The three faces of Eve: Strategic displays of positive ...ibavenegotiations.weebly.com/uploads/1/0/8/0/10805615/three_faces... · of positive, negative, ... managers strategically

Organizational Behavior and Human Decision Processes 99 (2006) 81–101

www.elsevier.com/locate/obhdp

The three faces of Eve: Strategic displays of positive, negative, and neutral emotions in negotiations

Shirli Kopelman a,¤, Ashleigh Shelby Rosette b,d, Leigh Thompson c

a The Ross School of Business, University of Michigan, 701 Tappan St., Ann Arbor, MI 48109-1234, USAb The Fuqua School of Business, Duke University, Box 90120, One Towerview Rd. Durham, NC 27708-0120, USA

c Kellogg School of Management, Northwestern University, 2001 Sheridan Rd. Evanston, IL 60208, USAd Bauer College of Business, University of Houston, 334 Melcher Hall, Houston, TX, 77204-6021, USA

Received 17 September 2002Available online 21 September 2005

Abstract

In a series of laboratory experiments, we tested the inXuence of strategically displaying positive, negative, and neutral emotions onnegotiation outcomes. In Experiment 1, a face-to-face dispute simulation, negotiators who displayed positive emotion, in contrast tonegative or neutral emotions, were more likely to incorporate a future business relationship in the negotiated contract. In Experiment2, an ultimatum setting, managers strategically displaying positive emotion were more likely to close a deal. This eVect was mediatedby negotiators’ willingness to pay more to a negotiator strategically displaying positive versus negative emotions. In Experiment 3,display of positive emotion was a more eVective strategy for gaining concessions from the other party in a distributive setting. Nego-tiators made more extreme demands when facing a negotiator strategically displaying negative, rather than positive or neutral, emo-tions. Implications for strategic display of emotion in negotiations are discussed. 2005 Elsevier Inc. All rights reserved.

Keywords: Emotion; AVect; Negotiation; Strategy; Ultimatum bargaining; Distributive gains; Dispute resolution; Relationships

The scholarly inquiry of negotiation behavior and tional states as predictors of negotiated outcomes. Both
outcomes has been heavily inXuenced by cognition, bothfrom a prescriptive (e.g., RaiVa, 1982) and from adescriptive perspective (Neale & Bazerman, 1991;Thompson & Hastie, 1990). A new generation ofresearchers has challenged this cognitive tradition bystudying how emotions aVect interdependent decisionmaking (Allred, 1999; Baron, 1990; Barry & Oliver,1996; Barsade, 2002; Carnevale & Isen, 1986; Forgas,1998; Keltner, Ellsworth, & Edwards, 1993; Kramer,Newton, & Pommerenke, 1993; Kumar, 1997; Sutton &Rafaeli, 1988; Thompson, Medvec, Seiden, & Kopelman,2001; Thompson, Nadler, & Kim, 1999). They view emo-
* Corresponding author. Fax: +1 734 764 3146.E-mail addresses: [email protected] (S. Kopelman), arosette@

duke.edu (A.S. Rosette), [email protected](L. Thompson).

0749-5978/$ - see front matter 2005 Elsevier Inc. All rights reserved.doi:10.1016/j.obhdp.2005.08.003

cognitive and emotional approaches to negotiation havehighlighted not only outcomes related to the negotiableitems on the table, but also outcomes related to the socialrelationship between the parties at the table (Greenhalgh& Chapman, 1995; Kopelman & Olekalns, 1999; Kramer& Messick, 1995; Lawler & Yoon, 1995; Leung & Fan,1997; McGinn & Keros, 2002; Sondak, Neale, & Pinkley,1999; Valley, Neale, & Mannix, 1995). Barry (1999)stressed the important role that emotions can play at thenegotiating table and argued that “the willful use ofemotional display or expression [can be used] as a tacti-cal gambit by an individual negotiator” (p. 3). We buildon this assertion and conceptualize emotions as a delib-erate behavioral strategy that is available to negotiators.

Empirical research on negotiations has focused ontwo distinct yet related subsets of a more general cate-gory called aVect: emotions and mood. Mood is

mailto: [email protected]







82 S. Kopelman et al. / Organizational Behavior and Human Decision Processes 99 (2006) 81–101

considered a more diVuse psychological state, whereasemotion is a diVerentiated response to a speciWcsituation (e.g., Barry, 1999; Schwarz & Clore, 1996). Incontrast to enduring moods, emotions are relativelyshort-lived and are either triggered by an identiWableevent or brought on as a means to achieving an aspiredend. We extend previous research that either examinedisolated unrelated emotional stimuli (e.g., pleasant scent,funny cartoons, sweet food, or a gift) that induced posi-tive or negative emotion (e.g., Baron, 1990; Carnevale &Isen, 1986; Isen, Daubman, & Nowicki, 1987) or mea-sured negotiator’s mood (e.g., Forgas, 1998) by examin-ing the eVectiveness of strategically displayed emotionson both economic and relational negotiation outcomes.

Strategic display of emotion in negotiations

The study of strategically displayed emotions in negoti-ations rests on two fundamental assumptions: (1) negotia-tors can control their behavioral emotional display duringsocial interactions, and (2) emotional displays can be com-municated convincingly to the other party. In line with theWrst assumption, we propose that negotiators can expressa desired emotion and that the behavioral display of theemotion, like other strategies in their negotiation tool kit,may serve as a means to attain a desired outcome. Delib-erate regulation of emotion may involve diVerent levels ofacting (Coté, 2005; Grandey, 2000; Gross, 1998; Hochs-child, 1983). In deep acting, internally experienced andexternally displayed emotions are aligned and thereforeconsidered authentic. At other times, however, the negoti-ator’s emotional strategy may be at odds with how he orshe feels at that moment. This type of emotional regula-tion is called surface acting, and because it inXuences onlythe display of emotion, not the internal experience of emo-tion, it is considered to be inauthentic (Coté, 2005).Whether the strategic behavioral display of emotion rep-resents emotions psychologically experienced at thatmoment (i.e., the negotiator strategically “harnesses feltemotions,” perhaps exaggerating them as he displaysthem) or whether it represents deliberately feigned emo-tion (i.e., the negotiator “wears an emotional mask” hid-ing her current feelings), skilled negotiators mayintentionally adjust their emotional display in a desireddirection by either amplifying or suppressing their ownexperienced emotion (Hochschild, 1983; Levenson,1994a). We do acknowledge that there may be ethical con-cerns regarding the intentional display of emotions inwhat may range from a less than forthright or honestmanner to an outright manipulative tactic; but, just asmisrepresentation of information is a negotiation strategyworthy of investigation, we maintain that so is the studyof the deliberate display of emotion.

The second assumption upon which our research restsis that emotional displays can be communicated con-

vincingly to the other party. That is, behavioral displayof emotion by a focal negotiator will be perceived by atarget negotiator. Whether through primal processes likeemotional contagion (e.g., HatWeld, Cacioppo, &Rapson, 1992) or higher-level cognitive processes, weassume that the target negotiator’s actions can be inXu-enced by the strategic display of emotion. Our concern isnot how strategic display of emotion is processed by thetarget negotiator. Neither the debate on the primacy ofemotions versus cognitions (Lazarus, 1984; Zajonc,1984) is addressed by our empirical investigation nor iswhether strategically displayed emotions are perceivedas genuine or contrived by the target negotiator. Ourpresumption is simply that the strategic display of emo-tion will be perceived. In fact, recognition of emotionaldisplay in social interactions is considered an evolution-ary adaptive human characteristic (Ekman, 1993). Thus,implicit to the argument that expressed emotion can beused strategically in negotiations, is the assumption thatemotional experiences can be controlled and convinc-ingly displayed in support of strategic action.

In this paper, strategic behavioral display of emotionrefers to emotion intentionally expressed by the focalnegotiator to attain a desired outcome, and is experi-mentally manipulated with regard to three general emo-tional approaches: deliberately positive, negative, andneutral display of emotions. Categorizing strategic emo-tions into such broad categories may constitute an over-simpliWcation of emotional phenomena, but since, to thebest of our knowledge, this is the Wrst empirical investi-gation of strategic emotional display in negotiations webelieved it was an appropriate starting point for this lineof empirical research. In general, negative emotiondiVers categorically from positive emotion from an evo-lutionary perspective. Negative emotions represent spe-ciWc action tendencies that narrow in on a speciWc subsetof behavioral options and are necessary for survival inlife-or-death situations (Tooby & Cosmides, 1990). Theymobilize appropriate autonomic support, making run-ning, for example, possible when in the face of danger(Levenson, 1994b). In contrast, positive emotions are rel-atively vague and underspeciWed (Fredrickson &Levenson, 1998). They are more likely to occur whenpeople feel safe and satiated (Frijda, 1986) and facilitateapproach behavior, exploration, and activity engage-ment. Social and personal resources accrued duringstates of positive emotions are theorized to be durableand build long-term resilience (Fredrickson, 2001).

There are several mechanisms by which strategicallydisplayed emotions, whether positive, negative, or neu-tral, may impact the social interaction between negotia-tors. First, displayed emotions may convey informationand inXuence strategic information processing (e.g.,Forgas & George, 2001). For example, assuming thetarget negotiator accurately perceives the display of pos-itive emotion each time the focal negotiator discusses a

S. Kopelman et al. / Organizational Behavior and Human Decision Processes 99 (2006) 81–101 83

particular issue, he may infer that this speciWc negotiableissue is of supreme importance to the focal negotiator.Whether learning that this issue is important to the focalnegotiator leads to identifying a compatible issue or toleveraging a trade-oV, this information could helpexpand the pie of resources and thus would be valuableinformation exchange.

Second, emotions may serve as a means of persuasion(e.g., Forgas, 2001) and thus may constitute a manipula-tive negotiation tactic that leads the other party to act ina way he otherwise would not have otherwise chosen.For example, strategic emotional display by the focalnegotiator may engender an emotional reaction in thetarget negotiator that leads him to concede on a negotia-ble issue. Consider a focal negotiator displaying an out-pour of positive emotion and ingratiating himself uponthe target negotiator. Prior research suggests that a tar-get negotiator, even if she is completely aware that aningratiation tactic is put into play, is still likely torespond by acquiescing to the request made (Yukl &Falbe, 1990). In contrast, strategically negative emo-tional display could engender negative emotion in theother party and thus manipulate the target negotiatorinto early foreclosure through greater concessions,because exiting the socially unpleasant situationbecomes a more salient concern than gaining economicvalue.

Given the above conceptual framework, we set out toinvestigate how displayed emotion inXuences the negoti-ation process and outcomes. Next, we review negotiationtheory, research, and tactics that correspond to the posi-tive, negative, and neutral strategic emotionalapproaches.

Positive emotion

Although there have not been studies on strategicpositive display of emotion, numerous researchers haveexamined how induced positive emotion and measuredpositive mood inXuence negotiations. A general Wndingis that induced positive emotion and good moodincrease cooperative tactics and enhance the quality ofagreements. A large body of social psychologicalresearch suggests that positive aVect leads to better deci-sions and improved consequences for the social actor ina variety of settings (see Isen, 1987 for a review). In whathas become a seminal study of positive aVect and negoti-ation, (Carnevale & Isen, 1986), positive emotion wasinduced prior to a negotiation task by having negotia-tors perform a seemingly unrelated task of sorting car-toons and oVering them a small gift. Negotiatorsexperiencing positive emotion subsequently reachedmore mutually beneWcial outcomes in a face-to-face bar-gaining task than did the control group. Subsequentstudies that similarly manipulated positive aVectreported similar Wndings, and together these suggest that

positive negotiators realize higher individual and jointgains on both integrative and distributive negotiationtasks than do negotiators in a neutral aVect condition(Baron, 1990; Carnevale & Isen, 1986; Hollingshead &Carnevale, 1990).

Favorable negotiated outcomes are attained becausepositive aVect inXuences how people process informa-tion and promote creative thinking (Isen et al., 1987;Kramer et al., 1993), which, in turn, makes negotiatorsmore likely to engage in innovative problem solving(Carnevale & Isen, 1986). People in a positive mood sethigher goals (Baron, 1990) so that they may be able toclaim a larger slice of the increased pie. The AVect Infu-sion Model (AIM, Forgas, 1995) is consistent with theview that positive emotions enhance negotiators’ eVec-tiveness. According to AIM, emotions inXuence cogni-tive evaluations and negotiators adopt aVect-congruentbargaining strategies. For example, happy negotiatorsdevelop more cooperative tactics than unhappy negotia-tors (Forgas, 1998).

Negative emotion

Social psychological research extols the beneWts ofpositive aVect, but another perspective argues that incontrast to positive emotional displays, negotiators whodisplay negative emotion can be extremely eVective atthe bargaining table. By ranting and raving and beingunpleasant, a negotiator can position the negotiation inhis favor and convince the other party to succumb to hisdemands. The good-cop bad-cop strategy (e.g., Brodt &Tuchinsky, 2000; Hilty & Carnevale, 1993) leverages theeVect of negative emotions. Criminal investigators capi-talize on the perceptual contrast (Cialdini, 1993)between a friendly negotiator who establishes rapportthrough ampliWed displays of compassion and is onlyable to extract concessions because of the demands pre-viously made by a negative co-conspirator (Rafaeli &Sutton, 1991). Similarly, Hilty and Carnevale (1993)found that when two people negotiated as a team andsequentially shifted between cooperative and competi-tive strategies (which may be viewed as positive and neg-ative approaches), greater concessions were made by theother party and the distance between oVers was reduced.Although in the latter study emotions per se were notmeasured, it suggests that negative emotional displaysmight be used strategically in the negotiation process.

A negative approach also is supported by the squeakywheel gets the grease principle, which states that a nego-tiator should demonstrate an unwillingness to moveaway from a stated position by vocally escalating thelevel of demands (Singelis, 1998). For example, a negoti-ator who issues more explicit threats toward the end of anegotiation, in contrast to implicit threats or no threats,is more likely to elicit concession from the target(Sinaceur & Neale, 2005). Whereas those who squeak


may be negatively perceived; self-regulation theory(Baumeister, Leith, Muraven, & Bratslavsky, 1998) sug-gests that people may concede to the demands of anopponent displaying negative emotion because peopleself-regulate to prolong positive moods and exposure tonegative stimuli minimizes positive moods. One underly-ing process supporting the eVectiveness of displayingnegative emotion, thus, seems to be the willingness of thetarget negotiator to make larger concessions to exit anunpleasant social situation.

Accordingly, one could expect negotiators strategi-cally displaying negative emotions to attain higher gains.In fact, recent experimental Wndings support this claim.Sinaceur and Tiedens (2005) found that strategic displayof anger was eVective in extracting value in face-to-facenegotiations, but only when the other party perceived hisalternatives to be weak. This occurred because the strate-gic display of anger communicated toughness to theother party, which supports the squeaky wheel gets thegrease view because targets made more concessions to anopponent they perceived as tough (Sinaceur & Tiedens,2005). In a series of studies, negotiators in a non-interac-tive setting received a verbal message describing theemotions of the focal negotiator. The Wndings suggestthat negative emotion could be economically beneWcialbecause participants who believed they faced a negativenegotiator made larger concessions (Van Kleef, DeDreu, & Manstead, 2004a, 2004b).

On the other hand, in some settings, display of nega-tive emotion may be detrimental to the negotiation pro-cess and outcomes. For instance, insulting oVers thatgenerate negative aVect are rejected in ultimatum bar-gaining settings (Pillutla & Murnighan, 1996). Deutsch(1973) described a series of bargaining games thatexplored the use of threat and suggested that althoughthe availability of coercive power increased aspirationlevels, the use of threat and aggression interfered withthe development of proWtable agreements. Consistently,Allred and co-workers (1997) found that negotiatorswho experienced high anger and low compassionachieved lower joint gains and had a reduced desire towork with each other in the future. Thus, strategic dis-play of negative emotion in a dispute setting may lead toa conXict spiral that is diYcult to break (Ury, Brett, &Goldberg, 1988).

Neutral emotion

A quite diVerent perspective cautions negotiators tosuppress and conceal all emotions. This strategy suggeststhat a negotiator is best advised to neither feel norexpress emotion at the bargaining table because emotionis viewed as a weakness and leads to vulnerability. Anegotiator who displays emotions, particularly those ofrelief, satisfaction, and approval, risks settling for aworse outcome than does the poker-faced negotiator

who deliberately masks any emotional display. It isimportant to note that the neutral emotional displaystrategy is not based in scientiWc theory or research, butis more of a lay-theory or prescriptive model of negotia-tions. Popular books advise that “ƒpeople in an emo-tional state do not want to think, and they areparticularly susceptible to the power of suggestion froma clever opponentƒ” (Nierenberg, 1968, p. 46) and sug-gest that being rational requires an absolute absence ofemotions.

Support for this approach can be drawn from Janisand Mann’s (1977) model of decision making. Theyargue that decision makers experiencing high levels ofemotional stress often conduct incomplete searches forinformation, complete sub-standard appraisals, and dis-regard contingency planning. These adverse negotiationstrategies often lead to defective decisions. The risk liter-ature also advises negotiators to be emotionally neutralbecause risk-seeking (thrill-based) or risk-averse (fear-based) behavior can lead to suboptimal decision makingand inferior negotiated outcomes (for reviews of risk-based negotiations see Bazerman & Neale, 1992; Neale& Bazerman, 1991). In fact, recent empirical Wndingssupport the perspective that people physiologicallydeprived of emotional reactions actually might makebetter decisions (Damasio, 1994). Patients with stablefocal lesions in brain regions related to emotion who donot experience emotions made more rational and ulti-mately more proWtable investment decisions thanpatients with focal lesions in brain regions unrelated toemotion and normal participants with no brain damage(Shiv, Loewenstein, Bechara, Damasio, & Damasio,2005). Hence, there may be advantages when negotiatorsdo not display emotions. However, there also could benegative implications for a poker-faced strategy. Forexample, a non-emotional approach may impede thedevelopment of rapport and trust between negotiators(Drolet & Morris, 2000).

Overview of experiments

Embracing a theoretical framework of goal-directedemotional display (Barry, 1999; Barry & Oliver, 1996;Thompson et al., 2001; Thompson et al., 1999), weempirically examined the eVect of strategically display-ing positive, negative, and neutral emotions. Experiment1 tested the eVect of strategic display of emotion in aninteractive dispute on immediate outcomes (dispute res-olution), and on expectations concerning a future busi-ness relationship. Negotiators were coached tostrategically display positive, negative, or neutral emo-tions during a one-hour-long negotiation process. Incontrast, Experiments 2 and 3 tested the eVect of strate-gic emotional display in a more controlled paradigm.Rather than having participants idiosyncratically


display strategic emotions, a professional actor uni-formly presented each emotional display to negotiatorsthrough a videotaped negotiation oVer. The oVer wasconsistent with respect to content. In Experiment 2, par-ticipants responded to a take-it-or-leave-it oVer made bythe professional actor and could either accept or rejectthe ultimatum. Experiment 3 extended the investigationto proposed counteroVers and concession-making in adistributive setting. Together, these three experimentswere designed to provide an initial test of the feasibilityand impact of strategically displaying emotion innegotiations.

Experiment 1

Disputes are a prime setting in which to examine theimpact of emotional display on negotiated outcomes. Ina dispute, each party is convinced that his version of thetruth is justiWed and assumes the other party is inten-tionally trying to harm him (Ury et al., 1988). Further-more, in contrast to deal-making, in disputes parties areinterdependent with respect to their alternatives to thenegotiated agreement; they either Wnd a mutually beneW-cial resolution or escalate beyond the negotiation pro-cess to a diVerent form of resolution. When the stakes ofthe perceived conXict are high and a negotiator cannotsimply disengage and walk away if negotiated terms areunacceptable to him, he is likely to experience and possi-bly display emotions. The question is which type of emo-tional display is strategically favorable to the negotiatoror whether deliberately concealing all emotions is prefer-able.

To overcome the conXict and move toward a con-structive future business relationship, negotiators maystrategically adopt a positive negotiation approach. Forexample, following a highly emotional argument with abusiness partner, a negotiator may deliberately begin thesubsequent discussion with a token of appreciation. Akind word, a compliment, an apology, or any display ofpositive emotional exchange would re-aYrm the goodrelationship and common goals that initially led bothparties to embark on the joint business opportunity.Even if a negotiator does not directly apologize forrecent acrimonious exchanges, being especially friendlyor otherwise positive may signal the desire to reframe thesituation and help create a constructive atmosphere fordispute resolution. By diVusing the bitter feelings, strate-gic display of positive emotion could increase the likeli-hood of a creative future-oriented and interest-basedsolution and may therefore increase the likelihood ofdispute resolution.

In contrast, negotiators who strategically displaynegative emotion may maintain an emotional tone con-sistent with the disputed problem and thus may be lesseVective in resolving disputes. A negative negotiator

may be more likely to hold to her positions and thusrisk reaching an impasse (Bazerman & Neale, 1982).Furthermore, when a negotiator displays negative emo-tion she risks falling into the trap of a conXict spiral. Ifthe approach generates anger, it could instigate retalia-tion by the other party (Allred, 1999) and subsequentlymay lead to a deadlock, or even an escalation of conXict(Ury et al., 1988). Thus, we make the followingprediction:

Hypothesis 1. In a dispute setting, negotiators who stra-tegically display negative emotion will be less likely toreach an agreement than negotiators who strategicallydisplay positive emotion.

Beyond the inXuence of emotional display on whetheran agreement will, altogether, be attained, strategic dis-play of emotion also may inXuence the quality of thenegotiated agreements. A central issue in dispute resolu-tion is developing a mutually agreeable future relation-ship (Greenhalgh & Chapman, 1995; Pruitt & Carnevale,1993). We suggest that the strategic display of emotionsalso will impact the likelihood that parties would antici-pate a constructive future relationship following the dis-pute. We argue that a positive approach more so than anegative approach may help negotiators not only over-come the uncooperative exchanges brought on by therejection of one party’s claim—the source of the con-Xict—but that display of positive emotions may mendthe relationship. Positive emotional convergencebetween negotiators (Anderson, Keltner, & John, 2003)is likely to broaden people’s self-conceptions to includeothers to a greater degree (Fredrickson, 2001; Waugh &Fredrickson, 2002) and thus may enhance the likelihoodof continued joint business projects. In contrast to posi-tive emotion, anger may diminish negotiators’ desire towork with each other in the future (Allred, Mallozzi,Matsui, & Raia, 1997). Thus, we predict that:

Hypothesis 2. In a dispute setting, negotiators who stra-tegically display positive emotion will be more likely toexpect a future business relationship with the other partyfollowing dispute resolution than negotiators who stra-tegically display negative emotion.

Method

ParticipantsA total of 274 executive MBA students enrolled in a

negotiation course participated in the study, thus 137dyads were created. The study design included three con-ditions: (1) positive emotional display (N D 47); (2) nega-tive emotional display (N D 44); and (3) neutralemotional display (N D 46). Participants were randomlyassigned to conditions. The design did not pit positivenegotiator against neutral or negative negotiator in thesame dyad. Instead, the comparison of emotional dis-


plays occurred between dyads. By restricting a single typeof emotional display to a negotiating dyad (positive, neg-ative, or neutral), we could better isolate the inXuence ofthe emotional strategy on the negotiated outcome.

ManipulationsTo manipulate each of the three emotional displays,

explicit written instructions were pre-tested and pro-vided to the participants. Participants in the positiveemotional display condition were coached to develop apositive bargaining style by using positive emotion, leav-ing the other party feeling good, avoiding hostility, cre-ating positive relationships, and framing the negotiationas a partnership. In the negative emotional display con-dition, participants were coached to develop a resolutebargaining style by being persistent, passionately com-mitted to their beliefs, beginning with extreme positions,employing negative reinforcement, and being tough.Finally, in the neutral emotional display condition, par-ticipants were coached to develop a rational bargainingstyle by controlling their emotional display, making stra-tegic decisions, being professional, developing rationalstrategies, and thinking logically. The materials used forcoaching strategic emotional display are presented inAppendix A.

In each experimental condition, half the dyads wererandomly assigned to a matched emotional display con-dition (e.g., both negotiators were coached to displaypositive emotion), whereas the other half were randomlyassigned to a single emotional display condition (e.g.,only one party was coached to display positive emotionand the other was not coached). By examining bothmatched and single display dyads, we were able to assesswhether the manipulation of emotional display wasadditive (i.e., whether outcomes would be more pro-nounced when both negotiators displayed the same emo-tional strategy whether positive, negative, or neutral). Inthe single display condition, the other party was asked todevelop a strategy for the upcoming negotiation, with nospeciWc mention of emotions. All roles were randomlyassigned and each participant saw only one set of mate-rials (positive, negative, neutral coaching or in the caseof single-display, no emotional display coaching). Partic-ipants were led to believe the experiment was about dis-pute-resolution strategies and no mention of emotionswas made until the experiment was debriefed.

Procedures and taskParticipants in all conditions were randomly paired

with another participant and then randomly assigned toplay the role of one of the two disputants in the negotia-tion. The dispute simulation, Viking Investments(Greenhalgh, 1993), concerned a conXict between a con-dominium developer and a carpentry contractor. Thebackground information provided to the participantsexplained that the developer and the contractor had a

multi-faceted business relationship and had conductednumerous business transactions in the past. The disputenegotiated was about the cost of work performed by thecontractor for the developer, the renewal of a loan, andthe renewal of an oYce lease. If they settled the dispute,there was an opportunity for them to work together onfuture engagements. If, no settlement was reached, how-ever, the background information implied that the con-tractor would go bankrupt and the developer wouldforego future investment opportunities. Participantswere given information describing the interactionbetween the two parties that led to the dispute and weretold that they would meet with the other party to discussthe disputed contract. The emotional display manipula-tion was attached to the front of the role materials.

After participants read their materials, they wereintroduced to the other party and given one hour tonegotiate. We determined through pre-testing that onehour was a suYcient amount of time to resolve the dis-pute. At the end of one hour, both parties wereinstructed to jointly complete a written document detail-ing the nature of their agreement (or lack thereof). Fol-lowing this, participants were separated and eachcompleted a post-negotiation questionnaire. Participantsthen were debriefed.

MeasuresManipulation check. To test our manipulation, followingthe negotiation, participants used items on a 7-pointLikert scale, anchored by 1 (not at all) and 7 (very much),to rate their own (“I was”) and their partner’s (“Partnerwas”) emotional display during the negotiation. Basedon ratings of each negotiator’s own emotional display,the following three constructs were created: self/positive,self/negative, and self/neutral. Positive emotional displayfor the negotiator (self-positive) was measured with oneitem, friendly. A high score indicated that the negotiatorperceived him/herself to have displayed positive emo-tion. Negative emotional display for the negotiator (self-negative) was measured with three items: aggressive,insistent, and angry. A high score indicated that thenegotiator perceived him/herself to have displayed nega-tive emotion (Cronbach’s �D .76). Neutral emotionaldisplay for the negotiator (self-neutral) was measuredwith one item, rational. A high score indicated that thenegotiator perceived him/herself to have displayed neu-tral emotion. Based on ratings of each negotiator regard-ing their negotiation partner, the following threeconstructs were created: partner/positive, partner/nega-tive, and partner/neutral. Positive display of emotionsfor the negotiator’s partner (partner-positive) was mea-sured with one item, friendly. Negative display of emo-tion for the negotiator’s partner (partner-negative) wasmeasured with three items: aggressive, insistent, andangry (Cronbach’s �D .71). Neutral display of emotionfor the negotiator’s partner (partner-neutral) was


measured with one item, rational. Both self and partneremotional constructs were used to test if the display ofemotions was successfully manipulated. Given our fun-damental assumptions about behavioral display of emo-tion (i.e., displayed emotion can be controlled andconvincingly communicated), we created two measuresto test the manipulation.

The Wrst measure, emotional contagion (EC), testedthe emergence of positive, negative, or neutral emotionssubsequent to strategic display of positive, negative, orneutral emotions, respectively by one or both negotia-tors. Emotional contagion was conceptualized as a prim-itive, unconscious process that results from a tendency toautomatically mimic and synchronize movements,expressions, postures, and vocalizations of another per-son (e.g., HatWeld et al., 1992). To test whether dyadscoached in one of the three strategic emotionalapproaches (whether single or matched display) weremore positive/negative/neutral following a one hourface-to-face interaction, we created a dyad level measureof emotional contagion by adding the self-ratings madeby the focal negotiator (self-positive/negative/neutral) tothe self-ratings made by the target negotiator (self-posi-tive/negative/neutral). For example, a dyad’s positiveemotional contagion score was calculated by adding thefocal negotiator’s self-positive rating to the target nego-tiator’s self-positive rating; the higher the score, thehigher the dyad’s positive emotional approach. Thus, wecreated three measures of emotional contagion: (1) posi-tive emotional contagion (positive EC); (2) negativeemotional contagion (negative EC); and (3) neutral emo-tional contagion (neutral EC). We note that the neutralcondition is not a pure control, since negotiators’ werecoached, but it is emotionally neutral and serves as abenchmark. The values for the emotional contagionmanipulation check ranged from 2 to 14, whereby a highscore indicated a high reciprocal display of the indicatedemotion by the two members in the dyad.

We expected emotional contagion to be higher in thematched- versus single display condition. If, for example,only one of the dyad members was coached to displaypositive emotion, the displayed emotions could seepthrough and inXuence the other negotiator, who in turnalso would become more positive. When both partieswere coached to display positive emotion, the same pro-cess of emotional contagion could result in an exponen-tial intensiWcation of the emotions. But, even in thesingle display condition, when only one negotiator wascoached to display an emotion, we expected dyads in thesingle-positive display condition to engender more posi-tive emotion than single display of neutral and negativeemotions. Likewise, we expected single-negative displaydyads to engender more negative emotion than singledisplay neutral and positive conditions.

The second measure, emotional accuracy (EA) testedwhether the focal negotiator’s displayed emotions were

correctly assessed by the target negotiator. To constructan emotional accuracy index, we Wrst measured thediVerence between the negotiators’ displayed emotion(as indicated by their self-positive/negative/neutral) andthe ratings given to them by the other party (partner-positive/negative/neutral). This resulted in threemeasures: positive emotional accuracy (positive EA),negative emotional accuracy (negative EA), and neutralemotional accuracy (neutral EA). Then, these three mea-sures were combined into a single emotional accuracyindex by matching the emotional accuracy to the strate-gic emotional approach manipulated (i.e., positive dis-play to positive EA, negative display to negative EA, andneutral display to neutral EA). A positive score on anEA index indicated that displayed emotion assessed bythe focal negotiator (as indicated by participants’ self-positive/neutral/negative) was more intense than dis-played emotion as judged by the target negotiator(under-perceived). A negative score indicated that dis-played emotion as judged by the target was more intensethan believed to be displayed by the focal negotiator(over-perceived). A score not signiWcantly diVerent fromzero would indicate a perfect correspondence betweenemotional display and perception.

Dispute resolution. Dispute resolution was measured aswhether or not negotiators came to a mutual agreement,also known as an impasse. The impasse rate was codedas a dichotomous variable where agreement was codedas one (1) and impasse as zero (0).

Expected future relationship. We measured the negotia-tors desire to preserve their future business relationshipby coding the content of the negotiated agreement thatthe two parties completed together at the end of thenegotiation. There were two parts to the agreement, theterms of the deal and an open-ended question. If eitherthe terms of the agreement or the responses to the open-ended question indicated a concern for the futurerelationship, a value of 1 was assigned to the future rela-tionship measure. If neither the terms nor the responsesreferenced a future relationship, a value of 0 wasassigned to the future relationship measure. An inter-rater reliability of 0.96 (Cohen’s �) was established bythe coders who assessed the future relationship concernin the open-ended question.

Results

Manipulation testOur analysis revealed that the positive and negative

emotional display manipulation was successful. To testwhether emotional display engendered emotional conta-gion, we conducted a series of three ANOVAs andplanned comparisons using positive EC, negative EC,and neutral EC as dependent variables and emotional


display (collapsed across matched and single displayconditions) as the independent variables. For positiveEC, the mean was highest for the positive display condi-tion. The overall model was signiWcant (F (2,134)D11.06,p <.001) and the planned comparisons demonstrated thatthe positive condition (MD11.87, SDD 1.85) was signiW-

cantly higher than the neutral (MD11.11, SDD 1.84;t (134)D2.00, p < .05) and negative conditions (MD9.72,SDD2.79; t (134)D4.28, p < .001). Similarly, for negativeEC, the mean was highest for the negative condition.The overall model was signiWcant, F (2, 134) D 18.85,p < .001, and the planned comparison demonstrated thatthe negative condition (M D 8.42, SD D 2.31) was signiW-

cantly higher than the positive (M D 5.87, SD D 1.95;t (134) D ¡5.67, p < .001) and neutral conditions(M D 6.53, SD D 1.86; t (134) D ¡4.26, p < .001) on thenegative EC measure. On the neutral EC measure, theoverall model was signiWcant, F (2,134) D3.97, p < .03 andplanned comparisons showed that the neutral condition(MD11.95, SD D1.28) signiWcantly diVered from the neg-ative condition (MD10.98, SD D2.14; t (134) D2.76,p <.01), but did not diVer signiWcantly from the positivecondition (MD11.65, SDD1.51; t (134)D¡1.04, ns). Thenegligible diVerence between the positive condition andthe neutral condition on the neutral EC measure sug-gests that it may be diYcult to manipulate an absence ofemotion. Individuals in general and negotiators speciW-

cally may more easily discern the valence of an emotion,either positive or negative, than a lack of emotion. It alsosuggests that positive negotiators may be perceived to besomewhat rational, whereas negative negotiators werenot. With respect to the hypotheses, this does not impedeour investigation because the paramount comparisons inthis study were between the positive and negative dis-plays of emotion. The neutral condition served as a base-line condition.

Although the manipulation check above conWrmedthat our manipulation of emotional contagion for thepositive and negative conditions was eVective, there wasno signiWcant diVerence between the single versusmatched display conditions. A dyad with two strategi-cally positive (negative) negotiators was not more posi-tive (negative) than a dyad in which only one negotiatordisplayed positive (negative) emotion. Planned compari-sons indicated that for the positive EC measure, the pos-itive matched display condition (M D 12.17, SD D 1.49)did not signiWcantly diVer from positive single displaycondition (M D 11.57, SD D 2.15; t (131) D 0.60, ns). Simi-larly, for the negative EC measure, the negative matcheddisplay condition (M D 8.74, SD D 2.50) did not signiW-cantly diVer from the negative single display condition(M D 8.05, SD D 2.06; t (131) D 1.11, ns). Also, for theneutral EC measure, the neutral matched condition(M D 11.83, SD D 1.30) did not signiWcantly diVer fromthe neutral single display condition (M D 12.08,SD D 1.26; t(131) D ¡0.58, ns). Consistently, negotiation

outcomes reported below did not signiWcantly diVerbased on single versus matched display, and these condi-tions were collapsed.

To test whether emotional display was accurately per-ceived, an ANOVA was conducted. With regard to emo-tional accuracy (EA), negotiators coached to displaynegative emotion were under-perceived (M D 1.26,SD D 1.28), meaning that they believed their display ofnegative emotion was more intense than judged by theirpartner (the mean was signiWcantly greater than zero,t (43) D 6.53, p < .001). However, negotiators coached todisplay positive emotion were neither over-perceived norunder-perceived (M D 0.00, SD D 0.92). Instead, negotia-tors were able to accurately detect and interpret the pos-itive display of emotion and achieve emotionalcongruence (the mean did not diVer signiWcantly fromzero t (46) D 0.00, ns). The overall ANOVA was signiW-cant (F (2,134) D 18.85, p < .001) and negotiators coachedto display neutral emotion also were neither over- norunder-perceived (M D 0.09, SD D 1.04).

Negotiation outcomesDispute resolution. We predicted there would be a higherimpasse rate in the negative emotional display (ED) con-dition than in the positive ED conditions (H1). Hypothe-sis 1 was not supported. A �2 analysis indicated therewere no diVerences between conditions with respect toimpasse rate. Dyads in the negative ED conditionimpassed in 21% of the cases (35 agreements and 9impasses), positive ED condition dyads impassed in 17%of the cases (39 agreements and 8 impasses), and neithercondition diVered from the neutral ED condition where24% of the dyads reached an impasse (35 agreementsand 11 impasses; �2 (2, N D 137) < 1, ns). These resultssuggest that regardless of the strategic emotional dis-play, negotiators were as likely to reach some form ofagreement and avoid potentially costly alternatives suchas resorting to the decisions of a court.

Expected future relationship. We hypothesized that thepositive ED would lead to a higher regard for preservingthe future business relationship than would the negativeED (H2). This hypothesis was conWrmed. Of the 39 posi-tive ED dyads who reached an agreement, 36 or 92%expected a future relationship; whereas only 71% (25 of35; �2 (1, N D 74) D 5.55 p < .02) of the negative ED dyadsexpected a future relationship. Interestingly, only 74%(26 of 35) of the neutral ED dyads expected a future rela-tionship, which also diVered signiWcantly from the posi-tive condition (�2 (1, N D 74) D 4.41, p < .04).

Discussion

The purpose of our Wrst experiment was to examinehow strategic emotions were displayed and perceived byparties in a negotiation and test their impact on


negotiated outcomes. The Wndings conWrmed ourhypothesis that strategic display of positive emotionwould have an impact on the expected long-term busi-ness relationship. Negotiators who displayed positiveemotion were more likely to anticipate a long-term rela-tionship than negotiators who displayed negative orneutral emotions, as indicated by the inclusion of futurejoint business endeavors in the negotiated contract. ThisWnding extends prior Wndings by Allred et al. (1997) whofound that negotiators who felt high anger and low com-passion had less desire to work with each other in thefuture. It suggests that in a dispute situation, even if youfeel angry there could be beneWts to displaying positiveemotion. Of course, surface acting (displaying emotionsincongruent with experienced emotions) may be diYcult,and furthermore may have ethical implications; how-ever, this experiment suggests that display of positiveemotion is beneWcial in a dispute setting. Instead of dis-solving a potentially fruitful business relationship, itfacilitated the potential of future value creation. A posi-tive approach did not, however, aVect the rate ofimpasse. In a dispute, the primary goal of negotiators isto avoid impasse because an impasse can be detrimentalto the outcome of both parties. Contrary to our hypoth-esis, negotiators strategically displaying negative emo-tion were not more likely to reach an impasse and, thus,were as successful as the positive and neutral negotiatorsin avoiding costly alternatives such as declaring bank-ruptcy or taking the battle to court.

An important contribution to the negotiation litera-ture actually may be found in the design of this study.SpeciWcally, strategic emotions were successfully dis-played by novice negotiators in a convincing fashion.The manipulation check conWrmed that negotiatorseVectively displayed positive and negative emotions dur-ing the negotiation simulation. Thus, deliberate displayof emotion is a feasible negotiation strategy.

In fact, it was suYcient for only one party to initiallydisplay positive or negative emotions during the negotia-tions. Accordingly, it may be adequate in future researchto manipulate the behavioral display of emotion of onlyone negotiator to impact the emotional experience of thedyad. These methodological implications make theexperimental Wndings more robust and the managerialimplications more straightforward. Even if only onenegotiator adopts a positive strategic emotionalapproach, it may be quite eVective in increasing the like-lihood of a proWtable future business relationship devel-oping once the currently disputed issue is resolved.Therefore, managers should consider focusing on theirown emotional display during negotiations because theiremotional disposition may inXuence the opposing partyand subsequently shape the nature of the negotiatedagreement.

Our Wndings about emotional accuracy, however, sug-gest that whereas strategic display of positive emotion was

accurately perceived, negotiators were overconWdent tothe degree in which they believed others accurately per-ceive the strategic display of negative emotion. Althoughthe negative display of emotion engendered more negativeemotion (as indicated by the emotional contagion manip-ulation check), the other party did not judge the focalnegotiator to be as negative as she believed she was (emo-tional accuracy was incongruent). The emotional accuracyWnding is consistent with research that showed that expe-rienced negative emotions were related to expression onlyfor people with dispositionally high expressivity (Gross,John, & Richards, 2000) and suggests that negative emo-tion may be relatively diYcult to feign. Thus, managerswho are low on emotional expressivity might need to betrained to convincingly display negative emotion in amanner that is consciously perceived by the other party.The ethical implications of training managers to displayemotion are perplexing, and seem more challenging wherenegative, rather than positive emotions are concerned.Ethical considerations would apply to both organizationsthat engage in regulation of emotional display of employ-ees and customers (Rafaeli & Sutton, 1991; Sutton &Rafaeli, 1988; Van Maanen & Kunda, 1989), as well as toindividual negotiators.

This experiment had several limitations. First, thiswas a negotiation simulation and it is not evident thatstrategically displaying emotions in the real world woulddirectly mirror doing so in a relatively risk-free class-room environment. Furthermore, the simulated negotia-tion was a qualitative dispute resolution case andoutcomes may not transfer to deal-making buyer–sellertransactions, whether distributive or integrative innature. Finally, other than being generally perceived aspositive and negative in contrast to a control condition,we had no real-time process measures of emotional dis-play, only assessments following the negotiation. Thus,emotional display may not have been consistent acrossdyads and may not have been consistent within dyadsthroughout the one-hour-long negotiation. For example,negotiators in the negative emotional display conditionmay have displayed negative emotion at the outset of thenegotiation, but to avoid impasse negotiators may havereverted to positive emotional display toward the end.To address the latter limitations, our next two experi-ments turned to a controlled display of emotion whereall participants in each condition viewed the same emo-tional display, which was consistently displayedthroughout the social interaction. We also shifted to anegotiation setting where we could exclusively focus oneconomic outcomes.

Experiment 2

In our next experiment, we investigated whether thepotential beneWts of strategic display of emotions


observed in Experiment 1 would carry over to a diVerentnegotiation context, a zero-sum setting. SpeciWcally, weinvestigated whether the strategic display of positive ornegative emotion would be the most eVective in an ultima-tum bargaining setting when no future business relation-ship was at stake and the parties could simply walk away.

Research on emotion in ultimatum bargaining sug-gests that the display of positive emotion may be moreeVective than negative emotion. Demanding statementssuch as “Take it or leave it!” increase rejection rates(Kravitz & Gunto, 1992; Leventhal & Bergman, 1969).In addition, Pillutla and Murnighan (1996) argued that,“Respondents who perceive ultimatum oVers as unfairmay also feel anger and, if they do, they are likely to actspitefully and reject an economically valuable oVer” (p.208). In contrast, research on persuasion tactics suggeststhat ingratiation and positive behaviors may help attaindesirable outcomes by getting the target in a good moodand by making the person think favorably about them-selves (e.g., Falbe & Yukl, 1992; Kipnis, Schmidt, & Wil-kinson, 1980; Yukl & Falbe, 1990). In our context, thiswould suggest that people may be more likely to acceptthe terms of an oVer when it is presented with a friendlydisplay of positive emotion rather than an insisting neg-ative display of emotion.

Hypothesis 1. Negotiators will be less likely to accept anoVer when the focal negotiator displays negative emo-tion and more likely to accept an oVer when the focalnegotiator displays positive emotion.

We also predict that a negative display of emotionwill adversely inXuence the maximum dollar amountthat a responding negotiator would be willing to pay.Willingness to pay is based on one’s reservation price,which represents a negotiator’s minimum or maximumacceptable settlement price within a current negotiationtaking into account alternatives and transaction costs(RaiVa, 1982). For respondents to an ultimatum, it rep-resents the most the target is willing to pay and, giventhe absence of a clear alternative, could be inXuenced bythe strategic emotional approach of the focal negotiator.According to Pillutla and Murnighan (1996), when afocal negotiator presents an oVer and strategically dis-plays negative emotion, the responding target negotiatormay react spitefully and reduce the amount that they arewilling to pay to make a deal. Thus, we make the follow-ing hypothesis.

Hypothesis 2. Negotiators who respond to an oVerextended by a focal negotiator displaying negative emo-tion will be less willing to pay a demanded settlementprice than negotiators who respond to an oVer extendedby a focal negotiator who displays positive emotion.

We further predict that willingness to pay will medi-ate the inXuence of strategic emotional approach onacceptance or rejection of the oVer. The target’s willing-

ness to pay may help to explain the relationship betweenstrategic display of emotion and outcomes.

Hypothesis 3. A negotiator’s willingness to pay willmediate the eVect of the emotional strategy on accep-tance rates such that a negotiator facing a focal negotia-tor displaying negative emotion will be willing to payless and therefore will be less likely to accept the oVer.

Methods

ParticipantsA total of 282 executive MBA students participated

as a part of a negotiation course. Individual participantswere assigned randomly to one of the three conditions:positive (N D 95), negative (N D 96), or neutral (N D 91)emotional display. One participant in the negative condi-tion did not complete the questionnaire and was subse-quently dropped from the analysis.

Task and procedureThe task used for this study involved an ultimatum

bargaining situation, in which the focal party (proposer)presented the recipient with a take-it-or-leave-it oVer,which the recipient (target) had to either accept or reject.Ultimatums diVer from negotiations in that the recipientcannot make a counteroVer (for a review on ultimatumbargaining see Roth, 1995). Because all target negotia-tors received an objectively equivalent oVer that onlydiVered in the strategic emotional approach, diVerencesin outcomes could be attributed to the strategic emo-tional display in this bargaining situation.

Furthermore, we created an ambiguous BATNA (bestalternative to the negotiated agreement, Fisher, Ury, &Patton, 1991) to introduce a degree of uncertainty in themind of the target so that the decision whether to acceptor reject the oVer could be swayed by the emotional dis-play of the oVer. On one hand, because there was no deW-

nite BATNA, the target negotiators had to seriouslyconsider any deal proposed by the focal negotiator. Onthe other hand, the presence of an ambiguous BATNAprovided some latitude for rejection of the oVer that couldbe attributed to the strategic emotional approach. Thus,based on the emotional display of the proposer, the recipi-ent negotiator could interpret the likelihood and attrac-tiveness of the BATNA diVerently, which couldsubsequently inXuence his/her Wnal decision.

Finally, we examined a wedding context that by itsnature would lend itself to emotion. Within the context ofan inherently exciting wedding arrangement, our choice ofa service provider also was intentional. We assumed thatan industry where customer satisfaction is vital would fur-ther make emotions salient (e.g., Oliver, 1997).

Participants were provided a half-page of writtenbackground information concerning the catering serviceunder consideration for their upcoming wedding. The


negotiation concerned the Wnancial arrangements for thecatering company. Participants were told to assume therole of the person getting married and that they wouldsoon “meet” with the business manager of the cateringcompany they had considered for their wedding. Thebusiness manager of the catering company had giventhem a good faith estimate of $14,000 several monthsago to provide catering service for their wedding recep-tion. The background information also mentioned analternative catering company that would be available ontheir wedding date. However, participants were told thatthey did not have direct contact with that company, nordid they fully trust the person who recommended it.Finally, participants were told that their goal was toWnalize the catering arrangements with the businessmanager at the upcoming meeting.

When participants Wnished reading the backgroundinformation, the experimenter told participants that theywere going to meet with the manager of the catering com-pany and then played a 2-min videotape. Participants wereseated at business desks to watch the Wlm, which wasshown on an 8�£8� screen. Participants viewed one ofthree versions of the oVer corresponding to the strategicemotional display by a professional actor playing the roleof the business manager. In the videotape, the businessmanager explained that the price of the reception increasedfrom $14,000 to $16,995 due to market price Xuctuations.The business manager ended the meeting by stating thatanother couple was interested in the same date, indicatingthat if the participant did not sign the contract immediatelythis option would no longer be available.

There were three versions of the videotape, each repre-senting one of the three strategic emotional approaches.The content of the message was the same in each condi-tion; however, the facial expressions, tone of voice, anddemeanor diVered in each condition (cf., Thompson &Kim, 2000). In the positive emotional display condition,the business manager in the videotape spoke with afriendly tone, smiled often, nodded her head in agreement,and appeared cordial and inviting. In the negative emo-tional display condition, the business manager spokeantagonistically, appeared intimidating, and was insistent.In the neutral emotional display condition, which servedas a benchmark condition, the business manager used aneven and monotonic voice, displayed little emotion, andspoke in a pragmatic manner. Otherwise, the content ofthe conditions, terms, and precipitating events were identi-cal across the three versions.

After viewing the videotape, participants were given aform that looked like the actual business contract in theWlm, which asked them to either accept and sign the pro-posed invoice at $16,995 or to reject the invoice. Onceparticipants made their decision to accept or reject theoVer made by the business manager they were asked tocomplete a “Customer Satisfaction” survey that servedas a manipulation check. Following this, participants

were asked the following: (1) Did you accept the oVerand sign the attached invoice (Yes/No)? and (2) What isthe most you would pay before walking away?

MeasuresManipulation check. The customer satisfaction surveywas used as a manipulation check for the displayed emo-tions of the videotaped business manager. The manipu-lation check measure included ratings of how friendly,insistent, and rational the manager was during the meet-ing. Each item was rated on a 7-point Likert scale,anchored by 1 (not at all) and 7 (very much).

Outcome measures. The acceptance or rejection of the oVerwas a dichotomous variable. The most the target negotiatorwould be willing to pay was a continuous variable.

Mediation analysisWe used the bootstrap framework and mediation

procedures recommended by Shrout and Bolger (2002)to test the predicted mediation in Hypothesis 3. Boot-strapping is a non-parametric approach and involves“re-sampling” the data many times to generate anempirical estimate of the sampling distribution of a sta-tistic. The bootstrap works well when it is likely that asample is not normally distributed such as the skeweddistribution that occurs in the distribution of indirecteVect estimates that is used in mediation testing (MacK-innon, Lockwood, & HoVman, 2002).

According to Shrout and Bolger (2002), the bootstrapdistribution of the indirect eVect, a £ b (where a is thepath from X, a predictor variable, to M, a mediator andb is the path from M to Y, an outcome variable) is con-structed in the following way. First, using the originaldata set, a bootstrap sample of N persons is created byrandomly sampling observations with replacement fromthe original data set. The bootstrap sample will have thesame size N as the original sample. Second, the productof the estimates a and b of the bootstrap sample is calcu-lated and saved to a Wle. Then, steps 1 and 2 are repeatedfor a total of J times, where J is a large number, usuallyat least 1000 to estimate conWdence intervals (Efron &Tibshirani, 1986). Next, a percentile conWdence intervalfor the indirect eVect in mediation is computed using thefollowing formula, (a £ b) § sab z�/2., where sab is theapproximate standard error for the indirect eVect a £ band z�/2 is equal to the constant 1.96. When the conW-dence interval does not include zero mediation hasoccurred. We used PLS (Chin, 2001), a statistical soft-ware package, to generate the bootstrap samples.1

1 In addition to the bootstrapping technique, we also used the inXu-ential mediation procedure originally proposed by Baron and Kenny(1986) and later updated by Kenny, Kashy, and Bolger (1998) to testthe predicted mediation. The results were consistent with that of thebootstrapping technique.


Results

Manipulation checkThe manipulation check revealed the emotional display

manipulations were eVective. Participants were more likelyto evaluate the manager in the videotape as friendly in thepositive (MD4.88, SDD1.76) than in the neutral(MD2.38, SDD1.33) and negative emotional display con-ditions (MD1.28, SDD0.59; F(2,270)D182.65, p<.001).Planned comparisons showed that the positive conditiondiVered signiWcantly from both the neutral (t(270)D10.71,p<.001) and the negative condition (t(270)D7.12,p<.001). Participants were more likely to evaluate themanager as insistent in the negative (MD5.49, SDD2.04)than in the positive (MD4.87, SDD1.63) and neutral emo-tional display conditions (MD4.86, SDD1.83;F(2,270)D3.58, p<.05). Planned comparisons showed thatthe negative condition signiWcantly diVered from both thepositive (t(270)D¡2.30, p<.03) and the neutral(t(270)D¡2.20, p<.03) conditions. Planned comparisonsalso revealed that participants were more likely to evaluatethe manager as rational in the neutral emotional displaycondition (MD4.03, SDD1.67) than in the negative condi-tion (MD2.85, SDD1.90; t(268)D4.47, p<.001) but, justas in Experiment 1, neutral emotional display did not sig-niWcantly diVer from the positive emotional display(MD4.04, SDD1.59; t(268)<1, ns). The overall ANOVAthat compared the evaluation of the neutral manager in thethree conditions was signiWcant (F(2,268)D14.65, p<.001).The results of the positive emotional display manipula-tion check and the negative emotional display manipula-tion check suggest that participants did not consider theneutral business manager to be overly positive or overlynegative. Accordingly, we considered the neutral emo-tional display condition to be a baseline or benchmarkcondition.

Reactions to the ultimatumThe dependent variable was the target’s decision to

either accept or reject the oVer made by the businessmanager. Hypothesis 1 predicted the eVects of strategicemotional display on outcomes, speciWcally whether ornot participants accepted the ultimatum proposed by thebusiness manager. In support of our hypothesis, therewas a signiWcant eVect of emotional display on accep-tance of the oVer, such that participants were signiW-cantly less likely to accept the proposal oVered by thebusiness manager displaying negative emotion(M D 26%), than were participants who received thesame deal oVered by the business manager displayingpositive emotion (M D 53%) or the neutral businessmanager (M D 56%; �2 (2, N D 281) D 20.13, p < .001).The positive and neutral conditions did not signiWcantlydiVer from each other (�2 (1, N D 186) D 0.28, ns). SeeTable 1 for frequencies of acceptance rates byconditions.

Hypothesis 2 tested the inXuence of emotional condi-tion on willingness to pay and was also conWrmed. Theoverall ANOVA was signiWcant (F (2, 249) D 11.16,p < .001) and planned comparisons showed that whenasked, “What would be the most you are willing to pay,”participants who viewed the business manager display-ing negative emotion were willing to pay substantiallyless (M D $14,618, SD D $3618) than did those partici-pants who viewed a positive (M D $16,136, SD D $2754;t (249) D 3.47, p < .001) or neutral (M D $16,624,SD D $1924; t (249) D 4.49, p < .001) display of emotion.

Hypothesis 3 argued that the target’s willingness topay would mediate the eVect of strategic emotionalapproach on the likelihood of acceptance and was con-Wrmed. The mean for the bootstrap estimate which mea-sured the path from emotional condition to target’swillingness to pay times the path from target’s willing-ness to pay to acceptance of oVer was 0.15 with anapproximate standard error of 0.04. The conWdenceinterval (0.07, 0.23) excluded zero, and thus we con-cluded that the target’s willingness to pay mediated therelationship between strategic emotional display andacceptance of the oVer.

Discussion

We predicted that negotiators who strategically dis-played negative emotion would be less likely to achievetheir objectives in a take-it-or-leave-it ultimatum bar-gaining setting than those who displayed positive emo-tion. In support of our hypothesis, the business managerwas less eVective when displaying negative emotion thaneither positive or neutral emotions. Negotiatorsresponding to the positive display of emotion were twiceas likely to sign a deal than negotiators responding tothe negative emotional display. Thus, even though theiralternatives were identical, target negotiators chosediVerent courses of action as a result of the strategicemotional approach of the focal negotiator. In general,these results were consistent with the Wndings of Experi-ment 1, since both experiments indicated that the strate-gic display of positive emotion may be more eVectivethan negative emotion with regard to retaining andembarking on a future business relationship.

Our process measure provided some insight about thepsychological mechanism that underpins this eVect. Theresults showed that negotiators who dealt with a

Table 1Frequencies of acceptance versus rejection by strategic emotional dis-play (Experiment 2)

Note. Subscripts indicate signiWcant diVerences between groups.

Outcome Positive Neutral Negative Total

Accept 50a 51a 25b 126Reject 45a 40a 70b 155Total 95 91 95 281


business manager strategically displaying negative emo-tion were less willing to pay an increased settlement pricethan were the negotiators who faced a positive or neutralemotional display. It suggests that negotiators who dealtwith the negative manager may have had an uncomfort-able and somewhat perverse experience that caused themto devalue the oVer made by the manager. Though mostnegotiators rejected the oVer and turned to an uncertainBATNA, a small, but signiWcantly greater than zero,proportion of negotiators were willing to pay the pre-mium charged by the negative business manager andaccepted the oVer. Future research could examinewhether individual diVerences in preferences toward riskdiVerentiate between negotiators who succumbed todemands of a negative negotiator and those who did not.

This experiment had several limitations. First,although the use of a professional actor displaying emo-tions in a videotaped manipulation oVered more experi-mental control, this methodology created a unilateraldisplay of emotion that introduced new boundary condi-tions. Participants were constrained in their ability toreciprocally inXuence the target negotiator through theirown emotional display. Given the ultimatum setting,they were also constrained in the role of responder andtheir range of responses was limited to a dichotomousdecision of accept or reject. Finally, the manipulationcheck was restricted to a single-item response which less-ens the reliability of the manipulation. To address theselimitations, the next experiment used the same positive,negative, and neutral emotional display manipulationvideos, and tested these manipulations with 3-item or 4-item composite measures. Furthermore, the role of par-ticipants was changed from responder to demander.

Experiment 3

In Experiment 3, we examined the eVect of positive,negative, and neutral display of emotions in a distribu-tive setting. Similar to Experiment 2, negotiators faced abusiness manager strategically displaying emotions;however, in this experiment the participants played therole of “demander” rather than “responder.” Thischange allowed participants to actively engage in thenegotiation process by making a counteroVer thatreXected their own demands, in contrast to simplyaccepting or rejecting an ultimatum oVer. It also pro-vided a test of the potential advantages of displayed neg-ative, in contrast to positive, emotion.

Although a negative and demanding negotiator mayseem unreasonable, risky, reckless, or seemingly out ofcontrol, a negative emotional display can be highly judi-cious and calculating (Thompson et al., 2001). InSchelling’s (1960) discussion of political deterrence, hedescribed how threats can be viewed as either “hot-headed or cool-headed activities” (p. 16). He also dis-

cussed how “administering a punishment until the otheracts, rather than if he acts” (p. 196) can be eVective. In abusiness transaction, the implication is that a crediblethreat of “no sale” (turning to one’s BATNA) can com-pel the other party to concede to a less favorable price,even if an outcome of “no sale” has a chance of hurtingboth sides. Frank (1988) similarly pointed out that athreat that is coupled with an irrational appearance canbe an eVective sales technique. He noted that, “It is theexpression on his face that really makes his [the sales-men’s] point” (p. 55) and leads people to buy somethingthey had no intent of purchasing. Furthermore, aggres-sive threats may lead to concessions if the other partyfears that a “barking dog might bite”; or as Frank putsit: “a rival with a deep croak ‘intends’ to Wght if chal-lenged.” (p. 98)

Thus, negative emotion coupled with clear demandscan lead to a competitive advantage in negotiations, aslong as the commitment or threat made during an emo-tional outburst is credible. Since the initial oVer made bythe focal negotiator in the negotiation scenario simu-lated in this study ended with a credible threat to turn tothe next customer in line, we suggest that when coupledwith a negative display of emotion, the demand of ahigher price (albeit due to external market price Xuctua-tions) will be more eVective. Therefore, it is possible thatwhen facing a business manager displaying negativeemotion along with a demand in price, negotiators willbe more likely to concede. On the other hand, resultsfrom Experiment 2 suggest that the display of negativeemotion may be less eVective than that of positive emo-tion when attempting to secure a deal presented as atake-it-or-leave-it oVer. Experiment 2 thus suggests thatnegotiators will be more likely to yield to the requestsmade by a demanding positive opponent than to therequests made by a demanding negative opponent. Thequestion boils down to whether “you can catch moreXies with honey” or whether “the squeaky wheel gets thegrease,” which leads to the following competing hypoth-eses.

Hypothesis 1a. Target negotiators will be more likely toconcede to focal negotiators who display negative emo-tion than those who display positive emotion.

Hypothesis 1b. Target negotiators will be more likely toconcede to focal negotiators who display positive emo-tion than those who display negative emotion.

To better understand the impact of strategic emotionson negotiation behavior, we also assessed the degree towhich the negotiators were concerned about the negoti-ated outcome. Experiment 1 results revealed that wheninteracting with a negative opponent, negotiators mayhave devalued future outcomes. This suggests that whenfacing a negative opponent, negotiators may not beoverly concerned about whether their own demands


(i.e., counteroVer) is accepted or rejected. Thus, if negoti-ators facing a business manager displaying negativeemotion are either insulted by the negative tone, or if thenegative emotions are transferred to the other party viaemotional contagion (HatWeld et al., 1992), they maybecome insensitive to whether the other party actuallyaccepts or rejects their counteroVer.

Hypothesis 2. Target negotiators will be less concernedabout the outcome when facing a focal negotiator dis-playing negative emotion in contrast to positive or neu-tral emotion.

Methods

The negotiation task used in Experiment 3 was thewedding scenario described in Experiment 2. The proce-dure diVered on three dimensions: (a) the communica-tion medium, (b) a change in the participant’s role, and(c) the dependent variable. First, instead of being seatedat business desks and completing the negotiation taskusing paper and pen, participants in this study com-pleted the negotiation task via the Internet. An onlinesurvey was created that included instructions, back-ground information, the video vignette (a business man-ager conveying the same information displayingstrategically positive, negative, or neutral emotion), anda post-vignette questionnaire. The online survey ran-domly assigned one of the three versions to each of theparticipants. Second, the role of the participant changedin this negotiation from the role of a “responder” inExperiment 2 to the role of “demander.” By studydesign, participants in Experiment 2 could only respondto the actions of the business manager and either acceptor reject the ultimatum oVer presented to them. Experi-ment 3 allowed the participant to actively partake in thenegotiation process by making a counteroVer. By givingparticipants the role of demander and the ability tomake a counteroVer, it gave the business manager powerto reject the counteroVer and hurt both sides (no deal) oraccept the counteroVer and come to a mutually agree-able outcome. Participants were told that the businessmanager would either accept or reject the counteroVer;however, acceptance or rejection by the business man-ager was randomly assigned to the participants by afunction in the online survey. Participants were appro-priately debriefed. The third parameter that diVered wasthe dependent variable (counteroVer).

ParticipantsParticipants were MBA students enrolled in a negoti-

ation course at a university located in the midwesternUS. The 82 participants were randomly assigned by theonline program (which accounts for the unbalanced cellsizes) to one of the three conditions: (1) positive(N D 29), (2) negative (N D 30), (3) or neutral (N D 23).

MeasuresManipulation check. To ensure that the strategic emo-tional approach of the business manager depicted in thevideo was successful, we created composite measures toassess emotional display. Four items measured positiveemotional display, including friendly, cooperative, con-structive, and empathic (Cronbach’s �D 0.86). Fouritems measured the display of negative emotion, includ-ing emotional, angry, aggressive, and vindictive (Cron-bach’s �D 0.80). Three items measured neutralemotional display, including rational, analytical, andlogical (Cronbach’s �D 0.71). Items were measured on a5-point Likert-type scale anchored by 1 (stronglydisagree) and 5 (strongly agree). Thus, a high score onthe composite measures indicated that the participantsperceived the business manager to display the respectiveemotional strategy.

Outcome concern. After the participants made theircounteroVer, but before they were informed that theircounteroVer was rejected or accepted, participantsanswered the question, “How disappointed would yoube if your counteroVer was rejected?” Participantsresponded using a Likert-type scale anchored by 1 (notat all) to 7 (very disappointed). A high value indicated ahigh concern for the outcome.

CounteroVer. The dependent variable was the counter-oVer made to the business manager by the participants.The respondents’ counteroVer amount was coded as 0 ifthe amount was less than or equal to $14,000 and 1 if theamount was greater than $14,000. When run as a contin-uous variable, the same pattern of results emerged. In thevideotape, the business manager explained that the priceof the reception increased from $14,000 to $16,995 dueto market Xuctuation. Thus, a counteroVer of $14,000 orless would indicate that the respondent was not willingto yield and not willing to make concessions. A counter-oVer of more than $14,000 would indicate that therespondent was more willing to negotiate how to splitthe increase of $2,995 between themselves and the busi-ness manager. It was coded as a dichotomous variable tocapture whether or not a concession was made.

Results

Manipulation checkThe manipulation check conWrmed that both the pos-

itive and negative display of emotions were eVective.Participants who were assigned to the positive emotionaldisplay condition (M D 3.22, SD D 0.68) evaluated thebusiness manager to be more positive than did the par-ticipants in the neutral (M D 2.30, SD D 0.76) or the neg-ative conditions (M D 1.61, SD D 0.60; F (2,79) D 41.61,p < .001). Planned comparisons demonstrated that thepositive condition responses diVered signiWcantly from


the neutral condition (t (79) D 4.81, p < .001) and the neg-ative condition (t (79) D 9.58, p < .001). The negativeemotional display manipulation also was successful. Par-ticipants who viewed the business manager displayingnegative emotion (M D 3.39, SD D 0.72) evaluated thebusiness manager to be more negative than did partici-pants who viewed the neutral (M D 2.15, SD D 0.60) orpositive business manager (M D 2.41, SD D 0.61;F (2, 79) D 28.47, p < .001). Planned comparisons demon-strated that the negative condition diVered signiWcantlyfrom the neutral condition (t (79) D ¡6.92, p < .001) andthe positive condition (t (79) D ¡5.87, p < .001). Consis-tent with Experiment 2, planned comparisons showedthat participants who viewed the neutral display of emo-tion perceived the business manager to be as rational,logical, and analytical (M D 3.07, SD D 0.84) as those inthe positive condition (M D 3.05, SD D 0.84; t (48) < 1, ns)and also did not diVer from the negative condition(M D 2.78, SD D 0.85; t (48) D 1.26, ns). The overallANOVA was not signiWcant (F (2, 79) D 1.08, ns). How-ever, given that the manipulation check for the strategicpositive and negative emotional display indicated thatparticipants did not consider the neutral business man-ager to be overly positive or negative, the neutral condi-tion serves as an appropriate baseline condition.

CounteroVers. Hypothesis 1 presented competinghypotheses. Hypothesis 1a predicted that negotiatorswould be more likely to concede to negotiators who dis-played negative emotion. Hypothesis 1b predicted theywould be more likely to concede to a focal negotiatorwho displayed positive emotion. Hypothesis 1b was sup-ported. Only 13% of the participants in the positive emo-tional display condition made counteroVers that wereequal to $14,000 or less; whereas 52% of the participantsin the negative emotional display condition made coun-teroVers equal to $14,000 or less (�2 (1, N D 58) D 9.47,p < .01). In the neutral condition, 22% of the negotiatorsmade counteroVers equal to $14,000 or less which signiW-cantly diVered from the negative emotional display con-dition (�2 (1, N D 52) D 4.87, p < .03). See the frequenciesfor the counteroVers in Table 2.

Outcome concern. Hypothesis 2 predicted that in thenegative emotional display condition negotiators wouldhave lower outcome concern than in the positive emo-tional display condition. Hypothesis 2 was conWrmed.

Table 2Frequencies of counteroVers by strategic emotional display condition(Experiment 3)

Outcome Positive Neutral Negative Total

CounteroVer < or D $14,000

4 (14%) 5 (22%) 15 (52%) 24 (30%)

CounteroVer > $14,000 25 (86%) 18 (78%) 14 (48%) 57 (70%)Total 29 23 29 81

Participants who faced a business manager displayingnegative emotion (M D 3.03, SD D 1.50) were less likelyto be concerned about the outcome than were participantswho faced a business manager displaying positive emotion(MD4.11, SD D1.32; t (78)D2.97, p < .01) and the overallANOVA was signiWcant (F(2,78)D4.43, p < .02). The neu-tral emotional display condition (MD3.74, SDD1.37) didnot diVer signiWcantly from the negative (t (78)D1.82,p< .08) or the positive emotional display conditions(t (78)< 1, ns).

Discussion

The main Wndings of Experiment 3 indicated thatwhen negotiators were given the opportunity to makedemands, they made more extreme demands and wereless concerned about whether these demands would berejected, when facing a negotiator strategically display-ing negative rather than positive emotion. That is, ratherthan conceding to the demands of the business managerdisplaying negative emotion, negotiators mirrored thisstrategy by making extreme counteroVers. CounteroVerswere not only lower in the negative than in the positiveemotional display condition, they were also more likelyto be below the initial quote suggesting not only that theoVer was devalued, but negotiators made a low ballcounteroVer that may have symbolized an act of spite.Rather than creating a strategic advantage by gainingconcessions from the other party, the strategic display ofnegative emotion by a skilled negotiator (a professionalactor) led to what could be interpreted as the foundationof a conXict spiral. Of course, this experiment has somelimitations. Similar to Experiment 2, the social relation-ship is artiWcially constrained by the use of a videotapedmanipulation. Furthermore, both experiments link emo-tions to a professional setting—the service industry—that may prime and therefore favor positive emotionaldisplay. It may be that negative emotional display isadvantageous when the focal negotiator’s role is that ofa bill collector or a criminal investigator (Rafaeli &Sutton, 1991; Sutton & Rafaeli, 1988), where negativeemotions are expected and considered appropriate.Finally, there may be a cultural bias, given that all par-ticipants in this experiment were from the US where dis-play of negative emotion is considered unprofessional.In other cultures, for example Israel, display of negativeemotion may be normative and actually expected in awider array of professional settings (Rafaeli,Fiegenbaum, Maw-Der, & Hwee-Hoon, 2005).

General discussion

This study examined the eVect of strategic display ofemotions on both economic and relational outcomes. Allthree experiments highlighted the advantage of


behaviorally displaying positive in contrast to negativeemotions. In a dispute setting, negotiators strategicallydisplaying positive emotion were more likely to reach aninterest-based agreement that included parameters for acontinued future business relationship. In an ultimatumsetting, negotiators who strategically displayed positiveemotion were more likely to persuade their opponents toaccept their oVers and close a deal. In a distributivenegotiation setting, negotiators strategically displayingpositive emotion were better able to extract concessionsfrom the opposing party.

We acknowledge that the ineVectiveness of strategicdisplay of negative emotion in the current study may haveoccurred because of the following limitations of the exper-imental design: (a) the particular business contexts chosenfor the negotiation tasks (e.g., small scale business disputein Experiment 1 and the service industry and role inExperiments 2 and 3), (b) the type of negotiation (e.g., dis-pute versus deal-making and ultimatum setting), and (c)the manner in which emotions were displayed, includingthe lack of control for whether the displayed emotionswere interpreted as genuine or feigned.

First, the context of the negotiations in these studiesmay pose a threat to the external validity of the Wndingsbecause of the link between the type of service providedand the behavior that was modeled by the emotionaldisplay. In a dispute setting, such as in Experiment 1,negotiators may tend to respond favorably to positiveemotion because when settling a one-on-one interper-sonal dispute between business partners in a relativelyinformal setting, positive emotions are likely expected.If, however, the dispute occurred between multiple cor-porate groups in a more formal setting, strategic displayof negative emotion may have led to more favorableoutcomes. Furthermore, in the service industry (Experi-ments 2 and 3) and speciWcally with regard to a weddingcaterer, negotiators may be less likely to sign a deal witha service provider who displays negative emotion ascompared to industries where there is less emphasis onsocial interaction. In addition, had the service seekerdisplayed negative emotion as opposed to the serviceprovider, a diVerent pattern of results may haveemerged. We note that a signiWcant number of the par-ticipants in Experiment 2 did accept the oVer and thuswere willing to work with a catering manager who dis-played negative emotion. It may be that they diVerenti-ated between the role of the manager who coordinatedthe event and the hired staV who they expected woulddisplay appropriate positive emotion as they served thewedding guests. Alternatively, there may be culturaldiVerences in what is considered appropriate emotionaldisplay. New Yorkers may not be intimidated and mayactually prefer authentic display of negative emotionthat could, for example, signal eYciency and eVectivefollow-through; whereas in the Midwest it may be inter-preted as rude and unprofessional.

Second, we chose to evaluate our hypotheses in inter-active dispute resolution and ultimatum bargaining set-tings. However, deliberate emotional display also shouldbe tested in integrative settings. When focusing on inter-est based and integrative tactics a display of positiveemotion may strengthen the impact of a negotiator’sstrategy. At key moments during a mixed-motive negoti-ation, however, display of negative emotion may proveeVective if judiciously implemented. Third, although ourmanipulation check was conWrmed for both positive andnegative emotional display, the eVect of emotion maydiVer when it is judged to be genuine versus feigned andfuture research should control for perceived authenticity.

Despite these limitations, in addition to highlightingthe eVectiveness of positive emotional display on negoti-ated outcomes, this study also provided some insightsinto the inXuence of emotional display on social interac-tions in negotiations. An important contribution to thenegotiation literature is the actual feasibility of strategi-cally displaying emotions. The manipulation check inExperiment 1 conWrmed that novice negotiators eVec-tively displayed emotions that led to the emergence ofpositive or negative emotional contagion. Interestingly,the degree to which displayed emotions were accuratelyjudged by the other party was contingent on the type ofemotion. When negotiators displayed positive emotion,the opposing party accurately judged the emotional dis-play as positive—that is, there was congruence betweenwhat the focal negotiator believed she displayed andwhat the target negotiator judged. However, when nego-tiators displayed negative emotion the negotiatorbelieved she was more negative than judged by theopposing party. This pattern suggests that when negotia-tors choose to strategically display negative emotionthey may be likely to overestimate the extent to whichtheir displayed emotion is detectable by others. Alterna-tively, they may have been overconWdent in their abilityto display negative emotion, perhaps experiencing moreemotion than they actually displayed. Similar to situa-tions where negotiators are egocentric with respect totheir cognitive or behavioral abilities (Babcock,Lowenstein, IssacharoV, & Camerer, 1992), our Wndingssuggest that negotiators also may over-interpret theiremotional experiences in a way that is self-serving. Wecall this emocentrism. The concept of an emocentricnegotiator, who has a biased appraisal of his or her stra-tegic emotional abilities, merits further investigation.

Another important direction for future research is todeconstruct the more general categories of positive andnegative emotional display and test the impact of dis-crete positive or negative emotions on both negotiationprocess and outcomes. SpeciWc emotions that can bedetected in facial expressions such as happiness,surprise, anger, sadness, fear, and disgust/contempt(Ekman, 1972) might be good starting points. Further,the boundary conditions and situational factors that


moderate their eVectiveness promise fertile ground forfuture research on display of emotions in interactive set-tings. Recent work by Sinaceur and Tiedens (2005) hastaken the Wrst step in this endeavor by examining theeVect of strategic anger display in mixed-motive interac-tive negotiations. Whether discrete emotions are judgedas authentic or contrived may inXuence their impact oneconomic and relational outcomes.

The nexus between the display of positive and nega-tive emotion is also an interesting direction for futureresearch. In our study, we did not pit diVerent emotionalstrategies against each other in the same dyad. Contrast-ing emotional strategies, for example, a happy negotia-tor who faces an angry negotiator, may have a surprisingimpact on the negotiation process (does contagion ofone emotion outweigh the other?) and negotiation out-comes. Thus, the inXuence of emotional display may becontingent on the interaction between diVerent types ofemotions. Also, convincing strategic emotions may beshrewdly and advantageously aligned with other negoti-ation strategies such that emotions may moderate theeVect of invoking interest, right, or power based persua-sion tactics. Furthermore, future research will need toaddress questions such as: What are the boundary con-ditions for displaying emotions? Can negotiators as eas-ily display chronic manifestation of one emotion andalternate between contrasting emotional displays? Arethere short-term or long-term social costs to feigningemotions and do ethical concerns outweigh the potentialbeneWts? A recent theoretical paper proposes a socialinteraction model of how emotion regulation mayimpact work strain (Coté, 2005), but the ethical consid-erations have yet to be addressed.

Virtually any manager or executive will vehementlyacknowledge that emotion and mood are central ineven the most routine negotiation situations. However,the Weld of negotiation has, for over two decades,focused nearly exclusively on cognition. The recentrenewal of interest in aVect promises to bring emotionback into the most central of management activities:negotiation. With the tide of much research in all areasof management focusing on emotion, mood, and socialrelationships, the movement away from models exclu-sively focusing on cognitive factors presents unchartedterritory for the scholar. What our study suggests isthat when negotiators display emotions, they need toconsider how the strategic display of emotions willinXuence the other party’s own emotions, cognitions,and behaviors. Researchers must distinguish betweenbest practices that work in all situations and strategicpractices that work well in some and poorly in others(Allred, 2000) and focus not only on how displayedemotion inXuences the outcomes of negotiation asmeasured by traditional economic indices, but alsohow emotion impacts the relationships that developbetween negotiators.

Acknowledgments

We thank the participants of the International Asso-ciation of ConXict Management Conference in Seville in2005 and the AVect and Emotions in OrganizationalBehavior Meeting in Rotterdam in 2005, our three blindreviewers and editor, as well as our colleagues from ourrespective departments for feedback on earlier andrecent versions of this paper. We thank the Dispute Res-olution Research Center for its generous funding of thisproject.

Appendix A. Manipulations of strategic emotional display in Experiment 1

A.1. Expert negotiation advice: Developing a positive bargaining style

A.1.1. Use positive emotionEVective negotiation requires creative information

processing and it is positive, rather than negative emo-tion that instigates such cognitive processing (Forgas,1998). When people are experiencing a positive mood,they are more creative (Isen et al., 1987).

A.1.2. Leave the other party feeling goodƒpositive feelings in opponents beneWt future negoti-

ations with them. ƒif an opponent leaves the negotia-tion feeling good about the process and the outcome,that person will be likely to engage in a cooperative fash-ion in subsequent negotiations and to fulWll the terms ofthe current contract. ƒAn opponent who feels goodabout a negotiation may speak highly of [the negotiator]and portray the negotiator as fair and cooperative(Thompson et al., 2001).

A.1.3. Avoid hostilityƒhostility in a negotiation may breed further hostil-

ity that can spiral out of control (Thompson et al., 2001).Once an attack-defense cycle gets going the parties

queue up to get their thrust in. The faster the attacks,and their replies, the higher the emotional tension. Peo-ple in an emotional state make threats, not necessarilyintending to carry them out, but threats provokecounter-threats and the parties may end up in a mutualexchange of sanctions because they boxed themselvesinto corners from which a retreat would [seemingly] costtoo muchƒ The consequence is that parties get nowhereexcept further apart which is the antithesis of negotiat-ing” (Kennedy, Benson, & McMillan, 1980).

A.1.4. Create positive relationshipsWhereas this may seem completely obvious, it is

much easier said than done. People in disputes oftenbecome angry or act too detached and business-like.


Therefore, it is important to begin negotiations by focus-ing on the relationship you have with the other partyand attempt to create positive atmosphere—no matterwhat your opponent doesƒTake a few moments tofocus on and accentuate the positive aspects of yourrelationship if you Wnd yourself deviating from a con-structive approach.

A.1.5. Reframe from “You against Me” to “We”A simple and powerful way to reframe the situation

from “you” or “me” to “we” is through body language.When people argue, they usually stand or sit face-to-face, physically expressing their confrontation. So Wnd anatural excuse to sit side by side. Pull out a document orproposed agreement and sit down next to your counter-part to review itƒTalking side by side will not magicallytransform the situation, but it will reinforce the idea thatyou are partners facing a tough challenge together (Uryet al., 1988).

A.2. Expert negotiation advice: Developing a resolute bargaining style

A.2.1. Be persistentMost people are not persistent enough when negotiat-

ing. They present something to the other side, and if theother side does not ‘buy’ it right away, they shrug andmove on to something else. If that is a quality you have, Isuggest you change it. Learn to hang in there. You mustbe tenacious (Cohen, 1980).

A.2.2. Be committed to your beliefsPassionate commitment is the ability to convince the

other party that you feel deeply and strongly about yourown claimƒTo the extent that you can convince theother party that you will follow through with whatseems to be an extreme course of action, the more likelyyou are to attain your goalsƒDo not hesitate to use thepower of emotion and drama at critical junctures in thenegotiation.

A.2.3. Begin with extreme positionsTo the extent that a person makes what is perceived

as an outlandish, ridiculous request, he or she is morelikely to secure agreement to a following, smallerrequest. When two diVerent requests are compared, oneextreme and the other modest, the second request is per-ceived to be much more reasonable than if only the sec-ond request were made in isolation (Cialdini, 1993).

A.2.4. Employ negative reinforcementIf the radio is playing obnoxious, unpleasant music,

most people will turn the radio oV. Similarly, since mostpeople Wnd it unpleasant to be around someone who isopenly hostile and negative, they may be willing to givethe person what he or she wants just to remove them-

selves from the aversive situation (Thompson et al.,2001).

A.2.5. Be toughSignal toughness throughout the negotiation so that

the opponent will respect your position. Negotiatorswho make fewer concessions and make smaller conces-sions are indeed more eVective in terms of maximizingindividual gain compared to those who make largerand more frequent concessions (Siegel & Fouraker,1960).

A.3. Expert negotiation advice: Developing a rational bargaining style

A.3.1. Control your emotional displayEven though a negotiator may feel emotion, he or she

should not express it. The negotiator who expressesrelief, satisfaction, and/or approval risks settling for aworse outcome than does a rational (poker-faced) nego-tiator (Thompson et al., 2001).

ƒPeople in an emotional state do not want to think,and they are particularly susceptible to the power of sug-gestion from a clever opponentƒ [an] excitable person isputty in the hands of a calm, even-tempered negotia-torƒ (Nierenberg, 1968).

A.3.2. Make strategic decisionsThe aspect of negotiation that an executive can con-

trol most directly is how he or she makes decisions. Theparties, the issues, and the negotiation environment areoften beyond [the negotiator’s] control. Rather thanseeking to change them, you must improve your abilityto make eVective, more rational decisions—to negotiatesmarter (Bazerman & Neale, 1992).

A.3.3. Be professionalConduct yourself in a professional, dispassionate

fashion; consider [the negotiation] to be a business meet-ing—no matter how the other party approaches the situ-ation.

A.3.4. Develop rational strategiesThe desire to ‘win’ at any cost should not preempt a

rational negotiation strategy. Misdirected persistencecan lead to wasting a great deal of time, energy, andmoney. Directed persistence can lead to commensuratepayoVs. Rational analysis enables you to distinguishbetween the two (Bazerman & Neale, 1992).

A.3.5. Think logicallyIt is important for disputants to recognize that emo-

tions can overwhelm logic. In fact, people are sometimestrapped into acting against their own best interests, evenwhen they recognize that they are doing so (Susskind &Cruikshank, 1987, p. 89).


One psychological and emotional pattern thatemerges again and again is the dynamic of escalation.Once conXicts begin, the emotional levels of the partici-pants tend to rise, and the situation becomes more andmore diYcult to defuse. ƒAs the conXict intensiWes, theyare less likely to listen seriously and think clearly. Unfor-tunately, such behavior on the part of one party merelyencourages similar behavior by the other (Susskind &Cruikshank, 1987, p. 93).

References

Allred, K. G. (1999). Anger and retaliation: Toward an understandingof impassioned conXict in organizations. In R. J. Bies & R. J. Lew-icki (Eds.), Research in negotiation in organizations (Vol. 7, pp. 27–58). Stamford, CT: JAI Press.

Allred, K. G. (2000). Distinguishing best and strategic practices: Aframework for managing the dilemma between creating and claim-ing value. Negotiation Journal, 16(4), 387–397.

Allred, K. G., Mallozzi, J. S., Matsui, F., & Raia, C. P. (1997). The inXu-ence of anger and compassion on negotiation performance. Organi-zational Behavior and Human Decision Processes, 70(3), 175–187.

Anderson, C., Keltner, D., & John, O. P. (2003). Emotional conver-gence between people over time. Journal of Personality and SocialPsychology, 84(5), 1054–1068.

Babcock, L., Lowenstein, G., IssacharoV, S., & Camerer, C. (1992).Biased judgments of fairness in bargaining. American EconomicReview, 85(5), 1337–1343.

Baron, R. A. (1990). Environmentally induced positive aVect: Itsimpact on self-eYcacy, task performance, negotiation, and conXict.Journal of Applied Social Psychology, 20(5), 368–384.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator vari-able distinction in social psychological research: Conceptual, stra-tegic, and statistical considerations. Journal of Personality andSocial Psychology, 51(6), 1173–1182.

Barry, B. (1999). The tactical use of emotion in negotiation. In R. J.Bies & R. J. Lewicki (Eds.), Research in negotiation in organizations(Vol. 7, pp. 93–121). Stamford, CT: JAI Press.

Barry, B., & Oliver, R. L. (1996). AVect in dyadic negotiation: A modeland propositions. Organizational Behavior and Human DecisionProcesses, 67(2), 127–143.

Barsade, S. G. (2002). The ripple eVect: Emotional contagion and itsinXuence on group behavior. Administrative Science Quarterly, 47,644–675.

Baumeister, R. F., Leith, K. P., Muraven, M., & Bratslavsky, E. (1998).Self-regulation as a key to success in life. In D. Pushkar & W. M.Bukowski (Eds.), Improving competence across the lifespan: Buildinginterventions based on theory and research (pp. 117–132). New York:Plenum Press.

Bazerman, M. H., & Neale, M. A. (1992). Negotiating rationally. NewYork: Free Press.

Bazerman, M. H., & Neale, M. A. (1982). Improving negotiation eVec-tiveness under Wnal oVer arbitration: The role of selection andtraining. Journal of Applied Psychology, 67(5), 543–548.

Brodt, S. E., & Tuchinsky, M. (2000). Working together but in opposi-tion: An examination of the “good cop/bad cop” negotiating teamtactic. Organizational Behavior and Human Decision Processes,81(2), 155–177.

Carnevale, P. J., & Isen, A. M. (1986). The inXuence of positive aVectand visual access on the discovery of integrative solutions in bilat-eral negotiation. Organizational Behavior and Human Decision Pro-cesses, 37(1), 1–13.

Chin, W. W. (2001). PLS [Computer software]: Soft Modeling Inc.

Cialdini, R. B. (1993). InXuence: The psychology of persuasion. NewYork: William Morrow.

Cohen, H. (1980). You can negotiate anything. New York: BantamBooks.

Coté, S. (2005). A social interaction model of the eVects of emotion reg-ulation on work strain. Academy of Management Review, 30(3),509–530.

Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and thehuman brain. New York: Grosset/Putnam.

Deutsch, M. (1973). The resolution of conXict. New Haven, CT: YaleUniversity Press.

Drolet, A. L., & Morris, M. W. (2000). Rapport in ConXict Resolution:Accounting for how face-to-face contact fosters mutual coopera-tion in mixed-motive conXicts. Journal of Experimental Social Psy-chology, 36(1), 26–50.

Efron, B., & Tibshirani, R. (1986). Bootstrap methods for standarderrors, conWdence intervals, and other measures of statistical accu-racy. Statistical Science, 1, 54–77.

Ekman, P. (1972). Universals and cultural diVerences in facial expres-sions of emotion. In J. Cole (Ed.), Nebraska symposium on motiva-tion, 1971 (pp. 207–283). Lincoln: University of Nebraska Press.

Ekman, P. (1993). Facial expression and emotion. American Psycholo-gist, 48(4), 384–392.

Falbe, C. M., & Yukl, G. (1992). Consequences for managers of usingsingle inXuence tactics and combinations of tactics. Academy ofManagement Journal, 35(3), 638–652.

Fisher, R., Ury, W., & Patton, B. (1991). Getting to yes: Negotiatingagreement without giving in. New York: Penguin Books.

Forgas, J. P. (1995). Mood and judgment: The aVect infusion model(AIM). Psychological Bulletin, 117(1), 39–66.

Forgas, J. P. (1998). On feeling good and getting your way: MoodeVects on negotiator cognition and bargaining strategies. Journal ofPersonality and Social Psychology, 74(3), 565–577.

Forgas, J. P. (2001). On being moody but inXuential: The role of aVectin social inXuence strategies. In J. P. Forgas & K. D. Williams(Eds.), Social inXuence: Direct and indirect processes (Vol. xxi, pp.147–166). New York: Psychology Press.

Forgas, J. P., & George, J. M. (2001). AVective inXuences on judgmentsand behavior in organizations: An information processing perspec-tive. Organizational Behavior and Human Decision Processes, 86(1),3–34.

Frank, R. H. (1988). Passions within reason: The strategic role of theemotions. New York: W.W. Norton.

Fredrickson, B. L. (2001). The role of positive emotions in positive psy-chology: The broaden-and-build theory of positive emotions.American Psychologist, 56(3), 218–226.

Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speedrecovery from the cardiovascular sequelae of negative emotions.Cognition and Emotion, 12(2), 191–220.

Frijda, N. H. (1986). The emotions. Cambridge: Cambridge UniversityPress.

Grandey, A. A. (2000). Emotion regulation in the workplace: A newway to conceptualize emotional labor. Journal of OccupationalHealth Psychology, 5(1), 95–110.

Greenhalgh, L. (1993). Viking investments: Creative consensus.Greenhalgh, L., & Chapman, D. I. (1995). Joint decision making: The

inseparability of relationships and negotiation. In R. M. Kramer &D. M. Messick (Eds.), Negotiation as a social process. ThousandOaks, CA: Sage Publications.

Gross, J. J. (1998). Antecedent- and response-focused emotion regula-tion: Divergent consequences for experience, expression, and physi-ology. Journal of Personality and Social Psychology, 74(1), 224–237.

Gross, J. J., John, O. P., & Richards, J. M. (2000). The dissociation ofemotion expression from emotion experience: A personality per-spective. Personality and Social Psychology Bulletin, 26(6), 712–726.

HatWeld, E., Cacioppo, J. T., & Rapson, R. L. (1992). Primitive emo-tional contagion. In M. S. Clark (Ed.), Emotion and social behavior


Review of personality and social psychology (Vol. 14, pp. 151–177).Newbury Park, CA: Sage Publications.

Hilty, J. A., & Carnevale, P. J. (1993). Black-hat/white-hat strategy inbilateral negotiation. Organizational Behavior and Human DecisionProcesses, 55(3), 444–469.

Hochschild, A. R. (1983). The managed heart: Commercialization ofhuman feeling. Berkeley: University of California Press.

Hollingshead, A. B., & Carnevale, P. (1990). Positive aVect and decisionframe in integrative bargaining: A reversal of the frame eVect.Academy of Management Best Paper Proceedings, 385–389.

Isen, A. M. (1987). Positive aVect, cognitive processes, and socialbehavior. In L. Berkowitz (Ed.), Advances in experimentalsocial psychology (Vol. 20, pp. 203–253). San Diego: AcademicPress.

Isen, A. M., Daubman, K. A., & Nowicki, G. P. (1987). Positive aVectfacilitates creative problem solving. Journal of Personality andSocial Psychology, 52(6), 1122–1131.

Janis, I. L., & Mann, L. (1977). Decision making: A psychological analy-sis of conXict, choice, and commitment. New York: The Free Press.

Keltner, D., Ellsworth, P. C., & Edwards, K. (1993). Beyond simple pes-simism: EVects of sadness and anger on social perception. Journalof Personality and Social Psychology, 64(5), 740–752.

Kennedy, G., Benson, J., & McMillan, J. (1980). Managing negotiations.Englewood CliVs, NJ: Prentice-Hall.

Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in socialpsychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.),Handbook of social psychology (Vol. 1, pp. 233–265). New York:Oxford University Press.

Kipnis, D., Schmidt, S. M., & Wilkinson, I. (1980). IntraorganizationalinXuence tactics: Explorations in getting one’s way. Journal ofApplied Psychology, 65(4), 440–452.

Kopelman, S., & Olekalns, M. (1999). Process in cross-cultural negotia-tions. Negotiation Journal, 15(4), 373–380.

Kramer, R., & Messick, D. M. (Eds.). (1995). Negotiation as a socialprocess. Thousand Oaks, CA: Sage Publications.

Kramer, R., Newton, E., & Pommerenke, P. (1993). Self-enhancementbiases and negotiator judgment: EVects of self-esteem and mood.Organizational Behavior and Human Decision Processes, 56(1), 110–133.

Kravitz, D. A., & Gunto, S. (1992). Decisions and perceptions of recipi-ents in ultimatum bargaining games. Journal of Socio Economics,21(1), 65–84.

Kumar, R. (1997). The role of aVect in negotiations: An integrativeoverview. Journal of Applied Behavioral Science, 33(1), 84–100.

Lawler, E. J., & Yoon, J. (1995). Structural power and emotional pro-cesses in negotiation: A social exchange approach. In R. M. Kramer& D. M. Messick (Eds.), Negotiation as a social process. ThousandOaks, CA: Sage Publications.

Lazarus, R. S. (1984). On the primacy of cognition. American Psycholo-gist, 39(2), 124–129.

Leung, K., & Fan, R. M. (1997). Dispute processing: An Asian perspec-tive. In H. S. R. Kao & D. Sinha (Eds.), Asian perspectives on psy-chology. Cross-cultural research and methodology series (Vol. 19, pp.201–217). Thousand Oaks, CA: Sage Publications.

Levenson, R. W. (1994a). Emotional control: Variations and conse-quences. In P. Ekman & R. J. Davidson (Eds.), The nature of emo-tion (pp. 273–279). New York: Oxford University Press.

Levenson, R. W. (1994b). Human emotions: A functional view. InP. Ekman & R. Davidson (Eds.), The nature of emotion: Funda-mental questions (pp. 123–126). New York: Oxford UniversityPress.

Leventhal, G. S., & Bergman, J. T. (1969). Self-depriving behavior as aresponse to unproWtable inequity. Journal of Experimental SocialPsychology, 5(2), 153–171.

MacKinnon, D. P., Lockwood, C. M., & HoVman, J. M. (2002). A com-parisons of methods to test mediation and other intervening vari-ables eVects. Psychological methods, 7(1), 83–104.

McGinn, K. L., & Keros, A. T. (2002). Improvisation and the logic ofexchange in socially embedded transactions. Administrative ScienceQuarterly, 47(3), 442–473.

Neale, M. A., & Bazerman, M. (1991). Cognition and rationality in nego-tiation. New York: Free Press.

Nierenberg, G. I. (1968). The art of negotiating. New York: Corner-stone Library.

Oliver, R. L. (1997). Satisfaction: A behavioral perspective on the con-sumer. Boston: Irwin McGraw-Hill.

Pillutla, M. M., & Murnighan, J. K. (1996). Unfairness, anger, andspite: Emotional rejections of ultimatum oVers. OrganizationalBehavior and Human Decision Processes, 68(3), 208–224.

Pruitt, D. G., & Carnevale, P. J. (1993). Negotiation in social conXict.PaciWc Grove, CA: Brooks/Cole Publishing.

Rafaeli, A., Fiegenbaum, T., Maw-Der, F., & Hwee-Hoon, T., (2005).Perceived relationships between anger, guild, and promotion deci-sions in Israel and Singapore. Working Paper.

Rafaeli, A., & Sutton, R. I. (1991). Emotional contrast strategies asmeans of social inXuence: Lessons from criminal interrogators andbill collectors. Academy of Management Journal, 34(4), 749–775.

RaiVa, H. (1982). The art and science of negotiation. Cambridge: Har-vard University Press.

Roth, A. E. (1995). A brief history of experimental economics. In J. H.Kagel & A. E. Roth (Eds.), The handbook of experimental economics(pp. 3–109). Princton, NJ: Princeton University Press.

Schelling, T. C. (1960). The strategy of conXict. Cambridge: HarvardUniversity.

Schwarz, N., & Clore, G. L. (1996). Feelings and phenomenal experi-ences. In E. Higgins & A. Kruglanski (Eds.), Social psychology:Handbook of basic principles (pp. 433–465). New York: GuilfordPress.

Shiv, B., Loewenstein, G., Bechara, A., Damasio, H., & Damasio, A. R.(2005). Investment behavior and the negative side of the emotion.Psychological Science, 16(6), 435–439.

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-experimental studies: New procedures and recommendations. Psy-chological Methods, 7(4), 422–445.

Siegel, S., & Fouraker, L. E. (1960). Bargaining and group decision mak-ing. New York: McGraw-Hill.

Sinaceur, M., & Neale, M. A. (2005). Not all threats are created equal:How implicitness and timing aVect the eVectiveness of threats innegotiations. Group Decision and Negotiation, 14(1), 63–85.

Sinaceur, M., & Tiedens, L. Z. (2005). Get mad and get more than even:When and why anger expression is eVective in negotiations. Journalof Experimental Social Psychology, in press.

Singelis, T. M. (1998). Teaching about culture, ethnicity and diversity:Exercises and planned activities. Thousand Oaks, CA: Sage Publica-tions.

Sondak, H., Neale, M. A., & Pinkley, R. L. (1999). Relationship, contri-bution, and resource constraints: Determinants of distributive jus-tice in individual preferences and negotiated agreements. GroupDecision and Negotiation, 8(6), 489–510.

Susskind, L., & Cruikshank, J. (1987). Breaking the impasse: Consen-sual approaches to resolving public disputes. New York: BasicBooks.

Sutton, R. I., & Rafaeli, A. (1988). Untangling the relationship betweendisplayed emotions and organizational sales: The case of conve-nience stores. Academy of Management Journal, 31(3), 461–487.

Thompson, L., & Hastie, R. (1990). Social perception in negotiation.Organizational Behavior and Human Decision Processes, 47(1), 98–123.

Thompson, L., & Kim, P. H. (2000). How the quality of third parties’settlement solutions is aVected by the relationship between negotia-tors. Journal of Experimental Psychology: Applied, 6(1), 3–14.

Thompson, L., Medvec, V. H., Seiden, V., & Kopelman, S. (2001).Poker face, smiley face, and rant ’n’ rave: Myths and realities aboutemotion in negotiation. In M. Hogg & S. Tindal (Eds.), Blackwell


handbook of social psychology: Group processes (pp. 139–163). Mal-den, MA: Blackwell Publishers.

Thompson, L., Nadler, J., & Kim, P. H. (1999). Some like it hot: Thecase for the emotional negotiator. In L. L. Thompson & J. M.Levine (Eds.), Shared cognition in organizations: The management ofknowledge (pp. 139–161). Mahwah, NJ: Lawrence Erlbaum Associ-ates.

Tooby, J., & Cosmides, L. (1990). The past explains the present: Emo-tional adaptations and the structure of ancestral environments.Ethology and Sociobiology, 11(4–5), 375–424.

Ury, W. L., Brett, J. M., & Goldberg, S. B. (1988). Getting disputesresolved: Designing systems to cut the costs of conXict. San Fran-cisco: Jossey-Bass.

Valley, K. L., Neale, M. A., & Mannix, E. A. (1995). Friends, lovers,colleagues, strangers: The eVects of relationships on the process andoutcome of dyadic negotiations. Research on Negotiation in Organi-zations, 5, 65–93.

Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2004a). Theinterpersonal eVects of anger and happiness in negotiations. Journalof Personality and Social Psychology, 86(1), 57–76.

Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2004b).The interpersonal eVects of emotions in negotiations: A motivatedinformation processing approach. Journal of Personality & SocialPsychology, 87(4), 510–528.

Van Maanen, J., & Kunda, G. (1989). “Real Feelings”: Emotionalexpression and organizational culture. In B. M. C. Staw, L. L. (Ed.),Work in Organizations (pp. 193–253). Greenwich: JAI Press.

Waugh, C., & Fredrickson, B. L. (2002). The beneWts of feeling good andfeeling close: Positive emotions, self-other overlap, and perspective-taking. Working Paper.

Yukl, G., & Falbe, C. M. (1990). InXuence tactics and objectives inupward, downward, and lateral inXuence attempts. Journal ofApplied Psychology, 75(2), 132–140.

Zajonc, R. B. (1984). On the primacy of aVect. American Psychologist, 39(2).

Documents

The three faces of Eve: Strategic displays of positive ...ibavenegotiations.weebly.com/uploads/1/0/8/0/10805615/three_faces... · of positive, negative, ... managers strategically