Sunteți pe pagina 1din 23

JOURNAL OF THE ACADEMY

Bagozzi etOF
al. MARKETING
/ THE ROLE OF
SCIENCE
EMOTIONS SPRING 1999

The Role of Emotions in Marketing


Richard P. Bagozzi
University of Michigan

Mahesh Gopinath
Tulane University

Prashanth U. Nyer
Chapman University

Emotions are mental states of readiness that arise from ap- in marketing behavior; and to provide suggestions for
praisals of events or one’s own thoughts. In this article, the future research.
authors discuss the differentiation of emotions from affect,
moods, and attitudes, and outline an appraisal theory of
emotions. Next, various measurement issues are consid- THEORY AND FUNCTION OF EMOTIONS
ered. This is followed by an analysis of the role of arousal
in emotions. Emotions as markers, mediators, and mod- Little consistency can be found in the use of terminol-
erators of consumer responses are then analyzed. The ogy related to emotions. For purposes of organization and
authors turn next to the influence of emotions on cognitive discussion, we begin with a definition of emotions and
processes, which is followed by a study of the implications then turn to a framework for interpreting emotional
of emotions for volitions, goal-directed behavior, and de- behavior.
cisions to help. Emotions and customer satisfaction are
briefly explored, too. The article closes with a number of Definitions
questions for future research.
The term affect will be conceived herein as an umbrella
for a set of more specific mental processes including emo-
tions, moods, and (possibly) attitudes. Thus, affect might
This article addresses emotional behavior in marketing. be considered a general category for mental feeling
In comparison to information processing and behavioral processes, rather than a particular psychological process,
decision research, we know much less about the role of per se.
emotions in marketing behavior. Much of what we do By emotion, we mean a mental state of readiness that
know is confined to consumer behavior, as opposed to the arises from cognitive appraisals of events or thoughts; has
behavior of salespeople or marketing managers. a phenomenological tone; is accompanied by physiologi-
Nevertheless, emotions are central to the actions of cal processes; is often expressed physically (e.g., in ges-
consumers and managers alike. Our goal in this article will tures, posture, facial features); and may result in specific
be to present a framework for thinking about emotions; to actions to affirm or cope with the emotion, depending on
discuss the measurement of emotions; to review how emo- its nature and meaning for the person having it. For a simi-
tions function as causes, effects, mediators, and moderators lar perspective, see Lazarus (1991) and Oatley (1992).
The line between an emotion and mood is frequently
difficult to draw but often by convention involves conceiv-
ing of a mood as being longer lasting (from a few hours up
Journal of the Academy of Marketing Science.
Volume 27, No. 2, pages 184-206. to days) and lower in intensity than an emotion. Yet, excep-
Copyright © 1999 by Academy of Marketing Science. tions to this construal can be found. Still another distinction
Bagozzi et al. / THE ROLE OF EMOTIONS 185

between emotions and moods is that the former typically is strength of felt subjective experience, plus magnitude of
intentional (i.e., it has an object or referent), whereas physiological response (e.g., autonomic nervous system
moods are generally nonintentional and global or diffused activity) and extent of bodily expression (e.g., facial dis-
(Frijda 1993). Also, moods are not as directly coupled with plays), when these latter reactions accompany an emotion.
action tendencies and explicit actions as are many Probably the most important factor differentiating
emotions. emotions from moods and attitudes is the way emotions
Finally, attitudes, too, are often considered instances of arise. Emotions are said to have a specific referent (e.g., a
affect, with the same measures used on occasion to indicate consumer becomes pleased when a new detergent removes
emotions and attitudes (e.g., pleasant-unpleasant, happy- grass stains from clothing; he or she is angered by poor ser-
sad, or interested-bored semantic differential items). How- vice in a restaurant). Specifically, emotions arise in
ever, some authors take a narrower view of attitudes and response to appraisals one makes for something of rele-
define them as evaluative judgments (measured, e.g., by vance to one’s well-being. By appraisal, we mean an
good-bad reactions) rather than emotional states. Cohen evaluative judgment and interpretation thereof. By some-
and Areni (1991), for instance, reserve the term affect for thing of relevance, we mean an incident or episode that
“valenced feeling states,” with emotions and moods as happens to oneself (e.g., an unplanned event); a behavior
specific examples. Attitudes are evaluative judgments in one performs or a result one produces (e.g., engaging in an
their view. Nevertheless, other researchers do not make a activity or receiving or failing to receive a planned out-
distinction between affect and evaluative judgments. For come); or a change in an object, person, or thought that has
example, Eagly and Chaiken (1993) point out that Fish- personal meaning.
bein and Ajzen (1975) and other social psychologists have It is important to stress that although categories of
“regarded affect as isomorphic with evaluation itself and events or physical circumstances are frequently associated
used the terms interchangeably” (p. 12). Still others pro- with particular emotional responses, it is not the specific
pose that attitudes have two distinct, but generally highly events or physical circumstances that produce the emo-
correlated, components: affective and cognitive (or tions but rather the unique psychological appraisal made
evaluative) dimensions. Some empirical support exists for by the person evaluating and interpreting the events and
this interpretation (Bagozzi and Burnkrant 1979; Batra circumstances. Different people can have different emo-
and Ahtola 1990; Breckler and Wiggins 1989; Crites, Fab- tional reactions (or no emotional reactions at all) to the
rigar, and Petty 1994; Eagly, Mladinic, and Otto 1994). same event or happening. Note, too, that appraisals can be
It should be recognized that the terms affect, emotions, deliberative, purposive, and conscious, but also unreflec-
moods, and attitudes have frequently been used inconsis- tive, automatic, and unconscious, depending on the person
tently in the literature. We will revisit this issue when we and eliciting conditions for emotional arousal. The central
consider both measurement issues and customer satisfac- role of appraisals in the formation of emotions has come to
tion research below. For now, we stress that when reading define what are aptly called appraisal theories in psychol-
the literature, it is important to pay attention to how ogy (e.g., Frijda 1986; Lazarus 1991; Ortony, Clore, and
authors define affective (and related) terminologies and Collins 1988; Roseman 1991; Smith and Ellsworth 1985).
how they measure the variables to which the terminologies Appraisal theorists maintain that the critical determi-
refer. One’s definition of terms permits an interpretation of nant of any emotion is the resultant evaluation and inter-
their meaning, but equally important is how the variables pretation that arise after comparing an actual state with a
to which the terms refer are operationalized. Some authors desired state. Two appraisals are particularly crucial at this
have defined key variables as emotions, moods, or atti- stage of emotion formation: goal relevance and goal con-
tudes but have used operationalizations corresponding to gruence (Lazarus 1991). That is, a necessary condition for
different concepts. Other authors have used operationali- an emotional response to an event or happening is that a
zations for a single variable that cut across two or more person has a personal stake in it and at the same time
instances of affect. To make clear our definition of emo- judges the event or happening to facilitate or thwart this
tions and how it differs from definitions of mood and atti- stake. Again, the appraisal can occur consciously or
tudes, we present the following point of view. unconsciously.
A distinctive feature of appraisal theories is their speci-
Organizing Framework fication of the conditions leading to discrete emotional
responses. Forced to be brief, we focus on Roseman’s
Above we noted that emotions are mental states of (1991) version of appraisal theories, which differs in rela-
readiness. But so, too, are moods and attitudes. How then tively minor ways from other leading theories. Roseman
might we distinguish between these affective states? For hypothesized that particular combinations of five apprais-
one thing, the state of readiness characterized by an emo- als determine which of 16 unique emotions will be experi-
tion tends to be more intense than that characterized by enced in any given situation. Figure 1 summarizes his the-
moods or attitudes. It is more intense in the sense of ory, where the five appraisals are labeled motive
186 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

FIGURE 1 Not every emotion is accounted for by Roseman’s


Roseman’s (1991) Appraisal framework (or by any other framework for that matter).
Theory of Emotions For instance, pride is regarded as a positive emotion in
Roseman’s framework, yet excessive or exaggerated pride
(sometimes termed hubris) can invite retribution. Like-
wise, shame and guilt are thought by Roseman to be pro-
duced by similar appraisals, but other researchers have
found important distinctions between shame, guilt, and
embarrassment (e.g., Lewis 1993). Likewise, disgust has
been studied extensively and found to differ from distress
(e.g., Rozin, Haidt, and McCauley 1993). Nevertheless, in
contrast to other theories of emotion that conceive of it in
bipolar terms (e.g., pleasure-displeasure and high arousal-
low arousal [Russell 1980] or high negative affect-low
negative affect and high positive affect-low positive affect
[Watson and Tellegen 1985]), Roseman’s framework and
other appraisal theories not only allow for many discrete
emotions but specify conditions for their occurrence.
An elaboration of appraisal theories that is especially
relevant for marketing is the treatment of goals, which may
SOURCE: Roseman (1991:193). Reprinted with permission. be defined as “internal representations of desired states,
where states are broadly defined as outcomes, events, or
processes” (Austin and Vancouver 1996:338). Oatley and
Johnson-Laird (1987) proposed what they termed a com-
municative theory of emotions wherein events are evalu-
consistent/motive inconsistent (i.e., positive emotions ver-
ated in relation to a person’s goals. Emotions are thought
sus negative emotions), appetitive/aversive (i.e., presence
to function to coordinate parts of one’s cognitive system so
of a reward vs. absence of a punishment), agency (i.e., out-
as to manage responses to events and in so doing change
come is perceived caused by impersonal circumstances,
from ongoing to new activities or to maintain desired states
some other person, or the self), probability (i.e., an out-
or activities. The self-regulation of goals is believed to be
come is certain or uncertain), and power (i.e., strong ver-
the main function of emotions:
sus weak coping potential).
For example, pride occurs when one evaluates his or her
own performance of an action or achievement of an out- Each goal and plan has a monitoring mechanism that
come in a positive light (e.g., a feeling of having done evaluates events relevant to it. When a substantial
change in probability occurs of achieving an impor-
well). Here the positive emotion is motive consistent,
tant goal or subgoal, the monitoring mechanism
either appetitive (e.g., having attained a positive goal) or broadcasts to the whole cognitive system a signal
aversive (e.g., having avoided a punishment), self- that can set it into readiness to respond to this
produced under weak or low coping potential, and either change. Humans experience these signals and the
certain or uncertain, depending on the circumstances. Sad- states of readiness they induce as emotions. (Oatley
ness happens when one experiences a loss for which one 1992:50)
recognizes that nothing can be done to restore it. The loss,
which is of something or someone valued, is experienced According to Oatley and Johnson-Laird (1987), emotions
negatively and with high certainty under conditions of low are evoked “at a significant juncture of a plan . . . typi-
coping power. It is perceived to be caused by impersonal cally . . . when the evaluation (conscious or unconscious)
circumstances. of the likely success of a plan changes” (p. 35). Positive
One value of appraisal theories is that it is possible to emotions (e.g., happiness, elation, joy) are associated with
account for most emotions. Indeed, subtle combinations of the attainment of a (sub)goal, which usually leads to a de-
appraisals yield discrete emotional responses. Anger and cision to continue with the plan, whereas negative emo-
regret, for example, differ primarily in only one type of tions (e.g., frustration, disappointment, anxiety) result
appraisal and share in the other four, namely, anger occurs from problems with ongoing plans and failures to achieve
when a person sees another person as the source of injury desired goals (see also Stein, Liwag, and Wade 1996).
to oneself or to another person viewed as a victim of injus- Emotions have implications for action and goal attain-
tice, whereas regret results when one’s negative outcome ment. Lazarus (1991) identifies coping responses as
is attributed to actions or inactions of the self. important mechanisms in this regard. When we experience
Bagozzi et al. / THE ROLE OF EMOTIONS 187

a negative emotion (e.g., anger, sadness, fear), we are in helplessness with sadness, assault with anger, withdrawal
disequilibrium and wish to return to our normal state. with shame).
Either one or both of two coping processes are typically Finally, it has been argued that many coping responses
used: problem-focused coping, where we attempt to alle- to emotions are volitional (Bagozzi 1992:186-189). The
viate the sources of distress, or emotion-focused coping, process begins with outcome-desire units and appraisals
where we either change the meaning of the source of dis- of changes or anticipated changes in goal attainment or
tress (e.g., deny that a threat exists, distance oneself from goal progress. Four appraisal classes can be identified.
the source of distress) or avoid thinking about a problem. Outcome-desire conflicts happen when one fails to
By contrast, coping with positive emotions often achieve a goal or when one experiences an unpleasant
involves sharing one’s good fortune, savoring the experi- event. One or more emotional reactions occur to outcome-
ence, working to continue or increase the rewards, and desire conflicts (e.g., dissatisfaction, anger, shame, guilt,
increasing physical activity. Positive emotions are some- sadness, disappointment, disgust, regret), depending on
times accompanied as well by higher levels of physiologi- attributions of the source of goal failure or the unpleasant
cal arousal, expanded attention, increased optimism, event (i.e., self, other person, or external cause). The cop-
enhanced recall, and a shift from self- to other-centered ing response(s) to these emotions, in turn, is selected from
orientations (e.g., becoming friendlier, caring about oth- the following: intent to remove or undo harm, obtain help
ers), when compared, say, to sadness. Indeed, positive or support, decrease outcome, reevaluate goal, or redouble
emotions, particularly happiness, frequently stimulate effort, if appropriate, depending on the specific emotion
helping or altruistic actions. Why? Schaller and Cialdini involved.
(1990) offer two explanations: “First, we may propose that Outcome-desire fulfillment takes place when one
positive mood leads to enhanced helping via the more achieves a goal, experiences a pleasant event, or avoids an
positive outlook and enhanced activity that appear to unpleasant event. One or more emotional reactions come
spring automatically from the experience of happiness,” about when outcome-desire fulfillment happens (e.g., sat-
and second, “we argue that happiness is associated with a isfaction, joy, elation, pleasure, pride, relief, caring, love),
motivation toward disequilibrium—toward the possible again depending on attributions of the source of good for-
attainment of additional personal rewards that transcend tune. The coping responses to these emotions include an
the basic concern over one’s mood” (pp. 284-285). The intention to maintain, to increase, to share, or to enjoy the
personal rewards referred to here concern such self- outcome.
enrichment motives as affiliation, achievement, compe- Outcome-desire conflicts and fulfillment refer to out-
tence, and esteem. comes in the past or present. The following two appraisal
Closely related to coping responses are action tenden- classes go on with regard to planned outcomes. Outcome-
cies. An action tendency is “a readiness to engage in or dis- desire avoidances transpire in anticipation of unpleasant
engage from interaction with some goal object” and outcomes or goals. Fear or its variants (e.g., worry, anxiety,
includes “(i)mpulses of ‘moving towards,’ ‘moving away,’ distress) are the emotional reactions to this appraisal. The
and ‘moving against’ ” (Frijda, Kuipers, and ter Schure coping responses to these emotions entail either an intention
1989:213). Some theorists maintain that emotions are not to avoid undesirable outcomes or to reinterpret the threat.
merely reactions to appraisals of events but also include The final class of appraisals, outcome-desire pursuits,
action tendencies as part of their meaning (Frijda 1986). happen in anticipation of pleasant goals or outcomes.
Others go further and maintain that action tendencies are Hope is the emotional reaction to such appraisals. The
automatic, “prewired” responses connected to emotions coping response(s) to hope includes intentions to realize or
(LeDoux 1996). And in Frijda’s (1986) treatment, emo- facilitate outcome attainment and to sustain one’s commit-
tions are conceived as the entire process from stimulus ment and vigilance.
event to action and arousal: The theory of self-regulation suggests that unique voli-
tional responses underly coping for each particular emo-
Stimulus event→event coding→appraisal (evaluation of tion or class of emotions (Bagozzi 1992). In addition, the
specific intention enacted depends on one’s degree of self-
relevance, context, and urgency/difficulty/seriousness of efficacy in executing the coping responses. Somewhat
analogous (appraisal→emotional reactions→coping)
action
event)→action readiness responses occur for outcome-identity conflicts, fulfill-
arousal ments, avoidances, and pursuits in social situations related
to normative expectations (Bagozzi 1992:191-194).
Much as emotions arise in response to patterns of apprais- Returning to the distinction between emotions and
als, Frijda (1986; Frijda et al. 1989) has shown that pat- moods and attitudes, we might say that in addition to the
terns of action readiness correspond to distinct emotion things mentioned earlier, emotions differ from moods and
categories (e.g., avoidance with fear, helping with caring, attitudes in the manner in which they arise and in their
188 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

representation in memory. Emotions occur in response to Other researchers who interpret emotions in broader
changes in specific plans or goal-relevant events. As Oat- terms, as either the whole process from the coding of
ley (1992) points out, emotions are manifest as “transi- events to action responses (e.g., Frijda 1986) or as com-
tions from one sequence of action and another,” but moods plex patterns of physiological responses (e.g., Cacioppo,
occur “when the cognitive system is maintained in an emo- Berntson, and Klein 1992 maintain that emotional experi-
tion mode for a period” (pp. 64, 91-92). Indeed, moods are ence is a function of somatovisceral activation, afferentia-
often resistant to changes in events surrounding them. One tion, and cognitive operations; LeDoux 1996 emphasizes
reason for this is that moods “depend on the dissociability brain processes, especially the role of the amygdala) stress
of control emotion signals from semantic information the need for measurement processes going beyond self-
about causation” (Oatley 1992:64). In general, moods are reports. Depending on the theorist, overt behaviors or
elicited by “(a) after effects of emotions; (b) organismic physiological reactions may be considered either a part of
conditions such as illness, fatigue, previous exercise, and what it means to have an emotion or antecedents, concomi-
good health, or pharmacological agents; (c) general envi- tants, or possibly even effects of an emotion. More behav-
ronmental conditions and side-effects of activities: heat, iorally or physiologically oriented researchers obviously
noise, environmental variety, stressful conditions” (Frijda employ measures of emotions consistent with these
1986:289). interpretations.
Like emotions, attitudes can arise from changes in Marketers have tended to take an empirical approach to
events, but attitudes also occur in response to mundane the measurement of emotions and to rely on self-reports
objects. In addition, arousal is a necessary part of emotions (i.e., either unipolar or bipolar items on questionnaires). In
but not necessarily attitudes. Moreover, attitudes seem to the typical application, many items cutting across numer-
have the capacity to be stored during long periods of time ous positive and negative emotions are administered to
and retrieved, whereas emotions are not experienced in measure reactions to a stimulus, and such methods as factor
this way (i.e., emotions are ongoing states of readiness; analysis, multidimensional scaling, or cluster analysis are
they are not stored and retrieved, per se, although it is pos- used to identify the underlying emotional dimensions for
sible to recreate the conditions originally producing them the sample at hand. The number of items investigated in
in our memory and react emotionally to the thoughts so this regard has been as large as 180 (Aaker, Stayman, and
generated, at least up to a point; emotions can, however, be Vezina 1988), while a paper-and-pencil technique (basi-
classically conditioned, but it is unclear whether attitudes cally a single item measured continuously while viewing
can). Finally, the connection of emotions to volition and an ad) has even been suggested to register “warmth”
action is stronger and more direct than it is for attitudes. toward a stimulus ad (Aaker, Stayman, and Hagerty 1986;
Emotions directly stimulate volitions and initiate action, see also Russell, Weiss, and Mendelsohn 1989).
but attitudes may require an additional motivation impe- Two influential studies in the measurement of emo-
tus, such as desire (Bagozzi 1992). tional responses toward advertisements are those by Edell
and Burke (1987) and Holbrook and Batra (1987). Edell
and Burke (see also Burke and Edell 1989) developed a 52-
MEASUREMENT OF EMOTIONS item scale for measuring emotions towards ads, while Hol-
brook and Batra worked with a 94-item scale, which was
The measurement of emotions could focus on a full set later reduced to 34 items (Batra and Holbrook 1990).
of signs or evidence, including evaluative appraisals, sub- With so many items measuring emotions, a question
jective feelings, body posture and gestures, facial expres- arises whether a small number of basic dimensions under-
sions, physiological responses, action tendencies, and lie people’s responses. Edell and Burke (1987) analyzed
overt actions. Whatever measurements one uses should, of the items in their scale and found three factors: upbeat feel-
course, be tied to an underlying theory of emotions. ings, negative feelings, and warm feelings. Likewise, Hol-
Some authors (e.g., most appraisal theorists) construe brook and Batra (1987) used factor analysis, but in a
emotions as mental states or processes, and thus it would slightly different way. Their 94 items were first generated
be prudent to directly measure the cognitive activities a priori to measure 29 emotional indices. For example,
comprising the emotional content of these states or joyful, happy, delighted, and pleased were hypothesized to
processes, from this perspective. Self-reports of one’s sub- indicate a joy index, and ashamed, embarrassed, and
jective experiences constitute the most frequently used humiliated were hypothesized to indicate a shame index.
procedures in this regard, although other methods for indi- Then, based on factor analyses of the 29 indices, a three-
cating emotional memory processes might be used as well factor solution for emotions was found: pleasure, arousal,
(e.g., response time, subliminal priming). From the point and domination. Similarly, Batra and Holbrook (1990)
of view of mental conceptualizations of emotions, physio- factor analyzed 12 emotional indices (largely a subset of
logical, motor, or biological indicators would be at best those revealed in Holbrook and Batra, with a few excep-
considered correlates or indirect measures of emotions. tions) and discovered three factors corresponding closely
Bagozzi et al. / THE ROLE OF EMOTIONS 189

to those discovered by Edell and Burke. Other researchers FIGURE 2


(e.g., Oliver 1994; Westbrook 1987), also using factor Watson and Tellegen’s Two-Factor
analysis, have found emotional items to load on two fac- Structure of Affect
tors: positive affect and negative affect.
Richins (1997) recently argued that consumption-
related emotions are more complex than the two- and
three-factor solutions observed in studies of reactions to
ads or customer satisfaction. Moreover, because explora-
tory factor analyses often yield a small number of factors,
she used a multidimensional scaling procedure, in con-
junction with examination of clusters based on location
and semantic similarity of emotional descriptors in two-
dimensional space. Sixteen clusters of emotions were
identified, each measured by 2 to 8 indicators (in Study 4):
anger, discontent, worry, sadness, fear, shame, envy, lone-
liness, romantic love, love, peacefulness, contentment,
optimism, joy, excitement, and surprise.
Although the approaches used by marketers to date
have been largely empirically driven (e.g., Edell and Burke
1987; Holbrook and Batra 1987; Oliver 1994; Richins,
1997; Westbrook 1987), they are consistent in certain senses
with leading perspectives on emotions in psychology. For
example, Holbrook and Batra’s (1987) three-factor
pleasure-arousal-domination findings are somewhat simi- SOURCE: Watson and Tellegen (1985:225). Copyright © 1985 by the
lar to Russell and Mehrabian’s (1977) three-factor American Psychological Association. Reprinted with permission.
pleasure-arousal-dominance model. However, some dif-
ferences can be pointed out. The most important are the
high loadings of sadness and fear on the domination factor
in Holbrook and Batra’s study, in contrast to the more com- degrees from their primary axes (but see Larsen and
mon outcome of sadness loading on or near the negative Diener 1992).
pole of a pleasure-displeasure factor, and fear loading The idea behind the circumplex model is that emotions
about 45 degrees away from displeasure and toward exist in bipolar categories (e.g., happy-sad, nervous-
greater arousal (e.g., Russell, 1997). relaxed) and can be arranged in a continuous order around
Similarly, the three factors found in Edell and Burke’s the perimeter of a two-factor space. The closer emotions
(1987) study (i.e., upbeat feelings, negative feelings, and are to each other on the perimeter, the more similar they
warm feelings), correspond roughly to the high positive are. For example, excited and aroused are more similar
affect-low positive affect, high negative affect-low nega- than are content and aroused (see Figure 2). The origin or
tive affect, and pleasantness-unpleasantness dimensions, center of the circumplex is thought to represent a neutral
respectively, of Watson and Tellegen’s (1985) circumplex point or adaptation level.
model; the positive-affect and negative-affect factors The circumplex model is appealing because it is intui-
observed by Oliver (1994) and Westbrook (1987) also tive, simple, and provides a description of which emotions
align to a great extent with factors on the circumplex are similar and which are dissimilar. However, it has draw-
model (see also Mano and Oliver 1993). backs. The most serious limitation is that it is based on
The circumplex structure of emotions is shown in Fig- empirical associations among experienced emotions and
ure 2. This representation of emotions has also been called has nothing to say about the conditions (e.g., appraisals)
the two-factor model, because, based on the techniques producing emotions. Then, too, the circumplex model can
used to generate it (e.g., factor analysis or multidimen- obscure subtle differences in emotions. Depending on
sional scaling), emotions can be arranged around two eliciting conditions and people’s appraisals, each of the
orthogonal axes. Russell (1997) terms the axes pleasure- emotions grouped together within any particular category
displeasure and arousal-sleepiness, while Watson and Tel- on the circumplex can be distinct from its cocategory
legen (1985) label them high positive affect-low positive members. For example, it is possible to feel fearful without
affect and high negative affect-low negative affect. Rus- feeling hostile (see “high negative affect” in Figure 2).
sell’s interpretation is essentially the same as Watson and Then, too, the circumplex contains categories that may not
Tellegen’s pleasantness-unpleasantness and strong correspond to emotions. For instance, surprised, drowsy,
engagement-disengagement axes, which are rotated 45 and sleepy do not seem to reflect emotions. On the other
190 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

hand, the circumplex fails to represent well instances of By contrast, research not based on manipulations of
emotion important in everyday life and marketing. It does not appraisal conditions or based on reactions to a single
accommodate love, disgust, pride, hope, guilt, shame, or stimulus frequently finds that emotions cluster in two and,
embarrassment very well, to name a few. The various two- on occasion, three factors (e.g., Edell and Burke 1987;
and three-factor summaries of emotions disclosed in mar- Holbrook and Batra 1987; Oliver 1994; Westbrook 1987).
keting also include variables not reflective of emotions and Furthermore, research examining the construct validity of
exclude instances generally recognized as emotions. measures of discrete emotions obtained in nonexperimen-
Richins’s (1997) Consumption Emotions Set (CES) tal survey settings shows that discriminant validity is often
with its 16 descriptors is appealing because it covers most lacking among measures of different positive or different
emotional reactions one encounters in consumption, and negative emotions (e.g., Bagozzi 1993) or between mea-
its measures achieved satisfactory reliability (except for sures within a particular subcategory of positive or negative
measure of envy, loneliness, peacefulness, and content- emotions, such as among measures of elation, gladness,
ment). The CES would be best used within the context of a and joy (Bagozzi 1991a).
particular theory of emotions to operationalize specific What accounts for the differences in findings between
categories of emotions hypothesized to serve as antece- experimental research based on appraisal theories and sur-
dents, consequences, or moderating variables. The CES vey research or research based on reactions to a single
could also be used to operationalize emotions in more stimulus measured by inventories of emotional items? One
empirically oriented studies, but whether measures of each possibility may be that discrete emotional reactions are
dimension would achieve discriminant validity is prob- short-lived or, once activated, stimulate other emotional
lematic. Most studies incorporating multiple instances of reactions closely related to them. Consider, for example,
both positive emotions and negative emotions find that the sadness: “[W]hen we experience loss, we rarely feel a sin-
measures load on two factors corresponding to positive gle emotion such as sadness. We grieve, are angry, anx-
and negative emotions (e.g., Bagozzi, Baumgartner, and ious, guilty, envious, even hopeful, and defensive” (Lazarus
Pieters, 1998; Oliver, 1994). 1991:250). A reason why these emotions may go in tan-
This raises the question of when one can expect discrete dem is that coping processes for sadness may involve
emotional reactions versus amalgamated groupings of, active struggle or even protest against loss, which results in
say, positive emotions and negative emotions (e.g., highly other emotions. Alternatively, the absence of discrete
correlated feelings of anger, sadness, and fear). The advan- emotions may simply reflect how difficult it is to create
tage of a theory-based approach to emotions is that spe- them. As Izard (1972) noted, pure emotions are “virtually
cific conditions can be specified for the occurrence of dis- impossible to obtain in the laboratory or in any research
tinct emotions, and these hypotheses can be tested. It is for setting” (p. 103). However, as noted above, researchers
these reasons that we used appraisal theories as our organ- have recently found that the use of scenarios in an experi-
izing framework for looking at emotions. Discrete emo- mental context can generate discrete emotional responses
tional reactions are likely to happen when one manipulates (e.g., Gopinath and Bagozzi 1999; Roseman 1991).
conditions producing specific appraisals or when naturally Another factor that might account for a coalescence of
occurring events correspond to unique appraisal condi- multiple positive emotions and multiple negative emotions
tions. Gopinath and Bagozzi (1999), for example, were in two corresponding groupings is the nature of the stimu-
able to induce independent emotional reactions toward lus under study. Most stimulus ads, products, or brands are
three targets in a moviegoing context. On the basis of complex, and the appraisals engendered are typically
Roseman’s (1991) theory, distinct emotional reactions variegated, but related. Also the way in which items are
were produced as a function of three-way interactions presented on some questionnaires makes it difficult to
between motive consistency-inconsistency, appetitive- uncover discrete emotional components. When multiple
aversive, and self-other agency conditions. For example, measures of a single discrete emotional response (e.g.,
admiration, affection, dislike, and contempt resulted happy, pleased, and joyful for “joy”) are interspersed
toward a group member in decision making with regard to throughout a questionnaire, this tends to reduce correla-
movie choice, and pride, shame/guilt, and regret resulted tions among items purported to indicate the same response
toward the self in group decision making, depending on and to increase correlations of these items with measures
the three appraisal conditions. Likewise, happiness, satis- of other responses. The result is predictably a reduction in
faction, annoyance, or frustration occurred toward the discriminant validity and high correlations among items
movie, and pleasure, contentment, irritation, or anger measuring positive emotions and among items measuring
occurred toward the theater, depending on the three negative emotions. The alternative is to group items by the
appraisal conditions. Scenarios were used to create the emotional response they are intended to tap, which tends
appraisal conditions. to increase correlations among measures of the same thing
Bagozzi et al. / THE ROLE OF EMOTIONS 191

and decrease correlations among measures of different It is perhaps too early to give definitive recommenda-
emotional responses. Thus, a trade-off is entailed by use of tions on which emotional scales to employ in empirical
either practice. work, but for now, we think that it is advisable to recom-
An issue that has received little attention in marketing is mend use of unipolar scales that ask respondents to
whether to use unipolar or bipolar items to measure emo- express to what extent each emotion describes their own
tions. The choice can influence findings and their interpre- subjective feelings, rather than bipolar scales that can
tation in fundamental ways. Some leading scholars claim obscure differences in emotional responses across the
that emotions are in the final analysis bipolar states or various dimensions. Also, at least five, preferably seven to
processes. We are either happy or sad, for example, and nine, scale steps should be used for each item to enhance
any other pattern (e.g., independence or concomitance) is the chances that optimal distributional properties of mea-
thought to be an artifact of measurement error. Although a sures will be achieved. In addition, at least three, prefera-
number of studies have shown that pleasant and unpleasant bly more, items should be used for each emotional
emotions are independent (e.g., Bradburn 1969; Diener subcategory.
and Emmons 1985; Zevon and Tellegen 1982), Green, A final measurement issue we wish to raise is the fol-
Goldman, and Salovey (1993) and Barrett and Russell lowing. To what extent are emotions blends of categories?
(1998) argue and present findings demonstrating that emo- For ease of discussion, we consider the categories of emo-
tions are bipolar, once random or both random and system- tions presented on the circumplex. Pleasantness-
atic errors are taken into account. unpleasantness might combine, for instance, with arousal
Bagozzi, Wong, and Yi (1998) challenge the conclu- to produce different kinds or intensities of emotion. To take
sions made by Green et al. (1993) and Barrett and Russell a particular example, consider happiness, an instant of
(1998). They hypothesize that bipolarity, independence, pleasantness on the circumplex. Intense forms of happiness
and concomitance depend on gender, culture, and the tar- occur when pleasantness combines with high arousal:
get of one’s emotions. Briefly, Bagozzi, Wong, and Yi elated, excited, enthusiastic, euphoric, gleeful, joyous, ec-
found that positive and negative emotions were highly static, and exultant are examples. Mild forms of happiness
negatively correlated for American women but highly occur when pleasantness combines with low arousal:
positively correlated for Chinese women. The former pat- peaceful, calm, serene, or quietude of mind are examples.
tern is evidence for bipolarity (i.e., either positive or nega- Happiness, itself, might be at an intermediate level of
tive emotions occur but not both), the latter for concomi- arousal. Other perspectives on blended emotions, based
tance (i.e., both positive and negative emotions occur at the not on a blend of arousal with emotion categories but
same time). For men, the correlations between positive and rather on combinations of “basic” emotions, can be found
negative emotions were much smaller in magnitude but in in Izard (1991, 1992) and Plutchik (1980).
the same direction across ethnicity, that is, slight negative What more general role, if any, does arousal play in
correlations were found for American men, slight positive emotions (Bagozzi 1991b)? We turn to this issue next.
correlations for Chinese men. Thus, the relationship
between positive and negative emotions for men was
nearly independent. Bagozzi, Wong, and Yi argued that AROUSAL
differences in culture (i.e., a tendency for Americans to
view things in dichotomies or discrete categories, i.e., in An early, influential point of view on emotions was pro-
opposition; and a tendency for Chinese to view things dia- fessed by James ([1890] 1950) who claimed that “bodily
lectically, i.e., in balance or harmony) interact with gender changes follow directly the perception of the exciting fact,
differences (i.e., a tendency for women to be more knowl- and . . . our feeling of the same changes as they occur IS the
edgeable and skilled in the use of emotions than men) to emotion” (p. 449, emphasis in original). For James, differ-
produce the divergent patterns. The above findings ent stimuli lead to different bodily responses (e.g., sweaty
resulted when people were asked to express how they felt palms, racing heart, etc.), these physiological responses
at the moment according to their idiosyncratic reasons, are then detected as bodily sensations in our mind, and the
which is the standard procedure used in the literature. That result is interpreted by us as emotional experiences. But it
is, the stimulus for each person could be considered het- is important to note that James reserved this interpretation
erogeneous across individuals. However, when Americans for what he termed the “coarser” emotions (e.g., “grief, fear,
and Chinese were asked to give their emotional reactions rage, love”), which involve strong bodily perturbations; he
to eating in fast-food restaurants (a common, singular was less clear about what he termed the “subtler” emotions
stimulus), positive and negative emotions were indepen- (e.g., “moral, intellectual, and aesthetic feelings” (James
dent for men and women alike. [1890] 1950:468).
An equally influential theory of emotions was pro-
posed by Schachter and Singer (1962), who argued that
192 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

emotion is essentially bodily arousal plus a cognitive label


one provides to diagnose his or her felt arousal, a perspec- meaning. Mere exposure, thus, loses its utility in such
tive consistent with James’s point of view. The idea is that cases, bothloosely
Based as a theory andnotion
on the a practical tactic.
that emotions or moods
we first experience physiological arousal, and, especially induced by one stimulus become attached to another, some
when we are unaware or uncertain of the origin of the researchers have investigated the effects of (a) music on
arousal, we look for evidence in the physical and social length of stay and money spent in supermarkets and res-
situation accompanying the arousal to label our emotional taurants (e.g., Milliman 1982, 1986) and (b) affective tone
state. Although this theory had considerable impact in psy- of stores on purchase intentions (e.g., Donovan and Rossi-
chology for nearly three decades, it has largely been dis- ter 1982) and evaluations (e.g., Gardner and Simokos
credited and has not received much supporting evidence 1986). How can the observed attachment of affect from
beyond Schachter and Singer’s original experiment (e.g., one stimulus to another be explained? Shimp (1991)
Manstead and Wagner 1981; Reisenzein 1983). One reviews seven studies in consumer research that test vari-
exception to the above observation is the frequently ous facets of classical conditioning explanations. The idea
repeated finding that arousal misattributed to an extrane- behind classical conditioning is that the repeated pairing
ous source intensifies emotions (e.g., Cantor, Bryant, and of a conditioned stimulus (e.g., a new brand name) with an
Zillman 1974; Dutton and Aron 1974; Zillman 1971). unconditioned stimulus (e.g., an attractive spokesperson)
An important contribution of James ([1890] 1950) and will eventually lead to the new brand name, on its own,
Schachter and Singer (1962) was the recognition that stimulating the unconditioned response (e.g., positive
arousal plays an essential role in emotion. Before we affect) originally induced by the unconditioned stimulus.
address arousal more fully, we should mention research in Very few studies have been performed in marketing that
marketing that addresses the acquisition of affect, without conform to the conditions required to test classical condi-
necessarily involving concepts of arousal or information tioning. It is unclear whether classical conditioning studies
processing. Both in the practice of marketing and market- can be designed to rule out such rival explanations as
ing research, considerable emphasis has been placed on demand characteristics or cognitive interpretations of the
the effects of various stimuli on consumer behavior. Retail results. Allen and Janiszewski (1989) provide some evi-
store environment cues, advertising, background music, dence that at least one type of cognitive mediation is neces-
brand names, packages, celebrity endorsers, and other sary for classical conditioning to occur: namely, subject
stimuli are frequently administered to produce emotional awareness of the contingency between the conditioned
reactions in consumers. The premise is that emotions or stimulus and unconditioned stimulus. On the other hand,
moods trigger buying responses (e.g., Gardner 1985; Hill classical conditioning, particularly for fear responses, has
and Gardner 1987). been shown to involve unconscious arousal processes con-
If not by appraisal processes or direct arousal, per se, nected with the amygdala (LeDoux 1996).
how does presentation of a stimulus under repetitive con- Another way to explain the observed attachment of
ditions induce affect? A seductively simple explanation affect from one stimulus to another is by Zillman’s (1971)
was provided by Zajonc, who argued that “when objects excitation-transfer model. Briefly, Zillman proposed that
are presented to the individual on repeated occasions, the exposure to one stimulus may produce arousal. If a second
mere exposure is capable of making the individual’s atti- stimulus is presented close on that also is capable of pro-
tude toward these objects more positive” (Zajonc and Mar- ducing arousal on its own, the two sources of arousal may
kus 1982:125). This “mere exposure” effect has been combine to produce intensely experienced arousal. Under
found primarily when the stimulus is simple and previ- certain conditions (e.g., unawareness of the source of
ously unknown or else has little or no semantic content arousal from the first stimulus, recency of the second), a
(e.g., nonsense syllables, foreign words of Chinese char- person may attribute the arousal to the second stimulus.
acters). One mechanism that has been offered to explain Zillman (1983) interpreted arousal as undifferentiated
the mere exposure effect is familiarity: we come to like sympathetic activation.
things that are familiar to us, perhaps because of feelings An issue in need of resolution is whether emotions can
of security (Zajonc 1968). However, a full theoretical occur without arousal. Cognitive theories of emotions
explanation for the mere exposure effect has not been (e.g., appraisal theories) seem to allow that emotions can
developed. When a person is exposed to meaningful stim- be produced by cognition alone, without arousal (e.g., Par-
uli, it has been more difficult to produce the mere exposure rott 1988). But does arousal always accompany the experi-
effect (e.g., Obermiller 1985). This is, in part, a conse- ence of emotions?
quence of the cognitive processing that occurs in response Recent research suggests that arousal is an essential
to awareness of the meaningful stimuli. Repeated expo- component of emotion and is manifest in neural systems in
sure to a meaningful stimulus can lead to increased or the brain. LeDoux (1996) reviews evidence suggesting
decreased positive or negative feelings, depending on its that there are at least five arousal systems in the brain con-
tributing complexly to emotional experience. Four of
Bagozzi et al. / THE ROLE OF EMOTIONS 193

these are in regions of the brain stem and rely respectively attitudes. In another study, Bagozzi (1996) found that for
on acetylcholine, noradrenaline, dopamine, and serotonin attitudes toward giving blood, high arousal tends to
for activation. A fifth (the nucleus basalis) is in the fore- enhance a halo effect from attitudes to positive beliefs and
brain and also relies on acetylcholine to arouse cortical reduce the halo from attitudes to negative beliefs about the
cells. LeDoux notes that the arousal systems act in nonspe- consequences of giving blood.
cific ways throughout the forebrain to make cells more Clearly, arousal is a fundamental aspect of behavior
sensitive to incoming signals. In a sense, the nonspecific related to emotions. We must acknowledge that appraisal
arousal interacts with the information processing of a par- theories have not done a good, or at least complete, job of
ticular stimulus. The amygdala acts as a kind of central incorporating arousal into their frameworks. In their
processor and interacts with the prefrontal cortex (working defense, however, we should mention the following.
memory and attention), hippocampus (long-term explicit Appraisal theorists recognize that the intensity of emo-
memory), and sensory cortex (perception and short-term tional experience consists of two components: arousal and
storage) to influence emotional responses. The amygdala self-control (e.g., Frijda 1994:120). Likewise, researchers
not only influences cortical areas of the brain but also accept that “autonomic nervous system and other physio-
receives input from arousal networks (which themselves logical processes” at least accompany subjectively felt
also influence the forebrain) and feedback from bodily emotions (e.g., Oatley 1992: 21) and that “[i]f the criterion
expression of emotions. In addition, signals from the of physiological activity was eliminated from the defini-
amygdala are sent to muscles and internal organs and glands. tion, the concept of emotion would be left without one of
Most of the research to date into the role of arousal sys- the important response boundaries with which to distin-
tems and the amygdala in emotional behavior has been guish it from nonemotion” (Lazarus 1991:58-59). But is
limited to a small number of emotions (e.g., fear). But it is there more to arousal in emotion than this?
believed that each emotional response is mediated by sepa-
rate neural systems, although each may overlap or resem- Some psychologists and marketers have been quick to
ble each other in many respects. In sum, LeDoux (1996) dismiss Zajonc’s (1980) claim that “preferences need no
and other brain researchers (e.g., Damasio 1994) construe inferences” (see Cohen and Areni 1991:215-216; Lazarus
emotions as biological functions of the nervous system 1982; Marcel 1983; Zajonc 1984). However, we believe it
(see also Zajonc 1998). is important, at the present, to recognize that emotional
To our knowledge, there has been little work to either meanings can be processed subconsciously, emotions can
integrate or reconcile cognitive theories of emotions with be activated automatically, and responses to emotions
neural and biological theories. Much remains to be done in (e.g., coping, action tendencies, actions) also can occur
psychological research before we can make definitive automatically. We leave open the possibility that “emotion
statements about the precise role of arousal in emotional and cognition are best thought of as separate but interact-
experience and behavior. ing mental functions mediated by separate but interacting
At least three studies have examined limited aspects of brain systems” (LeDoux 1996:69; see also Oatley 1992,
arousal in marketing-related contexts. Sanbonmatsu and chap. 1). It appears that arousal is a key part of emotional
Kardes (1988) found that arousal may govern attitude for- functions in the brain that underlies much of its automatic-
mation in persuasive message settings. Attitudes were ity. Cognitive appraisals and arousal need to be better
based on peripheral cues when respondents were highly incorporated into our theories of emotion.
aroused but on argument strength when they were moder-
ately aroused. It is unclear whether arousal functioned
here to reduce information-processing capacity of external EMOTIONS AS MARKERS, MEDIATORS,
arguments or focused attention on internal reactions. AND MODERATORS OF CONSUMER
Bagozzi (1994) found that consistent with predictions RESPONSES
by knowledge-assembly theory (Hayes-Roth 1977),
arousal transformed a two-dimensional, affective- Based on content, most advertisements can be divided
cognitive representation of evaluations of giving blood into two categories: (a) thinking ads, where focus is placed
into a one-dimensional, unitized representation. Likewise, on either factual information (e.g., product attributes) or
arousal increased the association between attitudes and utilitarian consequences of product/service use (e.g., sav-
positive beliefs about the consequences of giving blood ings in time or money) or (b) feeling ads, where concentra-
and decreased the association between attitudes and nega- tion is placed on the emotions one will experience through
tive beliefs. These predictions on the associations between use or ownership of a product (see Puto and Wells’s [1984]
attitudes and beliefs were explained by the implications of similar distinction between informational and transforma-
spreading activation effects of arousal and coping tional advertising). Rather than focusing on the stimulus,
responses, wherein individuals attempt to avoid negative per se, it is more important to emphasize the processes and
and facilitate positive associations of beliefs with experiences comprising a person’s response to ads, to
194 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

better understand the emotional meaning of ads (e.g., Fri- Rose 1990). Presumably, high involvement promotes cog-
estad and Thorson 1986). nitive processing of the usefulness of the ad and its
Paralleling the above differentiation between types of content.
ads and the emotional-cognitive division in mental A majority of research has addressed the effects of Aad,
processes mentioned earlier, Batra and Ray (1986) devel- especially on attitudes toward the brand (Ab). Classic atti-
oped a framework and coding scheme for classifying tude theory maintains that Ab is a function of beliefs about
affective responses to ads, as a complement to cognitive brand attributes or consequences of product use. Mitchell
responses. Specifically, Batra and Ray (1986) identified and Olson (1981) and Shimp (1981) were the first to find
three positive affective response categories: surgency- that Aad provided additional explanatory power for Ab over
elation-vigor/activation (SEVA), deactivation, and social and above brand beliefs (see also Edell and Burke 1987).
affection feelings. The SEVA feelings refer to upbeat, Batra and Ray (1986) found, however, that affective
happy mood reactions (e.g., “the ad’s music was ‘catchy,’ responses toward the ad influenced Ab only indirectly
the ad was ‘fun to watch or breezy,’ or . . . a likable use of through Aad.
humor”); deactivation includes soothing, relaxing, quiet, Some research has addressed the conditions under
or pleasing reactions; and social affection encompasses which Aad influences Ab. For example, Brown and Stay-
feelings of warmth, tenderness, and caring (Batra and Ray man (1992) revealed in their meta-analysis that the effects
1986:241). The three positive affective responses were of Aad on Ab are greater for novel than well-known brands
used along with six cognitive responses (i.e., support argu- and for durable and other goods versus nondurables. Some
ments, counterarguments, execution discounting, execu- evidence also exists showing that Aad influences Ab indi-
tion bolstering, neutral distracters, and other reactions) in rectly through its effect on beliefs about the brand (MacK-
a study of the impact of television commercials on con- enzie et al. 1986). Finally, Stayman and Aaker (1988)
sumers. A total of 12 percent of reactions to ads were clas- showed that repetition governs the feelings to Aad relation-
sified as positive affect: SEVA (3.7%), deactivation ship. Under levels of low versus high repetition, feelings
(2.5%), and social affect (6.1%). have a stronger effect on Aad. This may be a consequence of
One use of emotional reactions in the above sense greater information processing under high versus low
might be as markers or indicators of the effectiveness of repetition.
advertising copy, particularly with respect to the overall In addition to the transfer of affect from ad emotions to
persuasiveness of the ad, the appeal of spokespersons, Aad, research shows that brand names and feelings toward
evaluation of particular product claims, and appraisals of ads can become linked in memory. Stayman and Batra
other aspects of the execution (Wiles and Cornwell 1990). (1991) found that respondents exposed to an affective, as
Also, the program surrounding an ad (e.g., happy versus opposed to an argument, ad were able to retrieve brand atti-
sad content) has been found to have main effects on one’s tudes faster, when primed with the brand name. In addi-
evaluation of an ad and recall (e.g., Goldberg and Gorn tion, the positive retrieved affect had a stronger influence
1987). Future research is needed to identify how program on choice in low- than in high-involvement contexts. In a
content and advertising appeals interact to influence con- second study, Stayman and Batra demonstrated that view-
sumer emotional responses. ers of an ad who were in a positive affective state more
A research question that has received quite a bit of strongly evoked the affect when given the brand name as a
attention in recent years is how and to what extent emo- subsequent retrieval cue than viewers exposed to the ad
tional reactions to ads influence consumer decision mak- while not in a positive affect state.
ing. Most often these reactions have been measured as atti- Olney, Holbrook, and Batra (1991) investigated a hier-
tudes (e.g., liking) toward the ad (e.g., Brown and Stayman archical model explaining advertising viewing time. The
1992; Mitchell and Olson 1981; Shimp 1981). effects of ad emotions (i.e., pleasure and arousal) were
Attitude toward the ad (Aad) is thought to be a function mediated by Aad and reactions to ad content. Interestingly,
of feelings (and thoughts) about the ad itself (e.g., Batra arousal had both indirect and direct effects on viewing
and Ray 1986; MacKenzie, Lutz, and Belch 1986). In fact, time, even after controlling for ad content and Aad. Arousal
Batra and Ray (1986) found that the three affective was measured by self-reports.
responses discussed above significantly predicted Aad even Emotions have been found to serve as moderators in
after controlling for the effects of cognitive responses. A their impact on Ab. Batra and Stayman (1990), in one of the
number of researchers have examined the conditions few studies to examine mood and print ads, found that
under which emotions influence Aad. For instance, under positive moods enhance Ab through their interaction with
low-involvement viewing conditions, feelings about the two cognitive processes: “(1) a bias against the generation
ad have been found to be more important determinants of of negative thoughts (such as evoked by weak arguments),
Aad than thoughts (e.g., how informative or useful the ad leading to a more favorable evaluation of message argu-
is), but under higher involvement viewing, both feelings ments, and (2) a reduction in total cognitive elaboration,
and thoughts may be important (e.g., Miniard, Bhatla, and making processing more heuristic than systematic” (pp.
Bagozzi et al. / THE ROLE OF EMOTIONS 195

212-213). An interesting finding was that positive moods when a “proattitudinal/uplifting” position was taken, but
seem to reduce counterargumentation when weak argu- happy (versus sad) moods lead to less processing of argu-
ments are used in ads (see also Worth and Mackie 1987). ments when a “counterattitudinal/depression” position
Batra and Stephens (1994) also investigated the moderat- was taken. Wegener et al. (1995) explain the findings, in
ing effects of mood on Ab. Specifically, they showed that part, by suggesting that when in a happy mood, people try
mood and motivation (the latter conceived as degree of to maintain their mood and thus process less of the coun-
relevance of the product category for consumers) inter- terattitudinal/depression content.
acted to govern Ab when consumers watched television In addition, interactions of emotions sometimes occur
ads. The greatest impact on Ab occurred when positive with other variables, such as motivational or ability fac-
moods ensued under conditions of low motivation. The tors. For instance, Wegener, Petty, and Klein (1994) found
rationale is that positive moods and low motivation sup- that for people high in need for cognition, messages
press counterargumentation (and therefore lead to more framed positively (i.e., good things will happen if people
favorable A b ) in comparison with high-motivation adopt the advocacy) were more persuasive when the audi-
conditions. ence was happy (versus sad), but messages framed nega-
Following the elaboration likelihood model (ELM) tively (i.e., bad things will happen if people do not adopt
(e.g., Petty and Cacioppo 1986), we might expect emo- the advocacy) were more persuasive when the audience
tions to have one or both of two effects. Emotions might was sad (versus happy). Wegener et al. (1994) explained
operate centrally to influence cognitive processes (e.g., these findings by claiming that good outcomes seem better
with regard to argumentation), or emotions might function and more likely to occur while in a good (versus sad)
peripherally (perhaps through associative or affect transfer mood, and bad outcomes seem worse and more likely to
mechanisms). Early predictions under the ELM took an occur while in a sad (versus happy) mood.
either-or perspective and stressed that when the processing
Finally, even when the likelihood of information pro-
of information in a communication is low (e.g., due to low
cessing is low (due, e.g., to low need for cognition or low
motivation, distraction, low need for cognition, weak
relevance of a product for a consumer), the mood of the
arguments), emotional content in the communication
audience can have a direct effect on Ab, but when the likeli-
(e.g., an attractive spokesperson) is processed directly and
hood of information processing is high, the mood of the
transfers to, or influences, attitude toward the product or
audience affected message-generated thoughts consistent
message. When issue-relevant thinking is high, attitude
change is thought to be a function of the balance of pros with the mood (Petty, Schumann, Richman, and
and cons in the communication, a largely rational process, Strathman 1993). Mood was induced by a television pro-
and emotion may not be a factor. gram or music.
Over time, the role of emotion in persuasive communi-
cation has been found to be more complex than the simple
THE EFFECTS OF EMOTION 1:
central versus peripheral processing alternatives specified INFLUENCES ON COGNITIVE
in the ELM (e.g., Wegener and Petty 1996). For example, PROCESSES
when the likelihood of information processing is moder-
ate, emotions have been found to affect the extent to which A person’s emotional state can influence various
arguments in a communication become elaborated. Posi- aspects of information processing including encoding and
tive (versus neutral) moods tend to lead to less processing retrieval of information, different strategies used to
of arguments (e.g., Bless, Bohner, Schwarz, and Strack process information, evaluations and judgments, and crea-
1990; see also discussion below on the effects of emotions tive thinking. In this section we examine the influence of
on cognitive processes). By contrast, when people process affective states on various aspects of cognitive processes.
the arguments in a message closely, mood might bias
information processing or even function as an argument
Emotion/Mood Effects on Memory
itself (Forgas 1995). Another explanation for mood effects
is that sad or neutral (versus happy) moods lead to more
The influence of mood states on memory can be
effortful processing, which is believed to be done sponta-
broadly classified into three categories: retrieval effects,
neously (e.g., Bohner, Chaiken, and Hunyadi 1994). Note
encoding effects, and state-dependent learning effects. In
that this prediction seems to conflict with the observations
the following section, we review some of the studies that
made by Schaller and Cialdini (1990), who analyzed
have investigated these effects and discuss the mechanism
mostly nonpersuasive communication studies. Some clari-
by which affect influences memory.
fication for the discrepancy in interpretations can be seen
in a study by Wegener, Petty, and Smith (1995). Wegener Retrieval effects. Affect has been shown to influence
et al. (1995) discovered that happy (versus sad) moods retrieval of information, whereby persons in a positive
lead to the processing of more arguments in a message mood state at the time of retrieval have been found to show
196 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

superior recall of positive material learned during encod- a later time. The greater levels of associations evoked by
ing, relative to neutral or negative material (Isen, Shalker, mood-congruent material may have caused a more exten-
Clark, and Karp 1978; Laird, Wagener, Halal, and Szegda sive elaboration, which, in turn, requires more time. How-
1982; Nasby and Yando 1982; Teasdale and Russell 1983). ever, Isen et al. (1978) and Srull (1983) failed to find any
For example, Isen et al. (1978) had respondents study posi- encoding effects of affect.
tive, negative, and neutral words. Either positive, neutral,
or negative mood states were induced in these respon- State-dependent learning effects. A third memory ef-
dents. Respondents in the positive-mood condition re- fect of mood is the state-dependent learning effect of af-
trieved more positive words compared with neutral or fect, where any material regardless of its affective valence
negative words. Isen et al. (1978) suggest that thinking learned under a particular mood state is recalled better
about mood-incongruent material involves shifting one’s when the person is again in that affective state (Bartlett,
focus, which is cognitively taxing, and therefore people Burleson, and Santrock 1982; Bartlett and Santrock 1979;
are more likely to focus on mood-congruent material. An- Bower, Monteiro, and Gilligan 1978; Bower, Gilligan, and
other mechanism proposed to explain the retrieval effects Monteiro 1981). Bower et al. (1978) had respondents learn
of positive affect suggests that positive mood at the time of two sets of words, one while they were in a positive affec-
retrieval functions as a cue that primes the positive mate- tive state and the other in a negative mood. When respon-
rial in memory, making these material more accessible dents who learned the two lists in different moods recalled
(Isen 1989; Isen et al. 1978; also see Tulving and Pearl- the words in the wrong mood (e.g., when words learned in
stone 1966 for a discussion on the effects of priming on ac- a positive mood were recalled while respondents were in a
cessibility). The easier accessibility of positive material negative mood), they experienced interference and the av-
may then influence other cognitive processes such as erage recall rate was less than 50 percent. When respon-
evaluations and decision making, and also subsequent be- dents who learned two lists in different moods recalled the
haviors. While retrieval effects have been replicated by words in the correct mood, the average recall rate was
many researchers using different mood induction and test- more than 70 percent. Control respondents who learned
ing techniques, a few prominent studies have failed to de- and recalled both lists while in the same mood showed an
tect retrieval effects (Bower, Monteiro, and Gilligan 1978; average recall rate between 50 and 60 percent. Bower and
Bower, Gilligan, and Monteiro 1981), leading Isen (1984) Cohen (1982) suggest that the respondents’ mood at the
to speculate that this failure may have been caused by the time of learning becomes associated with the learned ma-
specific material and induction methods (such as hypno- terial and that these associations facilitate the recall of
sis) used in these studies. learned material when the mood state at recall matches the
mood state at encoding (also see Bower 1981 for a descrip-
Encoding effects. Mood states have also been shown to tion of his semantic-network theory).
exhibit encoding effects whereby the affective state at the
Evidence for mood state-dependent learning has been
time of learning is associated with superior memory for
ambiguous. Many studies have failed to find any state-
similarly valenced material (Bower and Cohen 1982; For-
dependent effects of mood state (Bower and Mayer 1985;
gas and Bower 1987). Nasby and Yando (1982) found that
Isen et al. 1978; Laird et al. 1982; Nasby and Yando 1982).
positive mood at the time of learning led to an improved re-
Eich and Birnbaum (1982) and Isen (1984, 1989) have
call of positive material at a later point in time regardless of
suggested that when the material to be learned has seman-
the mood state at the time of recall. Bower et al. (1981)
tic meaning, the stimulus will be encoded according to this
found evidence for the encoding effect of both positive and
meaning, and the influence of the mood state in the encod-
negative affect. Respondents were made to feel happy or
ing and subsequent retrieval processes will be minimal.
sad and then read descriptions of various psychiatric inter-
However, when the stimulus lacks meaning, contextual
views. Happy respondents learned many more happy facts
cues such as affective states at the time of learning may be
than sad facts, while sad respondents learned many more
more strongly encoded with the learned material. At the
sad facts than happy facts.
time of retrieval, these memory items, which have few
How can the encoding effects of mood be explained?
semantic associations, are more primed by the matching
Bower and colleagues (e.g., Bower and Cohen 1982) have
affective state at recall.
suggested that mood-congruent material is likely to be
more semantically elaborated relative to mood- Asymmetric effects of positive and negative moods.
incongruent material. Forgas and Bower (1987) found that While positive affective states have been shown to have
in impression formation situations, sad individuals spent significant influences on recall, negative affect has some-
more time examining negative rather than positive infor- times been found to have either no effect or a much smaller
mation, and they subsequently recalled the negative infor- effect on the recall of negative material from memory.
mation better. Conversely, happy individuals spent more Asymmetric effects of positive and negative moods have
time on the positive information and recalled that better at been found for retrieval effects (Isen et al. 1978; Nasby
Bagozzi et al. / THE ROLE OF EMOTIONS 197

and Yando 1982; Teasdale and Fogarty 1979), encoding solutions (Isen and Daubman 1984; Isen, Daubman, and
effects (Nasby and Yando 1982), and state-dependent Nowicki, 1987; Isen, Johnson, Mertz, and Robinson 1985;
learning effects (Bartlett and Santrock 1979; Bartlett et al. Isen, Niedenthal, and Cantor 1992; Murray, Sujan, Hirt,
1982). Isen (1984) speculated that positive affect is struc- and Sujan 1990). For example, respondents in positive
tured in a broad and extensive manner (i.e., highly inter- mood conditions tended to group a wider range of neutral
connected with other memories), while negative affect is stimuli together (Isen and Daubman 1984). They also
more narrowly and less well connected with other mate- rated words such as cane, ring, and purse as being better
rial, and that specific negative affective states such as an- exemplars of the category “clothing,” than did the neutral-
ger and sadness may be organized separately in memory. mood respondents. Murray et al. (1990) found that
This, in turn, would make it difficult for any given negative positive-mood respondents, compared with respondents
mood to act as an effective retrieval cue. It is not readily ap- in other mood states, formed broader categories when
parent why negative affect would be less well connected in focusing on similarities among exemplars and narrower
memory and positive affect more widely interconnected as categories when focusing on differences, prompting them
Isen suggests, considering that negative affect usually sig- to suggest that what positive mood promotes is not broader
nals problematic environmental conditions that may re- categorization but rather cognitive flexibility. Isen et al.
quire problem solving (Schwarz and Clore 1983; Wegener (1985) found that positive-mood respondents tended to
et al. 1995). One of the side effects of this hypothesized in- give more unusual responses to neutral words in word
terconnectedness of positive memories is the greater crea- association tests. For example, in response to the word
tiveness and cognitive flexibility demonstrated by people house, positive-mood respondents were more likely than
in positive moods, a topic we review briefly later in this were neutral-mood respondents to mention unusual first
article. associates such as security, residence, and apartment, sug-
Mood maintenance and repair have also been put for- gesting that positive-mood states may influence cognitive
ward as explanations for the asymmetric effects of positive organization, resulting in more flexible interpretation of
and negative moods. Isen (1984) has suggested that happy relationships among stimuli. This cognitive flexibility also
individuals attempt to prolong their positive affective state results in enhanced creativity. Isen et al. (1987) found that
by focusing on the positive aspects of their stimulus (mood respondents in positive-mood states outperformed those in
maintenance), while individuals in a negative mood try to neutral- and negative-mood states on tests requiring crea-
improve their situation by not focusing on negative memo- tive solutions. These researchers suggest that positive-
ries (mood repair). Isen (1989) noted that in some of the mood respondents were better at creative problem solving
studies showing symmetrical effects of positive and nega- since such tasks required the ability to see relatedness
tive mood (Bower et al. 1978, 1981), respondents were among seemingly unrelated stimuli, and as we have seen
instructed to maintain their induced moods, and this may earlier, positive affect results in more flexible cognitive
have discouraged them from engaging in mood repair organization.
strategies. However, the mood repair explanation is not
without problems. If sad respondents engage in mood Mood Effects on Evaluation
repair, why is there no evidence for mood-incongruent
One of the best-recognized and most robust effects of
recall effects? After all, an effective strategy to improve a
mood is its influence on evaluation. Individuals in
depressed mood state is to engage in pleasant thoughts and
positive-mood states have been shown to evaluate stimuli
memories. Yet, evidence seems to show that negative
more positively than individuals in neutral- or negative-
mood inhibits the recall of positive memories (e.g., Isen et
mood states, whether the stimuli being studied are other
al. 1978). The competing explanations of memory struc-
people (Clore and Byrne 1974; Forgas and Bower 1987),
ture differences and mood maintenance/repair have also
consumer goods (Isen et al. 1978; Srull 1983), life satis-
been used to explain differences in information-
faction (Schwarz and Clore 1983), or past life events
processing strategies of happy and sad individuals, a topic
(Clark and Teasdale 1982). The reliability of mood effects
we discuss later in this article.
on evaluation is perhaps best illustrated by the fact that
evaluations of the pleasantness of neutral/ambiguous
Mood, Categorization, and Creativity stimuli are used as checks for mood manipulation (e.g.,
Isen et al. 1985, 1987). Isen et al. (1978) found that respon-
dents in whom positive mood was induced were more
Various studies investigating the influence of mood on likely to rate their cars and televisions more favorably
categorization have found that people in positive mood compared with respondents in neutral-mood states. In the
states, compared with those in neutral or negative mood case of memory-based evaluations, if the recalled infor-
states, tend to be better at integrating information, finding mation is biased by the mood (as discussed in the
relationships among stimuli, and at finding creative
198 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

following section), then evaluations that follow will be It has been suggested that the mood maintenance strat-
biased too. On-line processing of evaluations can also be egy used by people in positive moods may also cause them
influenced by mood states through the retrieval of infor- to avoid investing cognitive effort in tasks unless doing so
mation congruent with the mood (Clore, Schwarz, and promises to maintain or enhance their positive mood (e.g.,
Conway 1994). Isen 1987; Wegener et al. 1995). Consequently, people in
A competing explanation based on the feelings-as- positive-mood states may not be motivated to engage in
information model (see Schwarz 1990; Schwarz and Bless systematic processing of information and may use heuris-
1991; Schwarz and Clore 1983) suggests that individuals tic processing instead. Positive affect usually denotes a
may assume that their mood states are affective reactions benign environment that does not require any action. On
to the object being evaluated and thus base their evalua- the other hand, negative affective states act as information
tions on their affective states. For example, a happy indi- signaling that the environment poses a problem and may
vidual when asked to evaluate a painting may ask the ques- motivate people to engage in systematic processing, which
tion, “How do I feel?” and infer that his or her positive is usually better suited to handling threatening situations
mood is a reaction to the painting and therefore come to the (Schwarz 1990; Schwarz and Clore 1983). Various stud-
conclusion that he or she likes the painting. The feelings- ies have found evidence suggesting the use of heuristic
as-information hypothesis suggests that when individuals processing by people in positive moods and systematic
attribute their mood state to something else other than the processing by people in negative moods (e.g., Bless et al.
object being evaluated, the effect of mood on evaluation 1990, 1996; Mackie and Worth 1989). Bless et al. (1990)
should disappear. Schwarz and Clore (1983) found sup- presented happy and sad individuals with either strong or
port for this hypothesis when they showed that although weak counter-attitudinal arguments. Sad individuals were
people called on sunny days reported more life satisfaction influenced only by strong arguments, while happy indi-
than people called on cloudy days, the differences disap- viduals were equally influenced by strong and weak argu-
peared when the interviewer casually mentioned the ments. These effects have been consistently replicated and
weather to the individuals. Presumably, the casual men- have been interpreted as providing evidence for reduced
tion of the weather made people attribute their mood to the systematic processing by individuals in positive affective
weather, and hence the mood lost any diagnostic value in states. The two major mechanisms that have been used to
evaluating life satisfaction. In other words, people may use explain mood effects on information-processing strategies
their moods as the basis for forming evaluations of objects are the same as those used to explain the asymmetric
unless the diagnostic value of the mood is discounted. effects of mood effects on memory, namely, the highly
Clore et al. (1994) compared the two explanations for interconnected nature of positive memories and mood
mood effects on evaluation and suggested that individuals maintenance.
may use feelings as information when the evaluation task In discussing the asymmetric effects of positive and
is affective in nature, when other information is lacking, negative affect on memory, we had briefly discussed Isen’s
when the information is complex, or when there are time (1984) contention that positive concepts are more highly
constraints. interconnected in memory relative to negative ideas. Isen
Although mood effects on evaluation have been repli- (1987) and Mackie and Worth (1989) have argued that
cated often, a few prominent studies have shown that mood since positive memories are highly interconnected, posi-
states will not influence evaluation when the object being tive mood will prime and activate some related and many
evaluated is highly familiar and for which past evaluations unrelated positive memories, thus leading to cognitive
exist in memory (e.g., Salovey and Birnbaum 1989; Srull capacity constraints. Because individuals do not have the
1983, 1984). Srull (1984, Experiment 3) found that the cognitive resources to engage in systematic processing,
evaluations of a car made by novices, but not by experts, they resort to the less demanding heuristic processing.
were influenced by mood state. Novices, by definition, are A second explanation for mood effects on cognitive
unfamiliar with the product category and are more likely to processing is based on the concept of mood maintenance.
engage in on-line evaluations, which are more susceptible Isen (1987) has suggested that individuals in a positive
to mood influences. Experts, on the other hand, have prior affective state are motivated to maintain their mood and
evaluations available in memory and so do not engage in may avoid cognitive activity that could interfere with their
on-line evaluations and are less likely to be influenced by positive mood. Thus, individuals in a happy mood are
mood. Similar findings have been obtained by Salovey and unlikely to engage in systematic processing of information
Birnbaum (1989, Experiment 3) and Schwarz, Strack, (also see Bohner, Crow, Erb, and Schwarz 1992).
Kommer, and Wagner (1987). We discuss this and other Another motivation-based explanation for the reduced
related issues in greater detail later in this article. processing under positive mood has been offered by
Schwarz and colleagues (e.g., Schwarz and Clore 1983)
Mood Effects on Information Processing using their feelings-as-information hypothesis. Negative
Bagozzi et al. / THE ROLE OF EMOTIONS 199

affective states inform people that they may be facing a familiar and when there are no strong cognitive, affective,
problem and this may provoke systematic processing of or situational factors that call for systematic processing.
information that is better suited to problem solving. Posi- Accordingly, individuals who evaluate very familiar
tive mood, on the other hand, informs the individual that objects would be using direct-access processing, and
the environment is benign and thus he or she may not be hence mood effects would not be found in such cases (cf.
motivated to engage in effortful cognitive processing. A Srull 1984). Another low-affect infusion strategy is moti-
fourth hypothesis suggests that the motivation to simplify vated processing, where the information search and the
processing is not what causes the reliance on heuristic evaluative outcome are guided by prior motivational goals
processing; rather, it is the increased use of heuristic pro- such as mood repair (see Forgas 1995 for a detailed discus-
cessing by these happy-mood individuals that results in sion). When the evaluative target is simple, the personal
simplified processing (e.g., Bless et al. 1996). relevance is low, the individual has limited cognitive
Although the increased reliance on heuristics by indi- capacity, or the accuracy requirements are not high, the
viduals in positive moods has been replicated many times, individual may resort to heuristic processing, a high-affect
there have been studies that have shown that positive infusion strategy. Under these circumstances, evaluations
moods could lead to both increased or diminished levels of may be based on the existing mood, as in the feelings-as-
cognitive processing (e.g., Martin, Ward, Achee, and Wyer information hypothesis (cf. Schwarz and Bless 1991).
1993). Wegener et al. (1995) have presented empirical When the judgmental situation requires the individual to
support for their hypothesis that positive moods could lead learn and process novel information, and when he or she
to either lesser or greater levels of information processing, has adequate cognitive capacity and motivation to process
depending on whether happy individuals believe that sys- the information, substantive processing is predicted to
tematic processing will lead to the maintenance or destruc- take place. According to Forgas (1995), this default strat-
tion of their positive mood. People in a positive-mood egy, which is used if other less effortful strategies are
state, who believe that systematic processing of a message inadequate, involves constructive processing and is one
would help maintain their mood, may engage in more that is susceptible to affect infusion (cf. Srull 1984).
detailed processing. However, if they view systematic
processing of that message as a threat to their positive-
mood state, they would avoid elaborate processing. This THE EFFECTS OF EMOTIONS 2:
hedonic contingency mood management is especially INFLUENCE ON VOLITIONS,
likely to be seen in positive-mood individuals who have GOAL-DIRECTED BEHAVIOR,
the most to lose by incorrectly engaging in systematic AND DECISIONS TO HELP
processing. Sad individuals, by contrast, have less to lose,
since processing messages might make them feel better Sometimes emotions spur one onto action; at other
(Wegener and Petty 1994). Wegener et al. (1995) suggest times emotions inhibit or constrain action. But only
that most of the previous research in which positive moods recently have researchers devoted much attention to study-
have been shown to lead to reduced message processing ing how this occurs.
have involved counterattitudinal or otherwise unpleasant Bagozzi, Baumgartner, and Pieters (1998) investigated
cognitions (e.g., Mackie and Worth 1989). Since the sys- the role of anticipatory emotions in goal striving. In a
tematic processing of counterattitudinal or unpleasant panel study of consumers’ efforts to lose or maintain their
messages is unlikely to help maintain a positive mood, body weight, people first reacted to the possibility of
individuals choose not to use systematic processing. achieving or not achieving their goals with well-defined
Forgas (1994, 1995), building on Fiedler’s (1990) dual- positive and negative anticipatory emotions. The anticipa-
force model, has recently presented the affect infusion tory emotions then energized volitions in the form of
model, which suggests that the various explanations for intentions, plans, and the decision to expend energy in the
mood effects on memory and information processing service of goal striving. Volitions, in turn, were shown to
(such as mood maintenance, feelings as information, influence goal-directed behaviors related to exercising
affect priming) are not necessarily competing models but and dieting. Next, the intensity of execution of these
could rather be complementary explanations. He suggests behaviors contributed to degree of goal attainment. Goal-
that affect infusion into judgmental processes takes place outcome emotions subsequently followed.
when the judgments require a high degree of constructive Anticipatory emotions functioned in the following
(on-line) processing. Affect does not influence those judg- manner. Respondents indicated which emotions they
ments that call for the retrieval of preexisting evaluations anticipated they would experience if they were to achieve
or for information processing that is highly specific and their goal (i.e., excitement, delight, happiness, gladness,
not requiring constructive processing. Forgas (1994, satisfaction, pride, and self-assurance) and which emo-
1995) suggests that direct-access processing (a low-affect tions they anticipated they would experience if they were
infusion strategy) is used when the evaluative target is to fail to achieve their goal (i.e., anger, frustration, guilt,
200 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

shame, sadness, disappointment, depression, worry, embarrassment, the greater the impulse to hide, to with-
uncomfortableness, and fear). The key processes are simi- draw, and to avoid contact with customers. These coping
lar to forward-looking counterfactual thinking processes responses, in turn, negatively influenced sales
(e.g., Roese and Olson 1995) or what Gleicher et al. (1995) performance.
term “prefactuals.” The decision maker begins by “imag- The study by Verbeke and Bagozzi (1998) illustrates
ining the possible” (i.e., identifying and evaluating the dysfunctional effects of negative emotions (see also Ver-
consequences occurring if one were to achieve one’s goal beke and Bagozzi 1999). Yet, negative emotions can have
or not). The alternative consequences to imagined goal pro-social consequences in certain instances. Bagozzi and
success and goal failure serve as input to appraisal Moore (1994) investigated the role of negative emotions
processes and felt emotional experiences. People are then and empathy on decisions to give help to abused children.
motivated to choose actions promoting the positive affect In Study 1, the effects of a negative emotional-appeal ad
and avoiding the negative affect associated with goal were compared with those of a rational-appeal ad; in
attainment and goal failure, respectively. Study 2, the effects of three levels of intensity of negative-
Four other studies support the above role for prefactuals emotion appeals were examined. The negative emotional
in decision making. Boninger, Gleicher, Hetts, Armor, and ads featured a young boy in his home running away from
Moore (1994) discovered that prefactuals with regard to his father in terror. Both studies demonstrated that four
the purchase of insurance in a laboratory game influenced negative emotions (anger, sadness, fear, tension) and four
subsequent action to take insurance. Gleicher et al. (1995) aspects of empathy (perspective taking, compassion/pity,
found that prefactuals concerning condom use affected protection motivation, fantasy elaboration) mediated the
positive attitudes; however, intentions to use condoms effects of viewing negative emotional ads on decisions to
were not influenced by prefactuals, although the differ- help. Stronger felt negative emotions in the audience led to
ence in means between experimental and control groups greater feelings of empathy, and this, in turn, enhanced the
was in the predicted direction. Perugini and Bagozzi decision to help victims of child abuse.
(1998) also showed that anticipated emotions were instru-
mental in influencing desires, volitions, and behavior in a A final area where emotions play a key role in promot-
study of exercising and an investigation of studying behav- ing action is in goal setting. We can think of goal setting in
iors (see also Parker, Manstead, and Stradling 1995; Rich- terms of a hierarchy of three classes of goals (Bagozzi and
ard, van der Pligt, and de Vries 1995). Finally, Brown, Dholakia 1998; Pieters, Baumgartner, and Allen 1995). A
Cron, and Slocum (1997) replicated the model proposed focal or basic-level goal answers the question, “What is it
by Bagozzi, Baumgartner, and Pieters (1998) in a study of that I strive for?” For example, a consumer may have a goal
goal pursuit by salespeople. to lose body weight. The means for achieving this goal can
Although emotions often function in broad categories be interpreted as subordinate goals and answer the ques-
of positive and negative affect, specific subcategories may tion, “How can I achieve that for which I strive?” Various
have special relevance in certain contexts. Verbeke and types of exercising behaviors and dieting activities are
Bagozzi (1998) studied the behavior of 458 salespeople in examples of subordinate goals for the focal goal of losing
Europe who sold financial services and found that two weight. Superordinate goals constitute reasons for pursu-
negative emotions were particularly important: shame and ing a focal goal and answer the question, “Why do I want to
embarrassment. These self-conscious emotions encom- achieve that for which I strive?” Bagozzi and Edwards
pass personal standards with regard to acceptable (1998) investigated the hierarchical structure for super-
thoughts, feelings, and actions. Shame occurs when one ordinate goals governing body weight maintenance and
fails to live up to an ideal and another person whose found that happiness was an important goal toward which
approval is important to oneself judges this failure criti- the focal goal and other more concrete superordinate goals
cally. Embarrassment is generated either when the self pointed (other superordinate goals included, for example,
does something inappropriate in public or the other with social acceptance, self-esteem, look good, feel good, and
whom one interacts does something inappropriate, but in fit into clothes). Happiness, or more generally personal
contrast to shame, the threat under embarrassment is less welfare, seems to be a central motive spurring one onto
fundamental and involves relatively trivial and even action. We turn now to the role of happiness and other
humorous behaviors (e.g., passing gas audibly while in a emotions in customer satisfaction.
sales negotiation). With embarrassment, the threat is to the
“presented self,” and one feels foolish and awkward; with
shame, the threat is to the “core self,” and one feels quite EMOTIONS AND CUSTOMER
pained, perhaps even humiliated or devastated. Verbeke SATISFACTION
and Bagozzi (1998) found that salespersons’ propensity to
experience shame and embarrassment led to particular Early research viewed customer satisfaction as a key
protective or coping responses: the greater the shame and outcome of product/service purchase, whereby a
Bagozzi et al. / THE ROLE OF EMOTIONS 201

comparison is made between expectations of performance measures of other positive emotions) or the lack of inclu-
and actual performance, and satisfaction arises when sion of a sufficient number of positive emotions. No study
actual performance is greater than or equal to expected to date has performed either a rigorous experimental or
performance, and dissatisfaction occurs otherwise (e.g., survey (e.g., multitrait-multimethod) examination of con-
Oliver 1980; Yi 1990). Westbrook (1987) explored the struct validity of measures of satisfaction, joy, and related
influence of positive affect and negative affect on satisfac- positive emotions.
tion, along with expectation beliefs and disconfirmation The centrality of satisfaction in marketing studies is
beliefs. Satisfaction with automobiles was a function of perhaps more due to being the first emotion to receive
positive affect, negative affect, expectations of receiving scrutiny in postpurchase behavior research than to consti-
benefits and liabilities, and disconfirmation beliefs. Satis- tuting a unique, fundamental construct in and of itself.
faction with cable television was determined by positive Indeed, it is likely that—depending on the situation, prod-
affect, negative affect, and disconfirmation beliefs. West- uct, and person—other positive and negative emotions are
brook and Oliver (1991) found that three emotional more important outcomes of purchase. Under certain con-
responses were important antecedents of satisfaction of ditions, frustration, anger, disappointment, alienation, dis-
newly purchased automobiles: pleasant surprise, interest, gust, anxiety, alarm, guilt, shame, joy, happiness, hope,
and hostility. pride, jubilation, excitement, relief, amusement, and
Oliver (1993) expanded the determinants of satisfac- pleasure, among many other negative and positive emo-
tion to include positive affect (interest and joy) and nega- tions, might be more valid reactions consumers have to
tive affect (anger, disgust, contempt, shame, guilt, fear, purchases. By the same token, the implications of emo-
sadness), as well as disconfirmation beliefs. In addition, tional reactions in purchase situations on complaint
attribute satisfactions and attribute dissatisfactions were behaviors, word-of-mouth communication, repurchase,
hypothesized as direct determinants of satisfaction, as well and related actions may differ for various positive and
as indirect (through positive and negative affect) determi- negative emotions and be of more relevance than reactions
nants. Satisfaction with automobiles was found to be a to satisfaction or dissatisfaction, per se.
function of attribute satisfaction, attribute dissatisfaction, Research by Nyer (1997a, 1997b) addresses appraisal
positive affect, negative affect, and disconfirmation. Satis- theories and their role in postconsumption responses.
faction with a required university course was also found to Nyer found that such postconsumption responses as repur-
be determined by these antecedents, except attribute chase intentions, word-of-mouth intentions, and other
dissatisfaction. reactions are predicted best by using measures of satisfac-
tion plus measures of other emotions. Other studies of note
The investigation of the impact of emotions on post-
investigating the role of specific emotions in customer sat-
purchase reactions is an important development in market-
isfaction include those done by Dubé, Bélanger, and
ing. However, it is unclear whether satisfaction is phenome-
Trudeau (1996); Folkes, Koletsky, and Graham (1987);
nologically distinct from many other positive emotions.
and Taylor (1994).
Satisfaction is neither a basic emotion nor a central
We are uncertain whether a single, summary emotional
emotional category in leading theories of emotions (e.g.,
response such as satisfaction is feasible or even desirable.
Frijda 1986; Lazarus 1991; Oatley 1992; Roseman 1991;
But if one is to be discovered, it may lie more in emotions
Smith and Ellsworth 1985). Furthermore, Shaver,
more closely connected to human welfare or emotional
Schwartz, Kirson, and O’Connor (1987) found that satis-
well-being (e.g., Diener and Larsen 1993). In this regard,
faction shares much common variance with such positive
for example, a case could be made for happiness as a fun-
emotions as happiness, joy, gladness, elation, delight, and
damental emotion related to the purchase of goods and ser-
enjoyment, among others. Likewise, Nyer (1997b) discov-
vices, in particular, and emotional well-being, in general
ered that measures of joy and satisfaction loaded on one
(Bagozzi forthcoming).
factor.
Although we leave open the possibility that measures of
satisfaction can achieve discriminant validity from mea- SOCIAL BASES OF EMOTIONS
sures of joy, happiness, and other positive emotions, we
believe that this will be very difficult to produce in prac- Although people can experience emotions privately,
tice. Also, no theory exists for specifying the conditions such as in response to physical danger, a case can be made
under which satisfaction exists uniquely from many other that emotions are most often interpersonal or group-based
positive emotions. We suspect that previous studies find- responses. Unfortunately, the vast majority of research
ing discriminant validity for measures of satisfaction can into emotional behavior has had an individualistic slant to
be explained by the way items were presented on the ques- it (e.g., Parkinson 1995). Marketing relationships seem to
tionnaire (e.g., separation of measures of satisfaction from be contexts where more social conceptualizations of
202 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

emotions would be worth pursuing. For example, Ruth, Area No. 6. How do emotions function to influence the
Otnes, and Brunel (1998) review studies where discrete behavior of salespeople and managers? When are they
emotions are central in gift exchanges, and they show how functional or dysfunctional?
appraisals lead to emotions and how emotions relate to Area No. 7. How do consumers and managers control
interpersonal relationships and disposition of gifts. A their emotions to advantage? What role do emotions play
related area in need of study is the management of emo- in self-regulation?
tions by organizations and by the self (e.g., Bagozzi 1992;
Hochschild 1983; Locke 1996). Finally, cultural aspects of Area No. 8. How should emotions be measured in mar-
emotions deserve further inquiry (e.g., Markus and Kita- keting? When are self-reports appropriate and inappropri-
yama 1994). ate? How can physiological measures be incorporated?
What are the consequences of treating emotions as unipo-
lar versus bipolar responses?
CONCLUSIONS Area No. 9. What are the distinctions between and rela-
tionships among emotions, affect, feelings, evaluations,
Emotions are ubiquitous throughout marketing. They moods, and attitudes?
influence information processing, mediate responses to Area No. 10. How are distinct emotions related to each
persuasive appeals, measure the effects of marketing stim- other? Under what conditions, for example, does frustra-
uli, initiate goal setting, enact goal-directed behaviors, and tion lead to dissatisfaction? Shame lead to anger? Or love
serve as ends and measures of consumer welfare. Yet, we lead to happiness?
are only beginning to understand the role of emotions in Area No. 11. An area neglected by marketers, but at the
marketing. heart of the discipline, is the role of emotions in marketing
The following areas constitute opportunities for future exchanges and relationships. How do emotions initiate,
research: maintain, or sever marketing relationships? Can emotions
Area No. 1. Exactly how are appraisals conducted and and marketing relationships be studied more dynamically
how do they lead to emotional reactions in consumers? In as they evolve in real time and in context? What are the
what sense are appraisals conscious and purposive versus implications of treating emotions in marketing as social
automatic? What is the role of the amygdala, hippocam- phenomena as opposed to strictly intrapsychic
pus, and other neural systems in appraisals? What are the phenomena?
essential elements of cognitive appraisals in emotional Area No. 12. Finally, is customer satisfaction a unique
behavior and how can they be influenced by marketing phenomenon or is it a subcategory of positive emotions?
stimuli? Likewise, is dissatisfaction a unique phenomenon, the
Area No. 2. Related to No. 1, but also more generally polar opposite of satisfaction, or a subcategory of negative
throughout the processes involved in emotional responses, emotions? How do satisfaction/dissatisfaction and other
what role does arousal play? What is physiological arousal emotions relate to customer loyalty, complaint behavior,
and how does it relate to appraisals, coping responses, and word-of-mouth behavior? What role do emotions play
action tendencies, and behavior? What do self-reports of in the diffusion of innovations?
arousal really indicate? Is arousal an essential component
of emotions? If so, how can marketers develop theories of ACKNOWLEDGMENTS
emotions, better incorporating arousal, and how should
arousal be measured? The authors would like to thank Professor Julie Ruth
Area No. 3. How do emotions affect information pro- for her comments on this article.
cessing in consumer decision making? In what ways do
emotions influence the encoding, storage, and retrieval of
information? What contribution do emotions make to con- REFERENCES
sumer decision making and choice?
Area No. 4. Are emotional reactions universal? In what Aaker, David A., Douglas M. Stayman, and Michael R. Hagerty. 1986.
“Warmth in Advertising: Measurement, Impact and Sequence Ef-
ways do emotions differ across cultures? What role do fects.” Journal of Consumer Research 12 (March): 365-381.
socialization, ethnicity, and culture play in the representa- , , and Richard Vezina. 1988. “Identifying Feelings Elic-
tion, experience, and effects of emotions? ited by Advertising.” Psychology & Marketing 5 (Spring): 1-16.
Allen, Chris T. and Chris A. Janiszewski. 1989. “Assessing the Role of
Area No. 5. Under what conditions do emotions func- Contingency Awareness in Attitudinal Conditioning With Implica-
tion in discrete categories and what are the implications of tions for Advertising Research.” Journal of Marketing Research 26
discrete emotions for consumer behavior? When do emo- (February): 30-43.
Austin, James T. and Jeffrey B. Vancouver. 1996. “Goal Constructs in
tions function in clusters or as homogeneous groups, such Psychology: Structure, Process, and Content.” Psychological Bulle-
as positive and negative categories? tin 120:338-375.
Bagozzi et al. / THE ROLE OF EMOTIONS 203

Bagozzi, Richard P. 1991a. “Further Thoughts on the Validity of Mea- , Shelly Chaiken, and Piroska Hunyadi. 1994. “The Role of Mood
sures of Elation, Gladness, and Joy.” Journal of Personality and So- and Message Ambiguity in the Interplay of Heuristic and Systematic
cial Psychology 61:98-104. Processing.” European Journal of Social Psychology 24:207-221
. 1991b. “The Role of Psychophysiology in Consumer Research.” Boninger, David S., Faith Gleicher, John Hetts, David Armor, and E.
In Handbook of Consumer Behavior. Eds. Thomas S. Robertson and Moore. 1994. “The Influence of Counterfactual Thinking on Inten-
Harold H. Kassarjian. Englewood Cliffs, NJ: Prentice Hall, 124-161. tions and Behavior.” Unpublished raw data, reported in Gleicher et al.
(1995).
. 1992. “The Self-Regulation of Attitudes, Intentions, and Behav-
Bower, Gordan H. 1981. “Mood and Memory.” American Psychologist
ior.” Social Psychology Quarterly 55:178-204.
36 (2): 129-148.
. 1993. “An Examination of the Psychometric Properties of  and Paul R. Cohen. 1982. “Emotional Influences in Memory and
Measures of Negative Affect in the PANAS-X Scales.” Journal of Thinking: Data and Theory.” In Affect and Cognition. Eds. Margaret
Personality and Social Psychology 65:836-851. Clark and Susan T. Fiske. Hillsdale, NJ: Lawrence Erlbaum, 291-
. 1994. “The Effects of Arousal on the Organization of Positive 331.
and Negative Affect and Cognitions: Application to Attitude The-  and John D. Mayer. 1985. “Failure to Replicate Mood-
ory.” Structural Equation Modeling 1:222-252. Dependent Retrieval.” Bulletin of the Psychonomic Society 23:39-42.
. 1996. “The Role of Arousal in the Creation and Control of the , Kenneth P. Monteiro, and Stephen G. Gilligan. 1978. “Emo-
Halo Effect in Attitude Models.” Psychology & Marketing 13:235- tional Mood as a Context for Learning and Recall.” Journal of Verbal
264. Learning and Verbal Behavior 17:573-585.
. Forthcoming. “Happiness.” In Encyclopedia of Human Emo- , Stephen G. Gilligan, and Kenneth P. Monteiro. 1981. “Selectiv-
tions. Eds. David Levinson, James Ponzetti, and Peter Jorgensen. ity of Learning Caused by Affective States.” Journal of Experimental
New York: Macmillan. Psychology: General 110 (4): 451-473.
 and Robert E. Burnkrant. 1979. “Attitude Organization and the Bradburn, Norman M. 1969. The Structure of Psychological Well-Being.
Attitude-Behavior Relationship.” Journal of Personality and Social Chicago: Aldine.
Psychology 37:913-929. Breckler, Steven J. and Elizabeth C. Wiggins. 1989. “Affect Versus
 and Utpal Dholakia. 1998. “Goal-Setting and Goal-Striving in Evaluation in the Structure of Attitudes.” Journal of Experimental
Consumer Behavior.” Unpublished working paper, University of Social Psychology 25:253-271.
Michigan. Brown, Steven P. and Douglas M. Stayman. 1992. “Antecedents and
 and Elizabeth A. Edwards. 1998. “Goal Setting and Goal Pursuit Consequences of Attitude Toward the Ad: A Meta-Analysis.” Jour-
in the Regulation of Body Weight.” Psychology and Health 13:593- nal of Consumer Research 19 (June): 34-51.
621. , William L. Cron, and John W. Slocum Jr. 1997. “Effects of Goal-
 and David J. Moore. 1994. “Public Service Advertisements: Directed Emotions on Salesperson Volitions, Behavior, and Perfor-
Emotion and Empathy Guide Prosocial Behavior.” Journal of Mar- mance: A longitudinal Study.” Journal of Marketing 61:39-50.
keting 58:56-70. Burke, Marian C. and Julie A. Edell. 1989. “The Impact of Feelings on
Ad-Based Affect and Cognition.” Journal of Marketing Research
, Hans Baumgartner, and Rik Pieters. 1998. “Goal-Directed Emo-
26:69-83.
tions.” Cognition and Emotion 12:1-26.
Cacioppo, John T., Gary G. Berntson, and David J. Klein. 1992. “What Is
, Nancy Wong, and Youjae Yi. 1998. “The Representation of Af- an Emotion? The Role of Somatovisceral Afference, With Special
fect in Independent- and Interdependent-Based Cultures.” Unpub- Emphasis on Somatovisceral ‘Illusions.’ ” Review of Personality and
lished working paper, University of Michigan. Social Psychology 14:63-98.
Barrett, Lisa Feldman and James A. Russell. 1998. “Independence and Cantor, Joanne R., Jennings Bryant, and Dolf Zillman. 1974. “Enhance-
Bipolarity in the Structure of Current Affect.” Journal of Personality ment of Humor Appreciation by Transferred Excitation.” Journal of
and Social Psychology 74:967-984. Personality and Social Psychology 30:812-821.
Bartlett, James C. and John W. Santrock. 1979. “Affect-Dependent Epi- Clark, Margaret D. and John D. Teasdale. 1982. “Diurnal Variation in
sodic Memory in Young Children.” Child Development 50:513-518. Clinical Depression and Accessibility of Memories of Positive and
, Georgia Burleson, and John W. Santrock. 1982. “Emotional Negative Experiences.” Journal of Abnormal Psychology 91:87-95.
Mood and Memory in Young Children.” Journal of Experimental Clore, Gerald L. and Donn Byrne. 1974. “A Reinforcement Affect
Child Psychology 34:59-76. Model of Attraction.” In Foundations of Interpersonal Attraction.
Batra, Rajeev and Olli T. Ahtola. 1990. “Measuring the Hedonic and Ed. T. L. Huston. New York: Academic Press, 143-170.
Utilitarian Sources of Consumer Attitudes.” Marketing Letters , Norbert Schwarz, and Michael Conway. 1994. “Affective
2:159-170. Causes and Consequences of Social Information Processing.” In
 and Morris B. Holbrook. 1990. “Developing a Typology of Af- Handbook of Social Cognition, 2nd ed., Vol. 1. Eds. Robert Wyer,
fective Responses to Advertising.” Psychology & Marketing 7 Thomas Srull, and Alice Isen. Hillsdale, NJ: Lawrence Erlbaum,
(Spring): 11-25. 323-417.
 and Michael L. Ray. 1986. “Affective Responses Mediating Ac- Cohen, Joel B. and Charles S. Areni. 1991. “Affect and Consumer Behav-
ceptance of Advertising.” Journal of Consumer Research 13 (Sep- ior.” In Handbook of Consumer Behavior. Eds. Thomas S. Robertson
tember): 234-249. and Harold H. Kassarjian. Englewood Cliffs, NJ: Prentice Hall, 188-
 and Douglas M. Stayman. 1990. “The Role of Mood in Advertis- 240.
ing Effectiveness.” Journal of Consumer Research 17:203-214. Crites Jr., Stephen L., Leandre R. Fabrigar, and Richard E. Petty. 1994.
 and Debra Stephens. 1994. “Attitudinal Effects of Ad-Evoked “Measuring the Affective and Cognitive Properties of Attitudes:
Moods and Emotions: The Moderating Role of Motivation.” Psychol- Conceptual and Methodological Issues.” Personality and Social Psy-
ogy & Marketing 11:199-215. chology Bulletin 20 (December): 619-634.
Bless, Herbert, Gerd Bohner, Norbert Schwarz, and Fritz Strack. 1990. Damasio, Antonio R. 1994. Descartes’Error: Emotion, Reason, and the
“Mood and Persuasion: A Cognitive Response Analysis.” Personal- Human Brain. New York: Avon.
ity and Social Psychology Bulletin 16 (2): 331-345. Diener, Ed and Robert A. Emmons 1985. “The Independence of Positive
, Gerald L. Clore, Norbert Schwarz, Verena Golisano, Christina and Negative Affect.” Journal of Personality and Social Psychology
Rabe, and Marcus Wolk. 1996. “Mood and the Use of Scripts: Does a 47:1105-1117.
Happy Mood Really Lead to Mindlessness?” Journal of Personality  and Randy J. Larsen. 1993. “The Experience of Emotional Well-
and Social Psychology 71 (4): 665-679. Being.” In Handbook of Emotions. Eds. Michael Lewis and Jeannette
Bohner, Gerd, Kimberly Crow, Hans-Peter Erb, and Norbert Schwarz. M. Haviland. New York: Guilford, 405-415.
1992. “Affect and Persuasion: Mood Effects on the Processing of Donovan, Robert J. and John R. Rossiter. 1982. “Store Atmospherics:
Message Content and Context Cues and on Subsequent Behaviour.” An Environmental Psychology Approach.” Journal of Retailing
European Journal of Social Psychology 22:511-530. 58:34-57.
204 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

Dubé, Laurette, Marie-Claude Bélanger, and Elyse Trudeau. 1996. “The Hill, Ronald P. and Meryl P. Gardner. 1987. “The Buying Process: Effects
Role of Emotions in Health Care Satisfaction.” Journal of Health of and on Consumer Mood States.” Advances in Consumer Research
Care Marketing 16 (Summer): 45-51. 14:408-410.
Dutton, Donald G. and Arthur P. Aron. 1974. “Some Evidence for Hochschild, Arlie R. 1983. The Managed Heart. Berkeley: University of
Heightened Sexual Attraction Under Conditions of High Anxiety.” California Press.
Journal of Personality and Social Psychology 30:510-517. Holbrook, Morris B. and Rajeev Batra. 1987. “Assessing the Role of
Emotions as Mediators of Consumer Responses to Advertising.”
Eagly, Alice H. and Shelly Chaiken. 1993. The Psychology of Attitudes.
Journal of Consumer Research 14:404-420.
New York: Harcourt Brace Jovanovich.
Isen, Alice M. 1984. “Toward Understanding the Role of Affect in Cog-
, Antonio Mladinic, and Stacey Otto. 1994. Cognitive and Affec- nition.” In Handbook of Social Cognition, Vol. 3. Eds. Robert Wyer,
tive Bases of Attitudes Toward Social Groups and Social Policies.” Thomas Srull, and Alice Isen. Hillsdale, NJ: Lawrence Erlbaum, 179-
Journal of Experimental Social Psychology 30:113-137. 236.
Edell, Julie A. and Marian Chapman Burke. 1987. “The Power of Feel- . 1987. “Positive Affect, Cognitive Processes, and Social Behav-
ings in Understanding Advertising Effects.” Journal of Consumer ior.” In Advances in Experimental Social Psychology, Vol. 20. Ed.
Research 14 (December): 421-433. Leonard Berkowitz. New York: Academic Press, 203-253.
. 1989. “Some Ways in Which Affect Influences Cognitive Pro-
Eich, Eric J. and Isabel M. Birnbaum. 1982. “Repetition, Cueing and cesses: Implications for Advertising and Consumer Behavior.” In
State-Dependent Memory.” Memory and Cognition 10:103-114. Cognitive and Affective Responses to Advertising. Eds. Patricia Caf-
Fiedler, Klaus. 1990. “Mood-dependent Selectivity in Social Cognition.” ferata and Alice M. Tybout. Lexington, MA: Lexington Books, 91-
In European Review of Social Psychology, Vol. 1. Eds. Wolfgang Stroebe 117.
and Miles Hewstone. New York: John Wiley, 1-32. ——— and Kimberly A. Daubman. 1984. “The Influence of Affect on
Fishbein, Martin and Icek Ajzen. 1975. Belief, Attitude, Intention, and Categorization.” Journal of Personality and Social Psychology 47
Behavior: An Introduction to Theory and Research. Reading, MA: (6): 1206-1217.
Addison-Wesley. , Thomas E. Shalker, Margaret Clark, and Lynn Karp. 1978. “Af-
Folkes, Valerie, Susan Koletsky, and John L. Graham. 1987. “A Field fect, Accessibility of Material in Memory, and Behavior: A Cognitive
Study of Causal Inferences and Consumer Reaction: The View From Loop?” Journal of Personality and Social Psychology 36 (1): 1-12.
the Airport.” Journal of Consumer Research 13 (March): 534-539. , Mitzi M. Johnson, Elizabeth Mertz, and Gregory F. Robinson.
Forgas, Joseph P. 1994. “The Role of Emotion in Social Judgments: An 1985. “The Influence of Positive Affect on the Unusualness of Word
Introductory Review and an Affect Infusion Model (AIM).” Euro- Associations.” Journal of Personality and Social Psychology 48 (6):
pean Journal of Social Psychology 24:1-24. 1-14.
. 1995. “Mood and Judgment: The Affect Infusion Model , Kimberly A. Daubman, and Gary P. Nowicki. 1987. “Positive
(AIM).” Psychological Bulletin 117 (1): 39-66. Affect Facilitates Creative Problem Solving.” Journal of Personality
 and Gordon H. Bower. 1987. “Mood Effects on Person- and Social Psychology 52 (6): 1122-1131.
Perception Judgments.” Journal of Personality and Social Psychol- , Paula M. Niedenthal, and Nancy Cantor. 1992. “An Influence of
ogy 53 (1): 53-60. Positive Affect on Social Categorization.” Motivation and Emotion
Friestad, Marian and Esther Thorson. 1986. “Emotion-Eliciting Adver- 16 (1): 65-78.
tising: Effect on Long-term Memory and Judgment.” Advances in Izard, Carroll E. 1972. Patterns of Emotions: A New Analysis of Anxiety
Consumer Research 13:111-115. and Depression. New York: Academic Press.
Frijda, Nico H. 1986. The Emotions. Cambridge, UK: Cambridge Uni- . 1991. The Psychology of Emotions. New York: Plenum.
versity Press. . 1992. “Basic Emotions, Relations Among Emotions, and
. 1993. “Moods, Emotion Episodes, and Emotions.” In Handbook Emotion-Cognition Relations.” Psychological Review 99:561-565.
of Emotions. Eds. M. Lewis and J. M. Haviland. New York: Guilford, James, William. [1890] 1950. The Principles of Psychology. Vol 2. New
381-403. York: Holt.
. 1994. “Emotions Are Functional, Most of the Time.” In The Na- Laird, James D., John J. Wagener, Mark Halal, and Martha Szegda. 1982.
ture of Emotion. Eds. P. Ekman and R. J. Davidson. New York: Ox- “Remembering What You Feel: The Effects of Emotion on Memory.”
ford University Press, 112-122. Journal of Personality and Social Psychology 42 (4): 646-657.
, Peter Kuipers, and Elisabeth ter Schure. 1989. “Relations Larsen, Randy J. and Ed Diener. 1992. “Promises and Problems With the
Among Emotion, Appraisal, and Emotional Action Readiness.” Circumplex Model of Emotion.” In Review of Personality and Social
Journal of Personality and Social Psychology 57:212-228. Psychology: Vol. 14. Emotional and Social Behavior. Ed. M. S.
Gardner, Meryl P. 1985. “Mood States and Consumer Behavior: A Criti- Clark. Newbury Park, CA: Sage, 25-59.
cal Review.” Journal of Consumer Research 12 (December): 281-300. Lazarus, Richard S. 1982. “Thoughts on the Relations Between Emotion
 and G. J. Simokos. 1986. “Toward a Methodology for Assessing and Cognition.” American Psychologist 35:1019-1024.
Effects of Instore Atmospherics.” Advances in Consumer Research . 1991. Emotion and Adaptation. New York: Oxford University
13:27-31. Press.
Gleicher, Faith, David S. Boninger, Alan Strathman, David Armor, John LeDoux, Joseph. 1996. The Emotional Brain: The Mysterious Underpin-
Hetts, and Mina Ahn. 1995. “With an Eye Toward the Future: The Im- nings of Emotional Life. New York: Simon & Schuster.
pact of Counterfactual Thinking on Affect, Attitudes, and Behavior.” Lewis, Michael 1993. “Self-Conscious Emotions: Embarrassment,
In What Might Have Been: The Social Psychology of Counterfactual Pride, Shame, and Guilt.” In Handbook of Emotions. Eds. M. Lewis
Thinking. Eds. Neal J. Roese and James M. Olson. Mahwah, NJ: and J. M. Haviland. New York: Guilford, 563-573.
Lawrence Erlbaum, 283-304. Locke, Karen. 1996. “A Funny Thing Happened! The Management of
Goldberg, Marvin E. and Gerald J. Gorn. 1987. “Happy and Sad TV Pro- Consumer Emotions in Service Encounters.” Organizational Science
grams: How They Affect Reactions to Commercials.” Journal of 7 (January-February): 40-59.
Consumer Research 14:387-403. MacKenzie, Scott B., Richard J. Lutz, and George Belch. 1986. “The
Gopinath, Mahesh and Richard P. Bagozzi. 1999. “Cognitive Appraisals Role of Attitude Toward the Ad as a Mediator of Advertising Effec-
Leading to Consumer Emotions and Action Tendencies.” Unpub- tiveness: A Test of Competing Explanations.” Journal of Marketing
lished working paper, University of Michigan. Research 23 (May): 130-143.
Green, Donald P., Susan L. Goldman, and Peter Salovey. 1993. “Mea- Mackie, Diane M. and Leila T. Worth. 1989. “Processing Deficits and the
surement Error Masks Bipolarity in Affect Ratings.” Journal of Per- Mediation of Positive Affect in Persuasion.” Journal of Personality
sonality and Social Psychology 64:1029-1041. and Social Psychology 57 (1): 27-40.
Hayes-Roth, B. 1977. “Evolution of Cognitive Structures and Pro- Mano, Haim and Richard L. Oliver. 1993. “Assessing the Dimensionality
cesses.” Psychological Review 84:260-278. and Structure of the Consumption Experience: Evaluation, Feeling,
Bagozzi et al. / THE ROLE OF EMOTIONS 205

and Satisfaction.” Journal of Consumer Research 20 (December): L. P. Mos, H. V. Rappard, and H. J. Stam. New York: Springer-Verlag,
451-466. 327-337.
Manstead, Antony S. R. and Hugh L. Wagner. 1981. “Arousal, Cognition, Perugini, Marco and Richard P. Bagozzi. 1998. “The Role of Desires and
and Emotion: An Appraisal of Two-Factor Theory.” Current Psycho- Anticipatory Emotions in Goal-Directed Behaviors: Extending the
logical Reviews 1: 35-54. Theory of Planned Behavior.” Unpublished working paper, Univer-
Marcel, Anthony J. 1983. “Conscious and Unconscious Perception: An sity of Michigan Business School.
Approach to the Relations Between Phenomenal Experience and Per- Petty, Richard E. and John T. Cacioppo. 1986. Communication and Per-
ceptual Processes.” Cognitive Psychology 15:238-300. suasion. New York: Springer-Verlag.
Markus, Hazel R. and Shinobu Kitayama. 1994. “The Cultural Con-
struction of Self and Emotion: Implications for Social Behavior.” In , David W. Schumann, Steven A. Richman, and Alan J. Strathman.
Emotion and Culture: Empirical Studies of Mutual Influence. Eds. 1993. “Positive Mood and Persuasion: Different Roles for Affect Un-
S. Kitayama and H. R. Markus. Washington, DC: American Psycho- der High and Low Elaboration Conditions.” Journal of Personality
logical Association, 89-130. and Social Psychology 64:5-20.
Martin, Leonard L., David W. Ward, John W. Achee, and Robert S. Wyer Jr. Pieters, Rik, Hans Baumgartner, and D. Allen. 1995. “A Means-End
1993. “Mood as Input: People Have to Interpret the Motivational Im- Chain Approach to Consumer Goal Structures.” International Jour-
plications of Their Moods.” Journal of Personality and Social Psy- nal of Research in Marketing 12 (3): 227-244.
chology 64 (3): 317-326. Plutchik, Robert 1980. Emotion: A Psychoevolutionary Synthesis. New
Milliman, Ronald E. 1982. “Using Background Music to Affect the Be- York: Harper & Row.
havior of Supermarket Shoppers.” Journal of Marketing 46 (Sum- Puto, Christopher P. and William D. Wells. 1984. “Informational and
mer): 86-91. Transformational Advertising: The Differential Effects of Time.” Ad-
. 1986. “The Influence of Background Music on the Behavior of vances in Consumer Research 11:638-643.
Restaurant Patrons.” Journal of Consumer Research 13:286-289. Reisenzein, Rainer 1983. “The Schachter Theory of Emotion: Two Dec-
Miniard, Paul W., Sunil Bhatla, and Randall L. Rose. 1990. “On the For- ades Later.” Psychological Bulletin 94:239-264.
mation and Relationship of Ad and Brand Attitudes: An Experimen- Richard, René, Joop van der Pligt, and Nanne de Vries. 1995. “Antici-
tal and Causal Analysis.” Journal of Marketing Research 27 pated Affective Reactions and Prevention of AIDS.” British Journal
(August): 290-303. of Social Psychology 34:9-21.
Mitchell, Andrew A. and Jerry C. Olson. 1981. “Are Product Attribute Richins, Marsha L. 1997. “Measuring Emotions in the Consumption Ex-
Beliefs the Only Mediator of Advertising Effects on Brand Atti- perience.” Journal of Consumer Research 24:127-146.
tude?” Journal of Marketing Research 18 (August): 318-332. Roese, Neal J. and James M. Olson, eds. 1995. What Might Have Been:
Murray, Noel, Harish Sujan, Edward R. Hirt, and Mita Sujan. 1990. “The The Social Psychology of Counterfactual Thinking. Mahwah, NJ:
Influence of Mood on Categorization: A Cognitive Flexibilty Inter- Lawrence Erlbaum.
pretation.” Journal of Personality and Social Psychology 59 (3): 411- Roseman, Ira J. 1991. “Appraisal Determinants of Discrete Emotions.”
425. Cognition and Emotion 5:161-200.
Nasby, William and Regina Yando. 1982. “Selective Encoding and Re- Rozin, Paul, Jonathan Haidt, and Clark R. McCauley. 1993. “Disgust.” In
trieval of Affectively Valent Information.” Journal of Personality and Handbook of Emotions. Eds. M. Lewis and J. M. Haviland. New
Social Psychology 43 (6): 1244-1253. York: Guilford, 575-594.
Nyer, Prashanth U. 1997a. “Modeling the Cognitive Antecedents of Post- Russell, James A. 1980. “A Circumplex Model of Affect.” Journal of
Consumption Emotions.” Journal of Consumer Satisfaction, Dissat- Personality and Social Psychology 39 (December): 1161-1178.
isfaction and Complaining Behavior 10:80-90. . 1997. “How Shall an Emotion Be Called.” In Circumplex Mod-
. 1997b. “A Study of the Relationships Between Cognitive Ap- els of Personality and Emotions. Eds. R. Plutchick and H. R. Conte.
praisals and Consumption Emotions.” Journal of the Academy of Washington, DC: American Psychological Association, 205-220.
Marketing Science 25 (4): 296-304.  and Albert Mehrabian. 1977. “Evidence for a Three-Factor The-
Oatley, Keith 1992. Best Laid Schemes: The Psychology of Emotions. ory of Emotions.” Journal of Research in Personality 11:273-294.
Cambridge, UK: Cambridge University Press. , Anna Weiss, and Gerald A. Mendelsohn. 1989. “Affect Grid: A
 and Philip N. Johnson-Laird. 1987. “Towards a Cognitive Theory Single-Item Scale of Pleasure and Arousal.” Journal of Personality
of Emotions.” Cognition and Emotion 1:29-50. and Social Psychology 57:493-502.
Obermiller, Carl. 1985. “Varieties of Mere Exposure: The Effects of Ruth, Julie A., Cele C. Otnes, and Frédéric F. Brunel. 1998. “Bonds and
Processing Style and Repetition on Affective Response.” Journal of Burdens: Gift Receipt and Relationship Reformulation.” Unpub-
Consumer Research 12:17-30. lished working paper, University of Washington.
Oliver, Richard L. 1980. “A Cognitive Model of the Antecedents and Salovey, Peter and Deborah Birnbaum. 1989. “Influence of Mood on
Consequences of Satisfaction Decisions.” Journal of Marketing Re- Health-Related Cognitions.” Journal of Personality and Social Psy-
search 17 (November): 460-469. chology 57:539-551.
. 1993. “Cognitive, Affective, and Attribute Bases of the Satisfac- Sanbonmatsu, David M. and Frank R. Kardes. 1988. “The Effect of
tion Response.” Journal of Consumer Research 20 (December): 418- Physiological Arousal on Information Processing and Persuasion.”
430. Journal of Consumer Research 15:379-395.
. 1994. “Conceptual Issues in the Structural Analysis of Con- Schachter, Stanley and Jerome E. Singer. 1962. “Cognitive, Social, and
sumption Emotion, Satisfaction, and Quality.” In Advances in Consumer Physiological Determinants of Emotional State.” Psychological Re-
Research, Vol. 21. Eds. Chris T. Allen and Deborah Roedder John. view 69:379-399.
Provo, UT: Association for Consumer Research, 16-22. Schaller, Mark and Robert B. Cialdini. 1990. “Happiness, Sadness, and
Olney, Thomas J., Morris B. Holbrook, and Rajeev Batra. 1991. “Con- Helping: A Motivational Integration.” In Handbook of Motiva-
sumer Responses to Advertising: The Effects of Ad Content, Emo- tion and Cognition: Foundations of Social Behavior, Vol. 2. Eds.
tions, and Attitude Toward the Ad on Viewing Time.” Journal of R. M. Sorrentino and E. T. Higgins. New York: Guilford, 265-296.
Consumer Research 17 (March): 440-453. Schwarz, Norbert. 1990. “Feelings as Information: Informational and
Ortony, Andrew, Gerald L. Clore, and Allan Collins. 1988. The Cognitive Motivational Functions of Affective States.” In Handbook of Motiva-
Structure of Emotions. Cambridge, UK: Cambridge University Press. tion and Cognition: Foundations of Social Behavior, Vol. 2. Eds.
Parker, Dianne, Antony S. R. Manstead, and Stephen G. Stradling. 1995. R. M. Sorrentino and E. T. Higgins. New York: Guilford, 527-561.
“Extending the Theory of Planned Behaviour: The Role of Personal  and Herbert Bless. 1991. “Happy and Mindless, But Sad and
Norm.” British Journal of Social Psychology 34:127-137. Smart? The Impact of Affective States on Analytic Reasoning.” In
Parkinson, Brian. 1995. Ideas and Realities of Emotion. London: Emotion and Social Judgments. Ed. Joseph P. Forgas. London: Perga-
Routledge. mon, 55-71.
Parrott, Gerrod W. 1988. “The Role of Cognition in Emotional Experi-  and Gerald L. Clore. 1983. “Mood, Misattribution, and Judg-
ence.” In Recent Trends in Theoretical Psychology. Eds. W. J. Baker, ments of Well-Being: Informative and Directive Functions of Affec-
206 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 1999

tive States.” Journal of Personality and Social Psychology 45 (3): , , and David J. Klein. 1994. “Effects of Mood on High
513-523. Elaboration Attitude Change: The Mediating Role of Likelihood
Judgments.” European Journal of Social Psychology 24:25-43.
, Fritz Strack, Detlev Kommer, and Dirk Wagner. 1987. “Soccer,
Rooms and the Quality of Your Life: Mood Effects on Judgments of , , and Stephen M. Smith. 1995. “Positive Mood Can In-
Satisfaction With Life in General and with Specific Life Domains.” crease or Decrease Message Scrutiny: The Hedonic Contingency
European Journal of Social Psychology 17:69-79. View of Mood and Message Processing.” Journal of Personality and
Shaver, Phillip, Judith Schwartz, Donald Kirson, and Cary O’Connor. Social Psychology 69 (1): 5-15.
1987. “Emotion Knowledge: Further Exploration of a Prototype Ap- Westbrook, Robert A. 1987. “Product/Consumption-Based Affective
proach.” Journal of Personality and Social Psychology 52:1061- Responses and Postpurchase Processes.” Journal of Marketing Re-
1086. search 24 (August): 258-270.
 and Richard L. Oliver. 1991. “The Dimensionality of Consump-
Shimp, Terence A. 1981. “Attitude Toward the Ad as a Mediator of Con- tion Emotion Patterns and Consumer Satisfaction.” Journal of Con-
sumer Brand Choice.” Journal of Advertising 10:9-15. sumer Research 18 (June): 84-91.
. 1991. “Neo-Pavlovian Conditioning and Its Implications for Wiles, Judith A. and T. Bettina Cornwell. 1990. “A Review of Methods
Consumer Theory and Research.” In Handbook of Consumer Behav- Utilized in Measuring Affect, Feelings, and Emotion in Advertising.”
ior. Eds. Thomas S. Robertson and Harold H. Kassarjian. Englewood Current Issues and Research in Advertising 13:241-275.
Cliffs, NJ: Prentice Hall, 162-187. Worth, Leila T. and Diane M. Mackie. 1987. “Cognitive Mediation of
Smith, Craig A. and Phoebe C. Ellsworth. 1985. “Patterns of Cognitive Positive Affect in Persuasion.” Social Cognition 5:76-94.
Appraisals in Emotion.” Journal of Personality and Social Psychol- Yi, Youjae. 1990. “A Critical Review of Consumer Satisfaction.” In Re-
ogy 48:813-838. view of Marketing 1990. Ed. V. A. Zeithaml. Chicago: American
Srull, Thomas K. 1983. “Affect and Memory: The Impact of Affective Marketing Association, 68-123.
Reactions in Advertising on the Representation of Product Informa- Zajonc, Robert B. 1968. “Attitudinal Effects of Mere Exposure.” Journal
tion in Memory.” In Advances in Consumer Research, Vol. 10. Eds. of Personality and Social Psychology Monograph Supplement 9:1-
Richard P. Bagozzi and Alice Tybout. Ann Arbor, MI: Association 27.
for Consumer Research, 520-525. . 1980. “Feeling and Thinking: Preferences Need No Inferences.”
. 1984. “The Effects of Subjective Affective States on Memory American Psychologist 35:151-175.
and Judgment.” In Advances in Consumer Research, Vol. 11. Ed. . 1984. “On the Primacy of Affect.” American Psychologist
Thomas C. Kinnear. Provo, UT: Association for Consumer Research, 39:117-123.
530-533. . 1998. “Emotions.” In The Handbook of Social Psychology. Eds.
Stayman, Douglas M. and David A. Aaker. 1988. “Are All the Effects of D. T. Gilbert, S. T. Fiske, and G. Lindzey. Boston: McGraw-Hill, 591-
Ad-Induced Feelings Mediated by Aad?” Journal of Consumer Re- 632.
search 15:368-373.  and Hazel Markus. 1982. “Affective and Cognitive Factors in
 and Rajeev Batra. 1991. “Encoding and Retrieval of Ad Affect in Preferences.” Journal of Consumer Research 9 (September): 123-
Memory.” Journal of Marketing Research 28 (May): 232-239. 131.
Stein, Nancy L., Maria D. Liwag, and Elizabeth Wade. 1996. “A Goal- Zevon, Michael A. and Auke Tellegen. 1982. “The Structure of Mood
Based Approach to Memory for Emotional Events: Implementations Change: An Idiographic/Nomothetic Analysis.” Journal of Personal-
for Theories of Understanding and Socialization.” In Emotion: Inter- ity and Social Psychology 43:111-122.
disciplinary Perspectives. Eds. R. D. Kavanaugh, B. Zimmerberg, Zillman, Dolf. 1971. “Excitation Transfer in Communication-Mediated
and S. Fein. Mahwah, NJ: Lawrence Erlbaum, 91-118. Aggressive Behavior.” Journal of Experimental Social Psychology
Taylor, Shirley. 1994. “Waiting for Service: The Relationship Between 7:419-434.
Delays and Evaluations of Services.” Journal of Marketing 58 . 1983. “Transfer of Excitation in Emotional Behavior.” In Hand-
(April): 56-69. book of Social Psychology. Eds. John T. Cacioppo and Richard E.
Teasdale, John D. and Sarah J. Fogarty. 1979. “Differential Effects of In- Petty. New York: Guilford, 215-230.
duced Mood on Retrieval of Pleasant and Unpleasant Events From
Episodic Memory.” Journal of Abnormal Psychology 88 (3): 248-257.
 and Louise M. Russell. 1983. “Differential Aspects of Induced ABOUT THE AUTHORS
Mood on the Recall of Positive, Negative and Neutral Words.” British
Journal of Clinical Psychology 22:163-171.
Tulving, Endel and Zena Pearlstone. 1966. “Availability Versus Accessi- Richard P. Bagozzi is the Dwight F. Benton Professor of Behav-
bility of Information in Memory for Words.” Journal of Verbal ioral Science in Management at the University of Michigan Busi-
Learning and Verbal Behavior 5:381-391. ness School. He is a graduate of Northwestern University and has
Verbeke, Willem and Richard P. Bagozzi. 1998. “Self-Conscious Emo-
been a faculty member at the University of California–Berkeley,
tions in the Context of Personal Selling: Performance Is a Function of
the Ability to Cope With Shame and Embarrassment.” Unpublished the Massachusetts Institute of Technology, and Stanford Univer-
working paper, University of Michigan. sity. He is currently doing research in emotions, goal-directed be-
 and . 1999. “Coping With Anxiety in Personal Selling: A havior, and social identity theory.
Study of Financial Service Salespeople.” Unpublished working pa-
per, University of Michigan. Mahesh Gopinath is an assistant professor of marketing in the
Watson, David and Auke Tellegen. 1985. “Toward a Consensual Struc-
ture of Mood.” Psychological Bulletin 98 (September): 219-235. A. B. Freeman School of Business at Tulane University. He is a
Wegener, Duane T. and Richard E. Petty. 1994. “Mood-Management graduate of the University of Michigan Business School and is
Across Affective States: The Hedonic Contingency Hypothesis.” doing research in emotions and customer satisfaction.
Journal of Personality and Social Psychology 66 (6): 1034-1048.
 and . 1996. “Effects of Mood on Persuasion Processes: Prashanth U. Nyer is an assistant professor of marketing in the
Enhancing, Reducing, and Biasing Scrutiny of Attitude-Relevant In-
formation.” In Striving and Feeling: Interactions Between Goals and
School of Business and Economics at Chapman University. He is
Affect. Eds. L. L. Martin and A. Tesser. Mahwah, NJ: Lawrence Erl- a graduate of the University of Michigan Business School and is
baum, 329-362. doing research in emotions and customer satisfaction.