a good mood, constraining their capacity for systematic thinking about other topics or persons. People in the smile pose condition, although happier than control subjects, were no more distracted by their posing task, nor were they any more likely to spontaneously generate random, distracting thoughts. Yet they were significantly more likely to render judgments in line with their stereotypic preconceptions. This implies that something else about happiness must be responsible for the tendency to use stereotypes to a greater extent. Revelle and Loftus (1992) have cautioned that so-called mood effects are often actually arousal effects. Perhaps happiness is associated with some kind of physiological arousal that disrupts coordinated thought. According to this arousal-disruption hypothesis, happiness may have some cortical concomitants that constrain the capacity of working memory or other-

HAPPY (SMILE)

wise disrupt the flow of thought, making the use of simple response strategies more appealing. Thayer (1989) has contrasted energetic arousal with tense arousal. He noted that, whereas tense arousal is strongly associated with negative affect, energetic arousal covaries strongly with positive affect. Obviously, the energetic arousal associated with being happy does not render us incapable of maintaining a train of thought, but it may provide some degree of disruption (for a more extensive review of relevant evidence, see Bodenhausen, 1993), and presumably the amount of such disruption would be in proportion to the amount of cortical arousal associated with the happy state. It is possible that in thinking back to an event that made them extremely happy, subjects in Experiment 1 may have experienced a measure of arousal that imposed some degree of constraint on their cognitive capacity. This is somewhat different from simply ruminating on happy thoughts, as it may involve a more physiological kind of capacity constraint. There is a body of evidence relating arousal, per se, to stereotyping (e.g., Bodenhausen, 1990; Kim & Baron, 1988), so if happiness produces its own kind of arousal, it may be by this means that it produces a tendency to stereotype others. This arousal disruption hypothesis is still viable in light of the data from Experiment 2, because it has been previously shown that facial feedback mood induction procedures produce the same syndrome of physiological responses as a reminiscence-based procedure (Ekman et al., 1983; Levenson, Ekman, & Friesen, 1990). Experiment 3 directly examined the plausibility of the arousal disruption hypothesis. Experiment 3 The notion that there is some kind of physiological disruption associated with happiness implies that happy states that are low in their level of energetic arousal (e.g., serenity or contentment) should not produce the same degree of heuristic (or stereotypic) thinking as happy states that are higher in this quality (e.g., excitement or exhilaration). If the hypothesis has merit, one would expect to observe a greater degree of stereotyping in social judgments made by those experiencing greater levels of happy arousal. For Experiment 3, we wanted tofinda mood induction procedure that would allow us to produce happiness that varied in its energetic arousal component, or more colloquially, in the level of excitement that accompanied the subjective happy state. Ideally, the overall amount of happiness would not differ between conditions, but the level of excitement associated with the happiness would differ. A musical mood induction seemed well suited for this purpose. As any music lover could testify, there exists happy music that produces a surge of energy and excitement, but there also exists happy music that produces a sense of tranquility and calm. Through extensive pilot testing, we identified musical selections that produced each of these desired effects. From the perspective of the arousal disruption hypothesis, happy states characterized by excitement and agitation

NEUTRAL (HAND) AFFECT CONDITION

I STEREOTYPE • NO STEREOTYPE

Figure 2. Mean perceived guilt as a function of stereotype activation and type of muscle pose (Experiment 2).

1 In a pretest, we also asked subjects engaging in these two poses to list any thoughts that came spontaneously to mind while contracting the assigned muscle groups. Very few thoughts were listed, and the number did not differ between the smile and thefistconditions.

626


should produce greater disruption, or processing constraints, therefore leading to an increased use of stereotypes, when available, as an efficient strategy for generating a response under the conditions of constraint. Calm, serene music, although producing equivalent amounts of happiness per se, should have much less disruptive influence on processing.

ditions as before. Subjects were randomly assigned within sessions to receive either a stereotypic or a nonstereotypic case. They completed the same dependent measures (case judgments, demographic profile) as did subjects in the previously reported experiments. Again, the crucial dependent variable was subjects' ratings of the apparent guilt of the student accused in the case they read. After a probe for suspicions about the purpose of the experiment, subjects received an educational debriefing and were dismissed.

Method Subjects Fifty-three undergraduates participated in return for credit in their introductory psychology course. Of these, 28 were men and 25 were women.

Procedures and Materials Music pretest. Several selections of music were pretested in terms of their influence on subjects' level of happiness and level of felt energetic arousal. Subjects sat in individual cubicles and listened to a single musical selection. Each selection lasted approximately 10 min. Subjects were instructed simply to close their eyes and listen to the music. They were told that they would be asked some questions about their reactions to the music afterward. When the musical selection hadfinished,they were given a questionnaire asking them to list the thoughts they had while listening to the music and to rate their momentary levels of several characteristics, including happy and energetic, on a 7-point scale ranging from 0 (not at all) to 6 (extremely). The questionnaire also included severalfilleritems concerning, for example, subjects' evaluations of various styles of music. Only data relevant to the chosen selections are described. For the happy and excited (HE) music, we chose a section of music from Borodin's Prince Igor. The selection comes from a vigorous dance sequence, and it is clearly very arousing and invigorating. For the happy and calm (HC) music, we chose a section of music from Delius'S/4 Village Romeo and Juliet ("The Walk Through Paradise Garden") that is particularly relaxing and serene. Both selections were instrumental only (i.e., no verbal content). For comparison purposes, we also had some subjects listen to a selection of neutral (N) music ("Solar Winds" by Hykes) that, although clearly not aversive, seemed minimal in its impact on the casual listener. Both the HE and HC musical selections produced higher levels of happiness than did the N music (Ms: HE = 5.67, HC = 5.91, N = 4.50; ps < .05). However, the HE and HC selections did not differ from each other in the amount of happiness they engendered in the listeners, (p > .75). Analysis of the energetic ratings showed that the HC and N selections did not differ from each other (p > .50), but the HE selection produced significantly higher levels of excitement (ps < .05; Ms: HE = 5.33, HC = 3.46, N = 2.80). Finally, analysis of thought listing data revealed no differences among the three music selections in terms of the number of thoughts reported while listening to the music (p > .50). Thus, the HE and HC selections met the criteria we established in that they produced comparable levels of happiness but differed in the level of felt energetic arousal associated with that state. Main study. Subjects were recruited for a study of music perception. When they reported to the laboratory in groups of four, they were told that because of the brevity of the music study, an additional, unrelated study would take place during the session. After being seated at individual cubicles, subjects were told that their task was simply to listen to a selection of music and then answer some questions about it. Approximately one half of the subjects listened to the HE music (i.e., the Borodin selection), and the remainder listened to the HC music (i.e., the Delius selection). After listening to the music, subjects made several bogus ratings of it and then went on to the second task, the Students' Court. The same materials were used again, under the same instructional con-

Results and Discussion Support for the arousal disruption hypothesis would emerge if subjects in the happy, excited condition showed stronger evidence of stereotypic judgments than did happy, calm subjects. The means relevant to this prediction are shown in Figure 3. Contrary to the arousal disruption hypothesis, the data revealed only a main effect of stereotype activation, F( 1, 49) = 4.91, p < .05. The size of the difference between ratings of stereotyped versus nonstereotyped targets was almost identical in both the happy, excited and the happy, calm conditions. Thus, as before, happy people show evidence of stereotypic bias in their judgments, but the amount of this bias appears to be independent of the degree of excitement or arousal inherent in their happy state. Although they were significantly less excited after listening to their musical selection, HC subjects were just as likely to stereotype as their HE counterparts. Given these results, it becomes fairly implausible to assume that there is any kind of disruptive arousal driving the stereotyping effect consistently observed among happy people. It appears that there is something about being in a pleasant state that promotes the tendency to stereotype, regardless of whether this state involves ruminating about happy ideas or feeling excited or energized by the experience. What, then, can explain this tendency? A third hypothesis relies less on arguments about constrained capacity and more on links between emotional states and patterns of cognitive motivation. According to the effort conservation hypothesis, happy people may simply be less motivated to think very systematically about the external environment. Schwarz (1990) has articulated an evolutionary rationale for this view (and provided empirical support for it; see, e.g., Bless, Bohner, Schwarz, & Strack, 1990; Schwarz & Bless, 1991). Whereas negative moods signal the existence of a prob-

HAPPY-ENERGETIC HAPPY-CALM AFFECT CONDITION I STEREOTYPE H NO STEREOTYPE

Figure 3. Mean perceived guilt as a function of stereotype activation and musical arousal level (Experiment 3).

627


lem that needs to be solved, happy moods signal satisfaction with the current state. Happiness is a kind of safety signal, indicating that there is no current need for problem solving. If one makes the plausible assumption that emotional states are associated with cognitive strategies that are most effective for the kinds of situations in which the emotion typically arises, it is logical to propose that unhappy people will think more deeply about their social environment (in an effort to solve their problems; see Edwards & Weary, 1993; Weary, 1990) whereas happy people can contentedly coast on cruise control, not bothering to think very deeply about surrounding events unless they impinge directly on their well-being. According to this view, happy people are not incapable of thinking systematically (because of constraints on cognitive capacity); rather, they simply often choose not to do so, unless their own outcomes are at stake (Forgas, 1989) or they enjoy the cognitive task and thus have some intrinsic motivation. Perhaps happy people, when asked to judge the likelihood of a defendant's guilt, simply prefer not to exert much effort, especially if there is a viable shortcut or heuristic cue available that provides the basis for a less effortful response. Regardless of the source or nature of a pleasant or happy state, it seems to signal a retreat from deeper, more systematic thinking, at least for tasks lower in interest and personal relevance. If a reduction in cognitive motivation is what is really behind the tendency for happy people to stereotype more, then a manipulation that makes task performance seem more consequential to subjects should be able to compensate for the motivational deficits characterizing happy people. This possibility was explored in a fourth and final experiment. Experiment 4 Research in the domain of persuasion has shown that, whatever the factors producing a tendency for happy people to process persuasive messages less systematically, happy people are indeed capable of responding appropriately to variations in argument quality if they are explicitly instructed to attend to the arguments (Schwarz et al., 1991). This clearly indicates that, if there are any cognitive constraints on processing ability imposed by the experience of happiness (as the cognitive distraction and arousal disruption hypotheses would have it), these constraints are easily surmountable. This strongly suggests that cognitive or epistemic motivation is what is really lacking in happy people. In thefinalstudy, we examined the issue of happy people's responsiveness to motivational cues in a stereotyping context. When social perceivers are concerned about the accuracy of their impressions and evaluations of social stimuli, they should be more attuned to the implications of specific available evidence and should rely less on simple cues or generalized preconceptions (Kruglanski, 1989). One way to induce this concern is to hold people accountable for their judgments and evaluations (Tetlock, 1983; Tetlock & Kim, 1987). Telling people that they will be held accountable for the quality of their judgments, and that they must be able to defend the decisions they make, should have the effect of motivating more systematic thought. What happens when happy people are told that they will be held accountable for the quality of their judgments? One possibility is that accountability cues will provide a sufficient motivational

impetus to overcome any resource-conserving tendency on the part of happy people. If so, they may show a pattern of judgments that is quite similar to neutral mood subjects. That is, all subjects may be evidence focused, and if they all have the same evidence, they will all form similar judgments. If, on the other hand, happy people still show a greater stereotyping tendency even in the presence of accountability cues, this would suggest either that (a) their resource-conservation mode of thinking is not easy to modify, or (b) perhaps there is some kind of basic constraint that makes systematic thought more problematic for happy people. The latter possibility seems unlikely, given the results of our previous studies, and both possibilities seem unlikely given the results of Schwarz et al. (1991) in their persuasion research. Thus, we expected that the stereotyping tendency evidenced by happy people in the previous three studies would be eliminated upon the activation of accountability cues.

Method Subjects and Design Participants were 131 undergraduates (93 women and 38 men) who received credit in their introductory psychology course for participating. Participants were randomly assigned to one of eight conditions defined by a 2 (Mood: happy vs. neutral) X 2 (Stereotype Activation: present vs. absent) X 2 (Accountability: low vs. high) between-subjects factorial design.

Materials and Procedures The same materials and procedures used in Experiment 1 were used again. That is, subjects participated in two ostensibly unrelated experiments. Thefirstwas actually a mood induction task based on a reminiscence procedure. The second was the Students' Court task. Under low accountability conditions, the methodology of Experiment 1 was replicated exactly. Subjects in the social perception task were simply asked to read the case they were given and make some judgments about it. At the end of the instructions for the social perception task, they were reminded that their judgments would be completely anonymous. In the high accountability conditions, however, the reminder about anonymity was deleted, and in its place was the following statement, in bold typeface: "Bear in mind that you will be held accountable for your judgment, just as if you were a judge on a real peer discipline panel. That is, you will have to be able to justify the decisions that you make about the case you read." After reading the instructions, subjects turned the page to find a description of a case concerning either a stereotypic or a nonstereotypic offender and the usual questionnaire items. Only one new item was added to evaluate the effectiveness of the accountability manipulation. For this item, subjects were asked to rate how strongly they felt that it was necessary to be able to justify their judgments about the case. This rating was made on a 10-point scale ranging from 0 (not at alt) to 10 (extremely). After completing the questionnaires and a probe for suspicions, subjects were debriefed and dismissed.

Results and Discussion Manipulation Checks Subjects' self-ratings of happiness confirmed that those in the happy condition were significantly happier than those in the neutral mood condition (Ms = 6.76 and 6.06, respectively, p < .025, one-tailed). In addition, subjects in the high accountability condition reported feeling a greater need to be able to justify

628


their judgments than those in the low accountability conditions (Ms = 5.90 vs. 5.20, respectively, p < .05, one-tailed). Thus, both the affect and the accountability manipulations were effective. Social Judgment Task The low accountability conditions constituted an empirical replication of Experiment 1, so we expected to find the same pattern of results (i.e., greater stereotyping in the social judgments of happy people). Under conditions of high accountability, however, we expected happy people to be no different from neutral mood subjects because they would become more databased in their social information processing. The data relevant to these predictions are presented in Table 1. It is clear that the judgments rendered by subjects in the low accountability conditions show the same pattern as that observed in previous experiments: Happy subjects made more stereotypic judgments than did those in the neutral condition (p < .05). However, the judgments of the accountable subjects showed quite a different pattern. The interaction of affect condition, accountability, and stereotype activation was statistically significant, F(l, 122) = 5.55, p < .025. Not only did the accountable happy people not show a greater stereotypic bias in their judgments, they showed a trend toward reduced perceptions of the guilt of the stereotyped target (p < . 15). In any case, it is clear that, as expected, happy subjects were quite capable of avoiding stereotypic judgments when given some motivational impetus to do so. General Discussion In each experiment reported, happy people who were simply asked to make some judgments about a case of alleged student misbehavior were significantly more likely to render harsher judgments about a stereotyped judgment target than were people in a neutral mood. Thisfindingis noteworthy in several respects. Whereas the bulk of prior research on affect and stereotyping has emphasized the role of negative affect of one kind or another in the elicitation of stereotypic responses to outgroup members (see Bodenhausen, Sheppard, & Kramer, 1994; Esses, Haddock, & Zanna, 1993), the presentfindingsshow that positive affect can also elicit these responses. Although these findings in no way cast doubt on the possibility that negative states such as frustration or anxiety can produce increases in prejudice and stereotyping, they do clearly show that negative states are not necessary for this to occur, and indeed, positive moods can (perhaps for very different reasons) produce similar sorts of

Table 1 Perceived Guilt as a Function of Accountability, Mood, and Stereotype Activation (Experiment 4) Not accountable

Accountable

Condition

Happy

Neutral

Happy

Neutral

Stereotype No stereotype

6.94 5.69

5.58 6.44

5.84 6.78

6.94 6.60

heightened stereotyping. Interestingly, Bodenhausen et al. (1994), using the same experimental paradigm as in the current Experiment 1, found that angry subjects produced a pattern of stereotypic judgments comparable to the happy subjects in the present experiments, whereas sad subjects did not. These findings collectively suggest that previous views about the connections between affect and stereotyping have been overly simplistic. The general claim that heightened stereotyping is associated with negative affect fails to recognize that (a) not all negative states seem to follow this pattern, and (b) positive states can also produce the same pattern. The findings reported also failed to match expectations derived from research on the mood congruency effect in social judgment. Although there were some mood congruency trends for judgments of the nonstereotyped target (i.e., happy subjects judged the nonstereotyped target somewhat more leniently than did neutral mood subjects), happy subjects considering a stereotypic case certainly did not show more leniency toward the target of the case. This suggests that the tendency toward mood congruency in social judgment is qualified by other factors, one of which maybe happy people's preferential use of simple judgment heuristics, even if these have negative implications. One implication of our main finding is that people who are in a good mood, for whatever reason, and who happen to hear about the alleged stereotypic behavior of an outgroup member (e.g., in a news report, overheard gossip), are likely to jump to stereotypic conclusions about the matter. They may decide that the allegation has substance, irrespective of the availability or implications of specific facts in the case. Such allegations, sufficiently documented in the imaginations of happy people, may constitute a type of mental confirmation of stereotypic beliefs that is not distinguished in memory from actual, proven incidents (cf. Slusher & Anderson, 1987). Although the present research examined only one judgment domain (perceptions of misbehavior), it is a particularly important one for social perception of many stigmatized outgroups. The most damaging social stereotypes are those that associate outgroup membership with tendencies toward engaging in undesirable, socially unacceptable, or criminal practices. The present research shows that feeling good tends to increase the extent to which such stereotypes are applied in making judgments of individual members of stigmatized outgroups. Why do happy people produce more stereotypic judgments? These experiments provided little support for the idea that happiness promotes stereotypic thinking by constraining the perceiver's capacity for more systematic thought. Given that the same results emerged even when the mood induction procedure did not involve any particular cognitive content (e.g., the facial feedback procedure of Experiment 2 or the music of Experiment 3) and did not involve any differential elicitation of random thoughts, it seems fairly unlikely that happy people are necessarily too caught up in happy ruminations to think deeply about the social world. By the same token, the idea that happy people experience some kind of disruptive arousal or excitement seems unlikely given that happy subjects do not stereotype to any greater extent than do calm, happy subjects (Experiment 3). The effort conservation hypothesis is compatible with the patterns emerging across all four experiments. According to this

629


view, happy people simply are not motivated to engage in cognitive effort, unless the tasks requiring such effort have some direct bearing on their own well-being or enjoyment. Experiment 4 showed that, if made to feel accountable, happy subjects become quite unlikely to render stereotypic judgments. Presumably, by making them feel that their own outcomes could be affected by the nature of their performance on the judgment task (e.g., if they could not justify a negative reaction and thus might be suspected of prejudice, which would be an undesirable outcome for most people), they became motivated to avoid stereotypic bias. Along similar lines, Forgas (1989) showed that the more heuristic or impulsive judgment strategies of happy people are only likely to emerge if their own outcomes were not contingent upon the quality of their judgments. The costs of effort conservation may be considerable for the targets of social stereotyping, but for the stereotyping perceiver, the costs of overgeneralization may often be minimal, or they may be outweighed by benefits such as preservation of cognitive resources (Macrae et al., 1994) or mood maintenance. The fact that accountable subjects did not show the same stereotyping tendency as their nonaccountable counterparts might also be interpreted in other ways. For example, it is possible that being told that one must justify one's judgments is somewhat threatening and unpleasant, and as such it may counteract the positive mood induction, eliminating its effects. Although plausible, it is unclear why this process would produce a tendency to avoid stereotypic thinking among the (formerly) happy subjects. Alternatively, the accountability manipulation may have simply aroused social desirability concerns, leading to an avoidance of stereotypic responses. This avoidance would therefore not necessarily be based on being more thoughtful or systematic in one's information processing as a result of the accountability manipulation. This account is also plausible, but the reasoning would seem to apply equally well to the neutral mood subjects. They, too, should be subject to social desirability concerns and should avoid harsh judgments of the stereotyped target. But they did not. Their judgments were among the harshest observed across the four experiments. Perhaps happy people have a particular desire to protect their good mood, and as such, they respond especially strongly to social desirability cues. This possibility remains to be examined in future research. The claim that happiness is associated with less thoughtfulness or cognitive effort has received a very interesting qualification from recent research by Martin, Ward, Achee, and Wyer (1993). Specifically, these researchers showed that good moods can have very different motivational implications, depending on a person's orientation to a current task. If one asks oneself "Have I done enough?" then a good mood may imply an affirmative answer, leading to a termination of cognitive effort. On the other hand, if one asks oneself "Am I still enjoying this task?" then a good mood may imply an affirmative answer, leading to continued effort. Task parameters and individual differences may determine which kind of orientation one has, but this research clearly implies that it is dangerous to generally assume that happiness implies a lack of cognitive motivation. Perhaps it is primarily for relatively uninvolving tasks that this is true. Nevertheless, the outcomes of uninvolving tasks (e.g., perusing evidence about the criminal culpability of a member of

a stereotyped group) can be consequential in building a social perceiver's database of stereotypic knowledge. The present findings may seem somewhat at odds with research indicating greaterflexibilityand creativity among happy persons (for reviews, see Isen, 1984, 1987). The work of Martin et al. (1993) provides one way of integrating these apparently disparate findings. If creativity-related tasks offer an opportunity for enjoyment or fun, then happy people may indeed be willing to expend some effort on them, because their orientation to task may take the form, "Am I still enjoying this?" Creativity tasks may have particular intrinsic appeal to happy people because they provide a means for enjoyment, and therefore for mood maintenance or protection. On the other hand, a social perception task like the one used here (and its real-life analogs) may provide little promise of enjoyment, and happy people may be correspondingly unwilling to exert much effort on it (unless, of course, other motivational concerns arise, as with the high accountability conditions). Happy people's use of heuristic shortcuts (e.g., Isen, Means, Patrick, & Nowicki, 1982) is probably limited to task environments that offer little promise of improving their current mood. If thinking deeply about the task might actually undermine or dissipate their good mood, then happy people have another motive to rely on a quick response strategy rather than engage in more systematic thought. Yet another interpretation of the present results is also possible. If positive moods are associated with enhanced cognitive accessibility of positive self-related information (and perhaps also negative information about outgroups, if this information adds to one's sense of positive distinctiveness; see Tajfel & Turner, 1986), then a positive mood induction may selectively prime such associations. If so, feeling good may also mean feeling quite self-confident or even overconfident (but see Allwood & Bjorhag, 1991). As a consequence, people who are feeling good may be more likely to disparage members of other groups rather automatically, without undertaking any effort to correct stereotypic judgments because of this confidence. Relatedly, Devine (1989) has suggested that stereotyping occurs automatically, but is often followed by an effortful attempt to correct one's judgments to avoid stereotypic influences (at least among those who are low in prejudice). Perhaps happy people do not bother to engage in this effortful correction process (see Bodenhausen, 1993). Given that the motivation for this corrective activity seems to lie in unpleasant feelings of guilt and compunction (Devine & Monteith, 1993; Devine, Monteith, Zuwerink, & Elliot, 1991), it may be that the happy mood induction undermines social perceivers' effort to avoid prejudice because it counteracts the feelings of compunction that might otherwise arise. This possibility is currently being investigated.

Conclusions What is the typical ecological relationship between emotional states and intergroup perception, judgment, and interaction? Bodenhausen (1993) recently argued that there are two different kinds of contexts in which affect may enter into intergroup perceptions. The first, termed integral affect, involves emotional reactions that are generated by thoughts and associations about the relevant outgroup(s). For many outgroups, the affect that is integral to people's thoughts about the group is decidedly nega-

630


tive (e.g., Dijker, 1987; Gaertner & Dovidio, 1986; Jackson & Sullivan, 1988; Stephan & Stephan, 1985; Wilder, 1993). If integral emotions were the only affective factor affecting social perception, it would be largely unnecessary to be concerned about happy moods producing stereotypic judgments. But there is a second context, the one of focal concern in the present research, which is termed incidental affect. Incidental affect consists of emotions elicited by factors beyond the intergroup context itself (e.g., the weather, professional accomplishments, one's love life). The present research demonstrates that when one is feeling good, for whatever incidental reason, the likelihood of stereotypic judgments increases, at least for judgmental tasks of minimal personal relevance. It may seem somewhat ironic, given the results of the present experiments, that researchers have touted the importance of a positive affective climate for the development of favorable intergroup relations (e.g., Amir, 1976; Brewer & Miller, 1984). If one considers the probable impact of integral positive affect on intergroup perception, however, there may be little contradiction between the implications of our findings and the recommendations of previous researchers concerning the best climate for intergroup contact. Because the accuracy of social perceptions in an actual intergroup interaction setting is relevant to one's own outcomes or well-being, it seems likely that happy people will be willing to expend the cognitive effort necessary to look beyond their general preconceptions about the outgroup in such settings (cf. Erber & Fiske, 1984). Incidental positive affect, then, may be a more likely culprit in the promotion of stereotypic thinking than integral positive affect. Incidental positive affect is more likely to occur in contexts in which thinking superficially about outgroups incurs no appreciable, immediate costs, whereas thinking more systematically seems to hold little promise of appreciable immediate benefits. A number of interesting questions remain for future research. The present findings have several intriguing implications. One is that people who have a chronic or dispositional tendency to experience happy states may correspondingly be more prone to stereotypic thinking. It also remains for future research to identify the various situational factors that are likely to moderate the impact of happiness on stereotypic thinking. Accountability cues seem to provide one basis for circumventing the stereotyping tendencies of happy social perceivers. Other factors impinging on epistemic motivation would undoubtedly have similar effects. After many years of successfully analyzing the cognitive bases of stereotyping and prejudice, it is becoming apparent that reintroducing affective and motivational issues, within the context of an information-processing framework, promises to yield valuable insights into the nature of intergroup processes (cf. Mackie & Hamilton, 1993). References Adelman, P. K., & Zajonc, R. B. (1989). Facial efference and the experience of emotion. Annual Review ofPsychology, 40, 249-280. Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New "Vbrk: Harper. Allwood, C. M., & Bjorhag, C. G. (1991). Mood and realism of confidence in one's own answers to general knowledge questions. Scandinavian Journal ofPsychology, 32, 358-37J.

Amir, Y. (1976). The role of intergroup contact in change of prejudice and ethnic relations. In P. Katz (Ed.), Towards the elimination ofracism (pp. 245-308). New York: Pergamon Press. Bless, H., Bohner, G., Schwarz, N., & Strack, F. (1990). Mood and persuasion: A cognitive response analysis. Personality and Social Psychology Bulletin, 16, 331-345. Bodenhausen, G. V. (1988). Stereotypic biases in social decision making and memory: Testing process models of stereotype use. Journal of Personality and Social Psychology, 55, 726-737. Bodenhausen, G. V. (1990). Stereotypes as judgmental heuristics: Evidence of circadian variations in discrimination. Psychological Science, 7,319-322. Bodenhausen, G. V. (1993). Emotion, arousal, and stereotypic judgment: A heuristic model of affect and stereotyping. In D. Mackie & D. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in intergroup perception (pp. 13-37). San Diego, CA: Academic Press. Bodenhausen, G. V., & Kramer, G. P. (in press). Social stereotypes and memorial accuracy in criminal justice contexts. In M. McLeod & C. N. Macrae (Eds.), Stereotypes in the criminal justice system: A social psychological perspective. London: Falmer Press. Bodenhausen, G. V., & Lichtenstein, M. (1987). Social stereotypes and information processing strategies: The impact of task complexity. Journal ofPersonality and Social Psychology, 52, 871-880. Bodenhausen, G. V., Sheppard, L. A., & Kramer, G. P. (1994). Negative affect and social judgment: The differential impact of anger and sadness. European Journal ofSocial Psychology, 24, 45-62. Bodenhausen, G. V., & Wyer, R. S., Jr. (1985). Effects of stereotypes on decision making and information-processing strategies. Journal of Personality and Social Psychology, 48, 262-282. Bower, G. H. (1991). Mood congruity of social judgment. In J. Forgas (Ed.), Emotion and social judgment (pp. 31-54). Oxford, England: Pergamon Press. Brewer, M. B., & Miller, N. (1984). Beyond the contact hypothesis: Theoretical perspectives on desegregation. In N. Miller & M. Brewer (Eds.), Groups in contact: The psychology ofdesegregation (pp. 281302). San Diego, CA: Academic Press. Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, 752-766. Chaiken, S. (1987). The heuristic model of persuasion. In M. P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social influence: The Ontario symposium (Vol. 5, pp. 3-39). Hillsdale, NJ: Erlbaum. Chaiken, S., Liberman, A., & Eagly, A. H. (1989). Heuristic and systematic information processing within and beyond the persuasion context. In J. Uleman & J. Bargh (Eds.), Unintended thought (pp. 212-252). New York: Guilford Press. Clark, M. S., & Williamson, G. M. (1989). Moods and social judgements. In H. Wagner & A. Manstead (Eds.), Handbook ofsocial psychophysiology (pp. 347-370). Chichester, England: Wiley. Clore, G. L., Schwarz, N., & Conway, M. (in press). Affective causes and consequences of social information processing. In R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook of social cognition (2nd ed.). Hillsdale, NJ: Erlbaum. Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-18. Devine, P. G., & Monteith, M. J. (1993). The role of discrepancy-associated affect in prejudice reduction. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 317-344). San Diego, CA: Academic Press. Devine, P. G., Monteith, M. J., Zuwerink, J. R., & Elliot, A. J. (1991). Prejudice with and without compunction. Journal ofPersonality and Social Psychology, 60, 817-830.


631

Wyer, Jr., & T. K. Srull (Eds.), Handbook ofsocial cognition (2nd ed.). Dijker, A. J. M. (1987). Emotional reactions to ethnic minorities. EuHillsdale, NJ: Erlbaum. ropean Journal ofSocial Psychology, 17, 305-325. Hamilton, D. L., Stroessner, S. J., & Mackie, D. M. (1993). The influDollard, J., Miller, N. E., Doob, L. W., Mowrer, Q H., & Sears, R. R. ence of affect on stereotyping: The case of illusory correlations. In (1939). Frustration and aggression. New Haven, CT: Yale University D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereoPress. typing: Interactive processes in group perception (pp. 39-61). San Duclos, S. E., Laird, J. D., Schneider, E., Sexter, M., Stem, L., & Van Diego, CA: Academic Press. Lighten, O. (1989). Emotion-specific effects of facial expressions and postures on emotional experience. Journal ofPersonality and Social Isen, A. M. (1984). Toward understanding the role of affect in cognition. In R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook ofsocial cognition Psychology, 57, 100-108. (Vol. 3, pp. 179-236). Hillsdale, NJ: Erlbaum. Eagly, A., & Chaiken, S. (1993). The psychology ofattitudes. Ft. Worth, Isen, A. M. (1987). Positive affect, cognitive processes, and social behavTX: Harcourt Brace Jovanovich. ior. In L. Berkowitz (Ed.), Advances in experimental social psychology Edwards, J. A., & Weary, G. (1993). Depression and the impression(Vol. 20, pp. 203-253). San Diego, CA: Academic Press. formation continuum: Piecemeal processing despite the availability of category information. Journal of Personality and Social Psychol- Isen, A. M., Means, B., Patrick, R., & Nowicki, G. (1982). Some factors influencing decision-making strategy and risk taking. In M. S. Clark ogy, 64, 636-645. & S. T. Fiske (Eds.), Affect and cognition: The seventeenth Carnegie Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nersymposium on cognition (pp. 243-261). Hillsdale, NJ: Erlbaum. vous system activity distinguishes among emotions. Science, 221, Jackson, L. A., & Sullivan, L. A. (1988). Cognition and affect in evalu1208-1210. ations of stereotyped group members. Journal of Social Psychology, Erber, R., & Fiske, S. T. (1984). Outcome dependency and attention to 129, 659-672. inconsistent information. Journal ofPersonality and Social PsycholKim, H., & Baron, R. S. (1988). Exercise and illusory correlation: Does ogy, 54, 709-726. arousal heighten stereotypic processes? Journal of Experimental SoEsses, V. M., Haddock, G., & Zanna, M. P. (1993). The role of mood in cial Psychology, 24, 366-380. the expression of intergroup stereotypes. In M. P. Zanna & J. M. Kruglanski, A. (1989). Lay epistemics and human knowledge. New Olson (Eds.), The psychology ofprejudice: The Ontario symposium \brk: Plenum. (Vol. 7, pp. 77-101). Hillsdale, NJ: Erlbaum. Levenson, R. W, Ekman, P., & Friesen, W. V. (1990). Voluntary facial Fiske, S. T, & Neuberg, S. L. (1990). A continuum of impression foraction generates emotion-specific autonomic nervous system activity. mation from category-based to individuating processes: Influences of Psychophysiology, 27, 363-384. information and motivation on attention and interpretation. In M. P. Mackie, D. M., & Hamilton, D. L. (Eds.). (1993). Affect, cognition, and Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. stereotyping: Interactive processes in group perception. San Diego, 1-74). San Diego, CA: Academic Press. CA: Academic Press. Forgas, J. P. (1989). Mood effects on decision-making strategies. AustraMackie, D. M., & Worth, L. (1989). Processing deficits and the medialian Journal ofPsychology, 41, 197-214. tion of positive affect in persuasion. Journal ofPersonality and Social Forgas, J. P. (1992a). Affect and social perception: Research evidence Psychology, 57, 27-40. and an integrative theory. In W. Stroebe & M. Hewstone (Eds.), EuMackie, D. M., & Worth, L. (1991). Feeling good, but not thinking ropean review of social psychology (Vol. 3, pp. 183-223). Chichester, straight: The impact of positive mood on persuasion. In J. P. Forgas England: Wiley. (Ed.), Emotion and socialjudgment (pp. 201-220). Oxford, England: Forgas, J. P. (1992b). Affect in social judgments and decisions: A Pergamon Press. multiprocess model. In M. P. Zanna (Ed.), Advances in experimental Macrae, C. N., Milne, A. B., & Bodenhausen, G. V. (1994). Stereotypes social psychology (Vol. 25, pp. 227-275). San Diego, CA: Academic as energy-saving devices: A peek inside the cognitive toolbox. Journal Press. ofPersonality and Social Psychology, 66, yi-Al. Forgas, J. P., & Moylan, S. J. (1987). After the movies: The effects of Martin, L. L., Ward, D. W, Achee, J. W., & Wyer, R. S., Jr. (1993). transient mood states on social judgments. Personality and Social Mood as input: People have to interpret the motivational implications Psychology Bulletin, IS, 478-489. of their moods. Journal of Personality and Social Psychology, 64, Gaertner, S. L., & Dovidio, J. F. (1986). The aversive form of racism. In 317-326. J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and Peak, H., Muney, B., & Clay, M. (1960). Opposites structures, defenses, racism (pp. 61-89). San Diego, CA: Academic Press. and attitudes. Psychological Monographs: General and Applied, Gilbert, D. T., & Hixon, J. G. (1991). The trouble of thinking: Activa74(8), Whole No. 495, 1-25. tion and application of stereotypic beliefs. Journal ofPersonality and Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model Social Psychology, 60, 509-517. of persuasion. In L. Berkowitz (Ed.), Advances in experimental social Greenberg, J., Pyszczynski, T, Solomon, S., Rosenblatt, A., Veeder, M., psychology (Vol. 19, pp. 124-203). San Diego, CA: Academic Press. Kirkland, S., & Lyon, D. (1990). Evidence for terror management Petty, R. E., Gleicher, F, & Baker, S. M. (1991). Multiple roles for affect theory: II. The effects of mortality salience on reactions to those who in persuasion. In J. P. Forgas (Ed.), Emotion and social judgments threaten or bolster the cultural worldview. Journal ofPersonality and (pp. 181-200). Oxford, England: Pergamon Press. Social Psychology, 58, 308-318. Pratto, E, & Bargh, J. A. (1991). Stereotyping based on apparently inGreenberg, J., Simon, L., Pyszczynski, T, Solomon, S., & Chatel, D. dividuating information: Trait and global components of sex stereo(1992). Terror management and tolerance: Does mortality salience types under attention overload. Journal of Experimental Social Psyalways intensify negative reactions to others who threaten one's chology, 27, 26-47. worldview? Journal of Personality and Social Psychology, 63, 212- Revelle, W., & Loftus, D. A. (1992). The implications of arousal effects 220. for the study of affect and memory. In S.-A. Christianson (Ed.), HandHamilton, D. L. (1979). A cognitive-attributional analysis of stereotypbook ofemotion and memory (pp. 113-149). Hillsdale, NJ: Erlbaum. ing. In L. Berkowitz (Ed.), Advances in experimental social psychol- Rothbart, M. (1981). Memory processes and social beliefs. In D. L. ogy (Vol. 12, pp. 53-84). San Diego, CA: Academic Press. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup Hamilton, D. L., & Sherman, J. W. (in press). Stereotypes. In R. S. behavior (pp. 145-181). Hillsdale, NJ: Erlbaum.

632


Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective states. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognition (Vol. 2, pp. 527561). New York: Guilford Press. Schwarz, N., & Bless, H. (1991). Happy and mindless, but sad and smart? The impact of affective states on analytic reasoning. In J. Forgas (Ed.), Emotion and social judgments (pp. 55-71). Oxford, England: Pergamon Press. Schwarz, N., Bless, H., & Bohner, G. (1991). Mood and persuasion: Affective states influence the processing of persuasive communications. In M. P. Zanna (Ed.), Advances in experimental social psychology {\o\. 24, pp. 161-199). San Diego, CA: Academic Press. Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal ojPersonality and Social Psychology, 45, 513-523. Schwarz, N., & Clore, G. L. (1988). How do I feel about it? The informative function of affective states. In K. Fiedler & J. Forgas (Eds.), Affect, cognition, and social behavior (pp. 44-62). Toronto, Ontario, Canada: Hogrefe. Sherif, M., & Sherif, C. W. (1953). Groups in harmony and tension. New \ork: Harper & Row. Sinclair, R. C, & Mark, M. M. (1992). The influence of mood state on judgment and action: Effects on persuasion, categorization, social justice, person perception, and judgmental accuracy. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 165193). Hillsdale, NJ: Erlbaum. Slusher, M. P., & Anderson, C. A. (1987). When reality-monitoring fails: The role of imagination in stereotype maintenance. Journal of Personality and Social Psychology, 52, 653-662. Stangor, C, & Duan, C. (1991). Effects of multiple task demands upon memory about social groups. Journal of Experimental Social Psychology, 27, 357-378. Stephan, W. C, & Stephan, C. W. (1985). Intergroup anxiety. Journal ofSocial Issues, 41(3), 157-175. Strack, F, Martin, L. L., & Stepper, S. (1988). Inhibiting and facilitating conditions of facial expressions: A nonobtrusive test of the facial feedback hypothesis. Journal of Personality and Social Psychology, 54, 768-777. Strack, F, Schwarz, N., & Gschneidinger, E. (1985). Happiness and

reminiscing: The role of time perspective, mood, and mode of thinking. Journal ofPersonality and Social Psychology, 49, 1460-1469. Stroessner, S. J., Hamilton, D. L., & Mackie, D. M. (1992). Affect and stereotyping: The effect of induced mood on distinctiveness-based illusory correlations. Journal ofPersonality and Social Psychology, 62, 564-576. Stroessner, S. J., & Mackie, D. M. (1992). The impact of induced affect on the perception of variability in social groups. Personality and Social Psychology Bulletin, 18, 546-554. Stroessner, S. J., & Mackie, D. M. (1993). Affect and perceived group variability: Implications for stereotyping and prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 63-86). San Diego, CA: Academic Press. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall. Terborg, J. R., & Ilgen, D. R. (1975). A theoretical approach to sex discrimination in traditionally masculine occupations. Organizational Behavior and Human Performance, 13, 352-376. Tetlock, P. E. (1983). Accountability and complexity of thought. Journal ofPersonality and Social Psychology, 45, 74-83. Tetlock, P. E., & Kim, J. I. (1987). Accountability and judgment processes in a personality prediction task. Journal ofPersonality and Social Psychology, 52, 700-709. Thayer, R. E. (1989). The biopsychology of mood and arousal. New York: Oxford Press. Weary, G. (1990). Depression and sensitivity to social information. In B. S. Moore & A. M. Isen (Eds.), Affect and social behavior (pp. 207230). Cambridge, England: Cambridge University Press. Wilder, D. (1993). The role of anxiety in facilitating stereotypic judgments of outgroup behavior. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 87-109). San Diego, CA: Academic Press. Wills, T. A. (1981). Downward comparison principles in social psychology. Psychological Bulletin, 90, 245-271. Worth, L. T., & Mackie, D. M. (1987). Cognitive mediation of positive affect in persuasion. Social Cognition, 5, 76-94.

Received July 1, 1993 Revision received October 19, 1993 Accepted October 25, 1993 •