Journal of Experimental Social Psychology 46 (2010) 543–550

Contents lists available at ScienceDirect

Journal of Experimental Social Psychology journal homepage: www.elsevier.com/locate/jesp

Report

Suspending judgment to create value: Suspicion and trust in negotiation Marwan Sinaceur INSEAD, Boulevard de Constance, 77305 Fontainebleau, France

a r t i c l e

i n f o

Article history: Received 31 March 2008 Revised 26 October 2009 Available online 5 November 2009 Keywords: Motive attributions Trust Negotiation Information search Integrative agreements

a b s t r a c t This paper introduces a distinction between suspicion and distrust. While distrust (trust) involves having negative (positive) expectations about another’s motives, suspicion is defined as the state in which perceivers experience ambiguity about another’s motives. Four experiments supported this distinction and showed that suspicion can present greater benefits than trust for generating information search and attaining integrative agreements in negotiation. In Experiment 1a, suspicious perceivers were characterized by consciously attributing more motives to a target compared to both distrusting and trusting perceivers. In Experiment 1b, suspicious perceivers were more willing to seek information. In Experiment 2a, Suspicious–Trusting dyads achieved greater joint outcomes in face-to-face negotiation than did Trusting– Trusting or Suspicious–Suspicious dyads. Experiment 2b showed that the suspicious participants’ ability to seek information in Suspicious–Trusting dyads mediated the superior performance of Suspicious– Trusting dyads over Trusting–Trusting dyads in attaining integrative agreements. Ó 2009 Elsevier Inc. All rights reserved.

Introduction Is trust always efficient in negotiation? According to philosophers, suspicion or skepticism is essential to acquire knowledge (Descartes, 1637/2006; Hume, 1748/2000). For example, in ancient povg  Þ, Greece the Skeptics philosophers recommended epochê ðe or the suspension of judgment that goes with recognizing how ambiguous (i.e., subject to rival interpretations) the world can be, rather than trust (Suber, 1996). In this paper I explore these ideas in the context of negotiations. Specifically, I investigate whether suspicion, defined as the state in which perceivers experience ambiguity about a target’s motives, can be more efficient than trust in generating information search in negotiation, and this greater information search can result in more mutually beneficial, or value creating, agreements.

Suspicion as ambiguity about another’s motives Social psychologists have defined suspicion or skepticism as the state in which a perceiver actively entertains different, plausibly rival, hypotheses about another’s motives (Fein, 1996; Fein & Hilton, 1994; Hilton, Fein, & Miller, 1993). As such, it means experiencing ambiguity about another’s motives. Specifically, suspicion lies in entertaining rival interpretations about a target’s behavior simultaneously. In this way, suspicion is more specific than uncertainty. While uncertainty is when people generally do not know for sure about something (Tiedens & Linton, 2001), suspicion goes one E-mail address: [email protected] 0022-1031/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jesp.2009.11.002

step further: it is specifically about thinking that something can have rival meanings – hence, inclining in no particular direction. Defined as such, suspicion entails specific attributions. It involves the belief that another might have hidden motives (Hilton et al., 1993). And, it induces more conscious attributional thinking (Fein, 1996; Hilton et al., 1993). Characteristically, suspicious perceivers consider carefully potential motives that underlie a target’s behavior (Fein, 1996). Further, arguments from previous research suggest that suspicion could increase perceivers’ rational information search about a target (Kramer, 1998; Schul, Burnstein, & Bardi, 1996). For example, suspicion makes perceivers better discriminate relevant information (Schul et al., 1996) and be less susceptible to biases in information search (Fein, Morgan, Norton, & Sommers, 1997). Although these effects have not been demonstrated in interpersonal interactions, it seems likely that they would extend to negotiation and they suggest a relationship between suspicion and extent of information search. Specifically, it is plausible that suspicion will generate greater information search about a target in negotiation. Because information search is an important negotiation process (De Dreu & Van Kleef, 2004; Thompson, 1991), suspicion-induced information search should, in turn, affect negotiation outcomes.

Suspicion vs. distrust/trust How does suspicion differ from distrust, however? Previous research (Deutsch, 1958; Fein & Hilton, 1994; Kramer, 1998) did not distinguish suspicion from distrust. It did not empirically disentan-

544

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

gle suspicion from distrust (Fein, 1996; Schul et al., 1996) and often used these terms interchangeably (Kramer, 1998). Trust (distrust) means having positive (negative) expectations about another’s motives (Deutsch, 1958; Kramer, 1999; Rubin & Brown, 1975). So, suspicion does presuppose a lack of trust. Specifically, this lack of trust means experiencing ambiguity about a target’s motives. Distrust presupposes a lack of trust too. Yet, this latter lack of trust is, I argue, different: it means not trusting a target; this implies a definitive judgment, albeit negative. Thus, in this paper I separate between suspicion (i.e., experiencing ambiguity) and distrust (i.e., making a negative, unidirectional judgment). Inspired by Fein’s (1996) work, I predict that suspicion differs from distrust through more conscious motive attributions.

pants’ spontaneous thoughts (see Malle, Knobe, O’Laughlin, Pearce, & Nelson, 2000). Experiment 1a explored whether suspicion differed from distrust/trust through motive attributions. Experiment 1b examined whether suspicion affected willingness to seek information. Experiments 2a and 2b investigated the impact of suspicion over trust in face-to-face negotiations. A second way to manipulate suspicion was pre-tested. Drawing from this second manipulation, Experiment 2a examined whether Suspicious–Trusting dyads created more value in face-to-face negotiations than did Trusting– Trusting dyads (and than did Suspicious–Suspicious dyads). Experiment 2b tested why Suspicious–Trusting dyads outperformed Trusting–Trusting dyads by examining the mediation by the suspicious party’s information search.

The benefits of trust do not include information search Design strategy Suspicion is likely to increase information search as suggested earlier. Conversely, trust is likely to decrease information search because it increases the willingness to accept information provided at face value (Parks, Henager, & Scamahorn, 1996). In fact, while trust has considerable benefits in negotiation, these do not include information search, but rather information providing. Research consistently demonstrated that trust enhances value creation (i.e., joint outcomes) in negotiation (De Dreu, Beersma, Stroebe, & Euwema, 2006; De Dreu, Giebels, & Van de Vliert, 1998; Rubin & Brown, 1975). One primary process underlying this is information exchange. That is, trust increases information exchanged in negotiation (De Dreu et al., 1998; Pruitt, 1981), which, in turn, facilitates identifying mutually favorable agreements (De Dreu et al., 2006; Pruitt, 1981; Thompson & Hastie, 1990). Behind information exchange is one particular mechanism though. Specifically, trust facilitates providing information because one has no fear that the counterpart may exploit this (De Dreu et al., 1998; Pruitt & Kimmel, 1977; Rubin & Brown, 1975). Thus, the mechanism through which trust enhances value creation is information providing, not information search. Suspicion, trust, and value creation This suggests that, compared to trust, suspicion will enhance information search about a target in negotiation. This is important because information search increases value creation in negotiation, independently of information providing (Thompson, 1991). Further, value creation increases even when only one member of the negotiating dyad seeks more information (Thompson, 1991). Hence, I predict that suspicious negotiators’ information search will increase value creation when the counterpart is trusting. Specifically, dyads in which one party is suspicious and the other trusting will create more value than dyads in which both parties are trusting. In Suspicious–Trusting dyads, the suspicious party will seek more information such that this will make these dyads create more value than the Trusting–Trusting dyads in which both parties are willing to provide information but neither engages in careful information search. Thus, the current research examines when trust, which has been deemed optimal for value creation in negotiation (Pruitt, 1981), may actually not be so. Specifically, it explores whether suspicion is more efficient than trust when trust leads both parties to take what the other says at face value, thus limiting information search. Overview Four experiments examined whether (1) suspicion is a mental state distinct from distrust/trust (Experiments 1a and 1b) and (2) suspicion creates more value than trust in negotiation (Experiments 2a and 2b). Experiments 1a and 1b documented partici-

Suspicion, as suggested previously, is theoretically distinguishable from uncertainty. It is likely to empirically involve some general uncertainty, however. This matters because uncertainty induces more sophisticated attributions (Weary, Vaughn, Stewart, & Edwards, 2006) and information search (Tiedens & Linton, 2001; Weary, Jacobson, Edwards, & Tobin, 2001). To address this, I controlled for uncertainty through the design in all experiments, i.e., by considering trust/distrust conditions that were associated with uncertainty. I expected that, compared to uncertain trust/distrust, suspicion would not differ in uncertainty, but would differ in motive attributions and information search.1 Experiment 1a: suspicion vs. distrust/trust Methods Participants and design Participants were 105 undergraduates in a mass-testing session. The between-subjects design was a 2 (trust vs. distrust)  2 (uncertainty associated with trust/distrust vs. certainty associated with trust/distrust) factorial design with an extra condition (suspicion). Materials Negotiation scenario Participants read a negotiation scenario in which they were made an offer by a target. The scenario was identical across conditions, except the information given about the target’s motives in making an offer. Trust/distrust associated with certainty In the trust [distrust] associated with certainty conditions, participants read that the target ‘‘undeniably had benevolent [malevolent] motives in making them this offer”. They were told to ‘‘trust the target very much [not to trust the target at all]”. The brackets above indicate the alternative wording for distrust. Trust/distrust associated with uncertainty In the trust [distrust] associated with uncertainty conditions, participants read that the target ‘‘was likely to have benevolent [malevolent] motives in making them this offer”. They were told to ‘‘trust the target moderately [not to trust the target much]”. The brackets above indicate the alternative wording for distrust. 1 Controlling for participants’ reported uncertainty did not affect the results in any of the experiments.

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

Suspicion Drawing from the definition of suspicion as thinking that another’s motives are ambiguous (Fein, 1996), suspicion was operationalized as thinking that another could have rival motives, both benevolent and malevolent. In the suspicion condition, participants read that the target ‘‘could both have benevolent and malevolent motives”. They were told to believe that the target ‘‘might have hidden motives (both benevolent and malevolent) that underlie their offer”. Specifically, these instructions were consistent with the notion that suspicion involves entertaining rival hypotheses as there could be hidden meaning in the target’s behavior (Fein, 1996, p. 1165). Measures Checks Participants assessed their uncertainty and their trust using 7point Likert scales. Uncertainty was the average of six items (certainty, knowledge of the target’s motives, judgment confidence, understanding of the issues, all four reverse scored; puzzlement; hesitation), Cronbach’s a = .65. Trust was the average of three items (trust; expectation of cooperativeness; expectation of deceitfulness, reverse scored), Cronbach’s a = .66. Participants’ explanations In an open-ended format, participants were asked to give possible explanations for the target making them an offer. This paradigm is common in attributions research (see Malle, Knobe, O’Laughlin, Pearce, & Nelson, 2000). Then, participants’ explanations were coded using Malle’s (Malle, 1999; Malle, Knobe, O’Laughlin, Pearce, & Nelson, 2000) detailed coding scheme for people’s attributions. Explanations were coded along two categories: conscious motive attributions (termed ‘‘marked desire reasons” by Malle); and any other attributions (e.g., stable traits or situational causes).2 Conscious motive attributions are explanations that describe the target’s motives, that is, what the target punctually tries to fulfill. They do so by mentioning verbs that indicate transient mental states, such as ‘‘she wants”, ‘‘he needs”, ‘‘she intends to” – hence, explicitly recognizing the target as a subjective, intentional agent (Malle, 1999). Overall, they mean that a perceiver consciously entertains a target’s motives.

conditions reported more trust than in the distrust conditions (b = .48; t(100) = 5.51, p < .001). There were no other effects. Finally, a regression with the suspicion vs. distrust and the suspicion vs. trust differences as two predictors showed that participants in the suspicion condition reported more trust than in the distrust conditions (b = .39; t(102) = 3.81, p < .001) and less trust than in the trust conditions (b = .42; t(102) = 4.20, p < .001). Participants’ explanations To examine participants’ explanations, I tested the suspicion condition against every other condition. Following the attributions literature (e.g., Malle, Knobe, O’Laughlin, Pearce, & Nelson, 2000), I treated categories of explanations as multiple measures and conducted multivariate analyses of variance (MANOVAs). All four MANOVAs on the contrasts comparing suspicion to each of the other four conditions yielded a significant effect for conscious motive attributions (all four multivariate Fs > 5.76, all four ps < .03), and no effect for any other explanation category. Discussion Suspicion was characterized by more conscious motive attributions and, consistent with my prediction, only by these attributions (see footnote 2). This result extends prior work (Fein, 1996; Schul et al., 1996) by disentangling for the first time suspicion from distrust. Further, while suspicion occupied the middle area of a trust/distrust continuum in trust judgments, it did not for conscious motive attributions. Rather, suspicion was the maximum of an inverted U-shape for conscious motive attributions. Departing from prior research that emphasized different degrees in trust (Kramer, 1999) or different bases for trust (Lewicki & Bunker, 1996), these results suggest that suspicion is of a distinct type compared to trust/distrust. Experiment 1b: suspicion and information search Experiment 1b probed whether suspicion induced greater propensity for information search. Because I was interested in an effect of suspicion beyond uncertainty, I examined differences between suspicion and distrust/trust associated with uncertainty.

Results

Methods

Table 1 presents means. Conscious motive attributions serve for testing the distinction between suspicion and other states.

Participants and design

Checks Uncertainty check First, a regression on participants’ uncertainty was conducted with four predictors based on contrast coding: the uncertainty and trust factors, their interaction, and the difference between suspicion and other states (these four predictors formed a set of orthogonal contrasts, which allowed to partition out the total variance, Howell, 1997, Section 12.3). Participants in the uncertain distrust/trust conditions reported more uncertainty than in the certain distrust/trust conditions (b = .26; t(99) = 2.71, p < .01). There were no other effects. Trust check Then, a regression on participants’ trust was conducted with the same four orthogonal predictors as above. Participants in the trust 2

A finer-grained categorization yielded no different results.

545

Participants were 99 undergraduates in a mass-testing session. They were randomly assigned to one of three conditions: suspicion; trust associated with uncertainty; or distrust associated with uncertainty. Materials The scenario and manipulations were the same as in Experiment 1a, except that there were three conditions instead of five. Measures Uncertainty Participants assessed their uncertainty using 7-point Likert scales. Uncertainty was the average of five items (uncertainty; hesitation; knowledge of the target’s motives, understanding of the issues, predictability of the target’s intentions, all last three reverse scored), Cronbach’s a = .67.

546

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

Table 1 Experiment 1a: mean values for measures. Trust  uncertainty Suspicion (N = 19) Uncertainty Trust Conscious motive attributions

Trust with uncertainty (N = 20)

.00a,b (4.44) 3.0a (2.69) 1.58a (1.22)

.75a (4.46) 3.9b (2.61) .35b (.67)

Distrust with uncertainty (N = 22) 1.33a (2.50) 1.43c (2.45) .82b (.80)

Trust with certainty (N = 22) 2.09b (3.82) 5.23b (3.07) .45b (.60)

Distrust with certainty (N = 22) .18b (3.05) 1.27c (2.62) .32b (.48)

Note: Means across rows that do not share a subscript differ significantly at p < .05. More negative means for the uncertainty scale signify lower uncertainty (higher certainty). Means for conscious motive attributions are based on Malle’s (1999) coding scheme. Standard deviations are in parentheses.

Strategy statements Participants were asked to describe the strategies they would intend using in a negotiation with the target from the scenario. Participants’ statements were coded for the number of strategies aimed at seeking information about the counterpart that they mentioned. These typically included ‘‘listening” and ‘‘asking questions”. A measure for information search strategies was obtained from averaging results by two coders (Cronbach’s a = .85). Results Table 2 presents means. As expected, suspicion did not elicit more uncertainty than did uncertain distrust/trust (ns). More importantly, controlling for the total of strategies, suspicious participants mentioned more information search strategies than did uncertain distrusting participants (F(1, 58) = 8.63, p < .005) and than did uncertain trusting participants (F(1, 65) = 4.55, p < .05), respectively. Discussion Experiments 1a and 1b documented cognitive differences between suspicion and other states, suggesting that suspicion did not occupy the middle area of a trust/distrust continuum (except for trust judgments; see Tables 1 and 2). Specifically, suspicion was characterized by more conscious motive attributions (Experiment 1a) and elicited greater propensity for information search (Experiment 1b). However, one question is whether this benefit of suspicion has behavioral consequences in negotiation. Experiments 2a and 2b explored this question. Previous research consistently showed that trust is more beneficial than distrust for value creation (De Dreu et al., 1998; De Dreu et al., 2006). Based on that, I wanted to explore whether suspicion could have benefits over trust for value creation. Thus, Experiments 2a and 2b examined the impact of suspicion compared to trust in face-to-face negotiations. Experiment 2a: suspicion and trust in face-to-face negotiations Pilot study Experiment 2a included a pilot study to pretest a second manipulation of suspicion that could be used in face-to-face negotiations. Table 2 Experiment 1b: Suspicion and information search. Trust/distrust with uncertainty Suspicion (N = 30) Uncertainty Information search

1.37a (4.52) .55a (.81)

Trust associated with uncertainty (N = 36)

Distrust associated with uncertainty (N = 28)

.56a (4.62) .19b (.52)

.18a (3.84) .12b (.33)

Note: Means across rows that do not share a subscript differ significantly at p < .05. More negative means for the uncertainty scale signify lower uncertainty (higher certainty). Standard deviations are in parentheses.

Suspicion means entertaining rival hypotheses about another’s motives (Fein, 1996). In Experiments 1a and 1b, suspicion was operationalized as thinking that another could have hidden, rival motives – both benevolent and malevolent. Yet another, perhaps more realistic, operationalization could be thinking that one’s counterpart is either benevolent or malevolent, but entertaining rival hypotheses about which category s/he falls into. Specifically, the perceiver could entertain two rival interpretations simultaneously: that the counterpart could be trustworthy, but that s/he also could be untrustworthy. Methods Participants and design Participants were 38 undergraduates in a mass-testing session. They were randomly assigned to one of two conditions: suspicion or trust associated with uncertainty. To simplify reading, trust with uncertainty is referred to as ‘‘trust” in this and subsequent experiments. Materials and procedure Negotiation preparation task Participants performed a negotiation preparation task. They were asked to imagine they would role-play a two-party job negotiation with a counterpart (an imaginary classmate), and that they needed to prepare for it. The job negotiation context was taken from Sinaceur and Tiedens (2006, Experiment 2). Suspicion manipulation Along with the context information, participants were given information regarding negotiation strategies. Specifically, they were advised to consider information about their (imaginary) counterpart’s motives. In the suspicion condition, participants were told to consider that about half of the counterparts assigned to the other party’s role had been instructed to provide inaccurate information about their interests, while the other half of the counterparts assigned to the other party’s role had been instructed to provide accurate information about their interests. Thus, there was an equal chance that the counterpart with whom the participants had been paired had received one among those two instructions. Participants were told their counterpart could be trustworthy, but also equally likely s/he could be untrustworthy. His/her behavior could both be interpreted as trustworthy and as untrustworthy. In the trust condition, participants were told to consider that trust was important to maximize the total pie. They were told their counterpart would probably have benevolent motives. S/he was likely to provide accurate information about her/his interests. Indeed, s/he was likely to think that ongoing cooperation was important. Thus, the participants’ counterpart was likely to be trustworthy. Note that the trust condition suggested that participants could maximize the pie, whereas the suspicion condition did not. This

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

is not optimal because it provides different task-information to participants. Yet, this transgression renders testing the hypothesis that suspicion increases value creation more conservative. Measures Strategy statements Participants were asked to describe the strategies they would intend using in the (imaginary) negotiation. Participants’ statements were coded by two coders (yielding an averaged measure, Cronbach’s a = .84) for the number of information search strategies that were mentioned. Most common examples were: ‘‘asking lots of questions”; ‘‘dropping notes”; ‘‘listening carefully”; ‘‘not interrupting the counterpart” (one participant mentioning one strategy did not mean s/he mentioned another). Close-ended items Then, participants answered items using 9-point Likert scales. Participants rated how much they would trust their counterpart. They also rated how much importance they gave to motives in explaining a counterpart’s behavior. Finally, negative uncertainty was the average of two items (experienced fear, hesitation), Cronbach’s a = .61. Results As expected, participants in the suspicion condition reported less trust (M = 3.86, SD = 1.39) than in the trust condition (M = 6.11, SD = 1.74; F(1, 35) = 18.29, p < .001). Also, reported uncertainty was similar across the two conditions (F(1, 35) = .003, ns). Central to the argument, participants in the suspicion condition gave more importance to motives (M = 6.06, SD = 1.47) than in the trust condition (M = 4.89, SD = 1.81, F(1, 35) = 4.49, p < .05). Further, controlling for the total of strategies, suspicious participants mentioned more information search strategies (M = 1.17, SD = .55) than did trusting participants (M = .36, SD = .18, F(1, 37) = 35.90, p < .0001). Discussion While suspicion was manipulated differently in this pilot study than in Experiments 1a and 1b, it elicited similar cognitive tendencies. In subsequent Experiments, I used this second manipulation to examine the impact of suspicion compared to trust on value creation in face-to-face negotiations.

547

venture negotiation was between two firms (AlphaPharma and BioPharma) who were considering setting up a venture to develop, manufacture and market a new drug in common. Role did not affect any of the negotiation outcomes (see Fig. 1, Note), so this factor is not discussed further. The negotiation was about the terms of the prospective business venture and included six issues: R&D investment, manufacturing facilities, marketing strategy, workforce, profit sharing, and sales coverage. Payoff schedules for each role specified the point values that participants could reach on each issue. They are presented in Appendix A. As can be seen, four issues were integrative (i.e., one party cared more about two issues and the other party more about the other two issues, which made trading-off possible to increase joint gain): R&D investment, manufacturing facilities, marketing strategy, and workforce. One issue was distributive (i.e., higher points for one party translated into lower points for the other): profit sharing. And one issue was congruent (i.e., the parties’ preferences were identical): sales coverage. Finally, participants started the negotiation with zero points. This payoff structure is consistent with previous research (e.g., De Dreu, Koole, & Steinel, 2000; Kray, Reb, Galinsky, & Thompson, 2004). Across conditions, participants were told their goal was to maximize their points: the more points they earned, the better for them. Suspicion manipulation The suspicion and trust (with uncertainty) manipulations were identical to those pre-tested in the pilot study. Procedure Participants had 20 min to read their instructions. Then, they were given 35 min to negotiate face-to-face in dyads. After negotiating, they indicated the agreement they reached, if any. Finally, they completed a questionnaire. Measures Checks Participants assessed their trust for the counterpart (one item) and their uncertainty (average of three items: uncertainty, hesitation, and puzzlement; Cronbach’s a = .73) using 7-point Likert scales. Outcomes Participants indicated the agreement, if any, they had reached. One impasse occurred in the Trusting–Trusting condition and, fol-

Main experiment Methods Participants and design Participants were 96 undergraduates in lab sessions. They were randomly assigned into 48 dyads. Dyads were randomly assigned to one of three conditions: Suspicious–Trusting dyads in which one participant was suspicious and the other trusting; Suspicious– Suspicious dyads in which both participants were suspicious; or Trusting–Trusting dyads in which both participants were trusting. As before, the ‘‘trust” condition was associated with uncertainty. Materials Negotiation exercise Participants role-played a two-party negotiation exercise. Participants were randomly assigned to either one of two equal-power roles in negotiating a new business venture. This new business

Fig. 1. Experiment 2a: value creation by dyads. Note: Dyads were randomly assigned to Suspicious–Trusting dyads (N = 18), Suspicious–Suspicious dyads (N = 12), or Trusting–Trusting dyads (N = 18). Among the Suspicious–Trusting dyads there was no difference in value creation between dyads with one role as the suspicious party and dyads with the other role as the suspicious party, t(43) = .14, ns. Also, no role claimed more than 50% of the value created, t(46) = 1.35, ns.

548

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

lowing previous research (e.g., Kray et al., 2004), was excluded when analyzing outcomes. The participants’ total of points was computed according to the payoff schedules.

one of two conditions: Suspicious–Trusting or Trusting–Trusting dyads. As before, trust was associated with uncertainty. Materials and procedure

Results Checks At the individual level of analysis, participants in the suspicion condition experienced less trust (M = 4.63, SD = 1.32) than in the trust condition (M = 5.20, SD = 1.32, F(1, 95) = 4.43, p < .05). Also, at the individual level of analysis, experienced uncertainty was similar across the suspicion (M = 3.52, SD = 1.14) and trust conditions (M = 3.86, SD = 1.32), F(1, 95) = 1.77, ns. Value claiming Value claiming was the percentage of the total points that one party got for him or herself. A one-sample t-test showed that the suspicious party in the Suspicious–Trusting dyads did not claim more (MSuspicious = 49.7%, SD = 3.9%) than 50% of the value created (ns). Value creation by dyads Value creation was measured by the total points earned by the dyad (possible range = 11,600–22,800). Fig. 1 displays means. Based on my hypothesis, I first conducted a regression with two predictors based on contrast coding (these formed a set of orthogonal contrasts, see Howell, 1997, Section 12.3): the difference between Suspicious– Trusting dyads and the other two dyads; the difference between the other two dyads (Trusting–Trusting and Suspicious–Suspicious). In this regression, the former difference was significant (b = .34; t(44) = 2.35, p < .03), but not the latter (p > .99). Then, I tested two planned contrasts. A first planned contrast showed that Suspicious–Trusting dyads (M = 21,406, SD = 1054) created more value than did Trusting–Trusting dyads (M = 20,553, SD = 1351; t(45) = 2.11, p < .05); that result constituted the key test of the argument. Another planned contrast showed that Suspicious–Trusting dyads created also more value than did Suspicious–Suspicious dyads (M = 20,550, SD = 1189; t(45) = 2.06, p < .05). Discussion Suspicious–Trusting dyads created more value than did either Trusting–Trusting or Suspicious–Suspicious dyads. Thus, suspicion could generate greater value creation than trust. No difference in value creation was found between Suspicious– Suspicious and Trusting–Trusting dyads. This is consistent with Thompson’s (1991) results that there is no difference in value creation between dyads in which both parties seek information and dyads in which both parties provide information. Also, the current results are consistent with other research that found too that one party’s mind state can induce greater value creation for the dyad (Galinsky, Maddux, Gilin, & White, 2008; Maddux, Mullen, & Galinsky, 2008). However, Experiment 2a did not test the proposed information search mechanism underlying the positive effect of suspicion. So in Experiment 2b I tested this proposed mechanism by examining whether the suspicious party’s information search mediated the greater value creation of Suspicious–Trusting dyads compared to Trusting–Trusting dyads. Experiment 2b: mediation by information search Methods Participants and design Participants were 64 undergraduates in courses. They were randomly assigned into 32 dyads. Dyads were randomly assigned to

Participants role-played the same negotiation exercise through the same procedure as in Experiment 2a, with one exception: they were given 20 min (rather than 35 min) to negotiate face-to-face. The suspicion and trust (with uncertainty) manipulations were the same as in Experiment 2a. Measures Uncertainty Uncertainty was measured through one item (experienced puzzlement) on a 7-point Likert scale. Outcomes As in Experiment 2a, outcomes were measured as points that participants reached. There were no impasses in this experiment. Information search Based on the coding from the pilot study, I measured four different information search behaviors through participants’ ratings: asking questions; taking notes about what the counterpart said; remaining silent to gather information; interrupting the counterpart when s/he talked (reverse scored). Asking questions, taking notes, and remaining silent resulted from participants rating their own behaviors; and interrupting resulted from participants being rated by their counterpart. The choice of these diverse items corresponded to the four different behaviors that were cited the most by participants from the pilot study (see earlier). Ratings were on 7point Likert scales. To create a total information search variable, I added together the four behavior ratings. Results Uncertainty At the individual level of analysis, experienced uncertainty was similar across the suspicion (M = 2.94, SD = 1.65) and trust conditions (M = 2.49, SD = 1.75), F(1, 56) = .78, ns. Value claiming A one-sample t-test showed that the suspicious party in the Suspicious–Trusting dyads did not significantly claim more (MSuspicious = 51.4%, SD = 3.2%) than 50% of the value created (ns). Value creation by dyads ANOVA showed that Suspicious–Trusting dyads created more value (M = 19,082, SD = 1210) than did Trusting–Trusting dyads (M = 18,213, SD = 1044, F(1, 31) = 4.66, p < .05). Mediation by information search Finally, I examined whether the suspicious party’s information search mediated the difference between Suspicious–Trusting and Trusting–Trusting dyads in value creation. I followed Baron’s and Kenny’s (1986) procedure. The results of the regressions are presented in Fig. 2. These indicate that information search by the suspicious party in Suspicious–Trusting dyads mediated the greater value creation of Suspicious–Trusting dyads compared to Trusting–Trusting dyads. A Sobel test that examined the reduced effect of the difference between dyads was significant, z = 1.95, p = .05. Supplementary analyses revealed that information search by the trusting party in Suspicious–Trusting dyads did not mediate the value creation effect (the trusting parties’ information search was similar across dyads).

549

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

Information search by the suspicious party β = .41 *

β = .66 *** / β = .58 **

Suspicious-Trusting dyads

Value

vs. Trusting-Trusting dyads

creation β = .40 * /

β = .17 ns

Fig. 2. Experiment 2b: mediation by information search. Note: Dyads were randomly assigned to Suspicious–Trusting dyads (N = 17) or Trusting–Trusting dyads (N = 15). * p < .05, **p < .01, ***p < .001.

General discussion Four experiments supported the argument that suspicion, defined as ambiguity about a target’s motives, can present greater benefits than trust for generating information search and creating value in negotiation. Experiments 1a and 1b showed that suspicion was characterized by more conscious motive attributions (Experiment 1a) and induced greater propensity for information search (Experiment 1b). Experiments 2a and 2b showed these cognitive differences had behavioral consequences in face-to-face negotiations. Suspicious–Trusting dyads created more value than did Trusting–Trusting dyads (Experiments 2a and 2b). The suspicious party’s information search mediated this effect (Experiment 2b). The current results extend the research on motive attributions to negotiation. Motives are key to negotiation dynamics (Carnevale & De Dreu, 2006). Granted negotiators often make dispositional attributions, which shape negotiation process and outcomes (Morris, Larrick, & Su, 1999). Yet, people also attribute motives to others, such as purposes and needs (Ames, Flynn, & Weber, 2004; Malle, 1999; Malle et al., 2000; Reeder, 2009; Reeder, Vonk, Ronk, Ham, & Lawrence, 2004; Struthers, Eaton, Santelli, Uchiyama, & Shirvani, 2008). Motive attributions affect person perception (Reeder, 2009; Reeder et al., 2004) and attitude (Ames et al., 2004; Struthers et al., 2008). The current research is the first to suggest that they impact behavior. The positive effect of suspicion over trust supports the general theory that motivated information processing augments value creation in negotiation (De Dreu et al., 2000, 2006): suspicion may be deemed a specific occurrence of motivated information processing as it involves effortful information search. Ambiguity about another’s motives provides a motivation for wanting to be more accurate about these. Additionally, while caution is needed when interpreting null-results, it seemed that suspicion did not affect impasses or value claiming (Experiments 2a and 2b), suggesting that suspicion as a state that motivates information seeking is independent of being competitive or Machiavellian. Incidentally, this further points to suspicion departing from distrust since distrust leads to being competitive (see Deutsch, 1958; Kimmel, Pruitt, Magenau, KonarGoldband, & Carnevale, 1980). Two concerns may limit the current findings, however. Initially suspicious perceivers may not remain in this state. They may come to a conclusion about their counterpart’s motives eventually. Hence, suspicion may be a transient, perhaps uncomfortable, state that is eventually replaced by distrust or trust. In addition, suspicion might induce (see Marchand & Vonk, 2005) distrust in the target of the suspicious perceiver over time. Nonetheless, the manipulation of suspicion as an initial state allowed causal inferences, and was consistent with other research (Kimmel et al., 1980; also, Galinsky et al., 2008; Sinaceur & Tiedens, 2006).

Trust is often useful in social interactions (Kramer, 1999). In negotiation trust has been deemed valuable for value creation (Pruitt, 1981). Indeed, research consistently demonstrated that trust generates higher value creation in negotiation than distrust (De Dreu et al., 2006). Yet trust, compared to suspicion, may also induce carelessness in seeking information. By separating between suspicion and distrust and showing that suspicion elicits greater information search than trust, the current research is the first to document that trust might limit value creation. Through greater information search suspicion may generate greater value creation in negotiation than trust. Acknowledgments I thank Maggie Neale for her numerous suggestions. I also owe to Jamie Barden, Max Bazerman, Byron Bland, Adam Galinsky, Deb Gruenfeld, Chip Heath, Rod Kramer, Bertram Malle, Liz Mullen, Bob Sutton, and Lara Tiedens for their insightful comments. Appendix A AlphaPharma payoff schedule

BioPharma payoff schedule Points

Points R&D Investment 40% made by AlphaPharma/60% made by BioPharma 45% made by AlphaPharma/55% made by BioPharma Equally shared 55% made by AlphaPharma/45% made by BioPharma 60% made by AlphaPharma/40% made by BioPharma

4000

0

3000

500

2000 1000

1000 1500

0

2000

2000 1500

0 1000

1000 500

2000 3000

0

4000

3200

0

2400

200

Manufacturing facilities Use BioPharma facilities Use BioPharma facilities 70% and AlphaPharma’s 30% Use both firms’ facilities equally Use AlphaPharma facilities 70% and BioPharma’s 30% Use AlphaPharma facilities Marketing strategy Marketing efforts focused on hospitals Marketing efforts focused on doctors

(continued on next page)

550

M. Sinaceur / Journal of Experimental Social Psychology 46 (2010) 543–550

Appendix A (continued)

AlphaPharma payoff schedule

BioPharma payoff schedule Points

Points Marketing efforts focused on healthcare insurance firms Marketing efforts focused on pharmacies and drugstores Marketing efforts focused on consumers

1600

400

800

600

0

800

800

0

600

800

400

1600

200

2400

0

3200

to

6000

0

to

4500

1500

to

3000 1500

3000 4500

to

0

6000

0 300 600 900 1200

0 300 600 900 1200

Workforce All workforce provided by BioPharma 70% Workforce provided by BioPharma Workforce provided by both firms equally 70% Workforce provided by AlphaPharma All workforce provided by AlphaPharma Profit sharing 70% to AlphaPharma/30% BioPharma 60% to AlphaPharma/40% BioPharma Profit split equally 40% to AlphaPharma/60% BioPharma 30% to AlphaPharma/70% BioPharma Sales coverage Global coverage USA–Asia USA–Latin America USA–Japan–European Union USA–Japan–France–Germany– UK

Note: Parties saw only their own payoff schedules and were not permitted to exchange them.

References Ames, D. R., Flynn, F. J., & Weber, E. U. (2004). It is the thought that counts: On perceiving how helpers decide to lend a hand. Personality and Social Psychology Bulletin, 30, 461–474. Baron’s, R. M., & Kenny’s, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. Carnevale, P. J., & De Dreu, C. K. W. (2006). Motive: The negotiator’s raison d’être. In L. L. Thompson (Ed.), Negotiation theory and research (pp. 55–76). Madison, CT: Psychosocial Press. De Dreu, C. K. W., Beersma, B., Stroebe, K., & Euwema, M. (2006). The interaction between social motives and epistemic motives in negotiation. Journal of Personality and Social Psychology, 90, 927–943. De Dreu, C. K. W., Giebels, E., & Van de Vliert, E. (1998). Social motives and trust in integrative negotiation: The disruptive effects of punitive capability. Journal of Applied Psychology, 83, 408–422. De Dreu, C. K. W., Koole, S. L., & Steinel, W. (2000). Unfixing the fixed pie: A motivated information-processing approach to integrative negotiation. Journal of Personality and Social Psychology, 79, 975–987. De Dreu, C. K. W., & Van Kleef, G. A. (2004). The influence of power on the information search, impression formation, and demands in negotiation. Journal of Experimental Social Psychology, 40, 303–319. Descartes, R. (1637/2006). A discourse on the method. New York, NY: Oxford University Press.

Deutsch, M. (1958). Trust and suspicion. Journal of Conflict Resolution, 2, 265–279. Fein’s, S. (1996). Effects of suspicion on attributional thinking and the correspondence bias. Journal of Personality and Social Psychology, 70, 1164–1184. Fein, S., & Hilton, J. L. (1994). Judging others in the shadow of suspicion. Motivation and Emotion, 18, 167–198. Fein, S., Morgan, S. J., Norton, M., & Sommers, S. R. (1997). Hype and suspicion: The effects of pretrial publicity, race, and suspicion on jurors’ verdicts. Journal of Social Issues, 53, 487–502. Galinsky, A. D., Maddux, W. W., Gilin, D., & White, J. B. (2008). Why it pays to get inside the head of your opponent: The differential effects of perspective-taking and empathy in strategic interactions. Psychological Science, 19, 378–384. Hilton, J. L., Fein, S., & Miller, D. T. (1993). Suspicion and dispositional inference. Personality and Social Psychology Bulletin, 19, 501–512. Howell, D. C. (1997). Statistical methods for psychology (4th ed.). Belmont, CA: Wadsworth. Hume, D. (1748/2000). Enquiry concerning human understanding. New York, NY: Oxford University Press. Kimmel, M. J., Pruitt, D. G., Magenau, J. M., Konar-Goldband, E., & Carnevale, P. J. D. (1980). Effects of trust, aspiration, and gender on negotiation tactics. Journal of Personality and Social Psychology, 38, 9–22. Kramer, R. M. (1998). Paranoid cognition in social systems: Thinking and acting in the shadow of doubt. Personality and Social Psychology Review, 2, 251–275. Kramer, R. M. (1999). Trust and distrust in organizations: Emerging perspectives, enduring questions. Annual Review of Psychology, 50, 569–598. Kray, L. J., Reb, J., Galinsky, A. D., & Thompson, L. (2004). Stereotype reactance at the bargaining table: The effect of stereotype activation and power on claiming and creating value. Personality and Social Psychology Bulletin, 30, 399–411. Lewicki, R. J., & Bunker, B. B. (1996). Developing and maintaining trust in work relationships. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 114–139). London: Sage. Maddux, W. W., Mullen, E., & Galinsky, A. G. (2008). Chameleons bake bigger pies and take bigger pieces: Strategic behavioral mimicry facilitates negotiation outcomes. Journal of Experimental Social Psychology, 44, 461–468. Malle, B. F. (1999). How people explain behavior: A new theoretical framework. Personality and Social Psychology Review, 3, 23–48. Malle, B. F., Knobe, J., O’Laughlin, M., Pearce, G. E., & Nelson, S. E. (2000). Conceptual structure and social functions of behavior explanations: Beyond person– situation attributions. Journal of Personality and Social Psychology, 79, 309–326. Marchand, M. A. G., & Vonk, R. (2005). The process of becoming suspicious of ulterior motives. Social Cognition, 23, 242–256. Morris, M. W., Larrick, R. P., & Su, S. K. (1999). Misperceiving negotiation counterparts: When situationally determined bargaining behaviors are attributed to traits. Journal of Personality and Social Psychology, 77, 52–67. Parks, C. D., Henager, R. F., & Scamahorn, S. D. (1996). Trust and reactions to messages of intent in social dilemmas. Journal of Conflict Resolution, 40, 134–151. Pruitt, D. G. (1981). Negotiation behavior. New York, NY: Academic Press. Pruitt, D. G., & Kimmel, M. J. (1977). Twenty years of experimental gaming: Critique, synthesis, and suggestions for the future. Annual Review of Psychology, 28, 363–392. Reeder, G. D. (2009). Mindreading: Judgments about intentionality and motives in dispositional inference. Psychological Inquiry, 20, 1–18. Reeder, G. D., Vonk, R., Ronk, M. J., Ham, J., & Lawrence, M. (2004). Dispositional attribution: Multiple inferences about motive-related traits. Journal of Personality and Social Psychology, 86, 530–544. Rubin, J. Z., & Brown, B. R. (1975). The social psychology of bargaining and negotiation. New York, NY: Academic Press. Schul, Y., Burnstein, E., & Bardi, A. (1996). Dealing with deceptions that are difficult to detect: Encoding and judgment as a function of preparing to receive invalid information. Journal of Experimental Social Psychology, 32, 228–253. Sinaceur, M., & Tiedens, L. Z. (2006). Get mad and get more than even: When and why anger expression is effective in negotiations. Journal of Experimental Social Psychology, 42, 314–322. Struthers, C. W., Eaton, J., Santelli, A. G., Uchiyama, M., & Shirvani, N. (2008). The effects of attributions of intent and apology on forgiveness: When saying sorry may not help the story. Journal of Experimental Social Psychology, 44, 983–992. Suber, P. (1996). Classical skepticism: Issues and problems. Philosophy Department, Earlham College. Available from http://www.earlham.edu/~peters/writing/ skept.htm#epoche. Thompson, L. L. (1991). Information exchange in negotiation. Journal of Experimental Social Psychology, 27, 161–179. Thompson, L. L., & Hastie, R. (1990). Social perception in negotiation. Organizational Behavior and Human Decision Processes, 47, 98–123. Tiedens, L. Z., & Linton, S. (2001). Judgment under emotional certainty and uncertainty: The effects of specific emotions on information processing. Journal of Personality and Social Psychology, 81, 973–988. Weary, G., Jacobson, J. A., Edwards, J. A., & Tobin, S. J. (2001). Chronic and temporarily activated causal uncertainty beliefs and stereotype usage. Journal of Personality and Social Psychology, 81, 206–219. Weary, G., Vaughn, L. A., Stewart, B. D., & Edwards, J. A. (2006). Adjusting for the correspondence bias: Effects of causal uncertainty, cognitive busyness, and causal strength of situational information. Journal of Experimental Social Psychology, 42, 87–94.