Case Study
scm41
Source-Forgas.pdf
ATTITUDES AND SOCIAL COGNITION
On Feeling Good and Getting ^bur Way: Mood Effects on Negotiator Cognition and Bargaining Strategies
Joseph P. Forgas University of New South Wales
Are happy people more likely to be cooperative and successful negotiators? On the basis of the Affect Infusion Model (ATM; Forgas, 1995a), Experiment 1 predicted and found that bodi good and bad moods had a significant mood-congruent effect on people's thoughts and plans, and on their negotiation strategies and outcomes in both interpersonal and intergroup bargaining. Experiment 2 replicated these results and also showed that mood effects were reduced for persons more likely to adopt motivated processing strategies (scoring high on machiavellianism and need for approval). Experiment 3 confirmed these effects and demonstrated that the mood of the opposition also produced more mood-congruent bargaining strategies and outcomes. The results are discussed in terms of affect priming influences on interpersonal behaviors, and the implications of these findings for real- life cognitive tasks and bargaining encounters are considered.
Conflict is an ever present feature of social life, and face-to- face negotiation is perhaps the most common and constructive method for resolving differences between people and between groups. Negotiation may be defined as a "discussion between two or more parties aimed at resolving incompatible goals" (Pruitt & Carnevale, 1993, p. xv). Although psychologists have long been interested in negotiation (Deutsch, 1973; Rubin & Brown, 1975), the influence of affect on negotiator cognition and bargaining has received surprisingly little attention to date, Recent cognitive approaches to negotiation suggest that biased perceptions and assumptions are a major cause of suboptimal bargaining strategies and outcomes (Neale & Bazerman, 1991; Thompson, 1990, 1995). These three experiments extended re- search on mood effects on cognition to the domain of interper- sonal behavior {Argyle, 1991), and explored the impact of short-term good or bad moods on people's plans, expectations, and reported performance in two consecutive negotiation tasks: an informal, interpersonal bargaining and a formal, intergroup negotiation (cf. Keenan & Carnevale, 1989).
Negotiator Cognition
Negotiation traditionally has been conceptualized by psy- chologists as a rational process, in which exchanges are guided
This project was supported by a Special Investigator Award from the Australian Research Council and by a Research Prize from the Alexander von Humboldt Foundation, Germany. I gratefully acknowledge the con- tribution of Stephanie Moylan, Joan Webb, and Joseph Ciarrochi.
Correspondence concerning this article should be addressed to Joseph P. Forgas, School of Psychology, University of New South Wales, Sydney 2052, Australia. Electronic mail may be sent [email protected]. Website: http://www.psych.unsw.edu.au/~joef/jforgas.htm.
by an enlightened attempt to gain the best possible outcome for the self by incurring as little cost as possible (Deutsch, 1994; Thompson, 1990). However, this rational view suffers from several shortcomings. By definition, not all participants in a negotiation can achieve the optimum outcome for themselves. This introduces an element of unpredictability, allowing a range of cognitive, motivational, and emotional factors to in- fluence negotiation processes. Expectations, often based on erroneous or incomplete assumptions, can play a critical role in negotiations and can be a major source of dysfunctional bargaining strategies (Bazerman & Carroll, 1987; Rubin, Kim, & Peretz, 1990; Thompson, 1990). As negotiators also suffer from "bounded rationality" (Kruglanski, 1989; March & Simon, 1958), their positive or negative thoughts and associations primed when contemplating a negotiation en- counter can exert a significant impact on their plans and bar- gaining strategies (Neale & Bazerman, 1991). Despite strong recent evidence for mood effects on many cognitive processes (Bower, 1991; Fiedler, 1991; Forgas, 1995a, 1995b; Isen, 1987; Mayer, Gaschke, Bravennan, & Evans, 1992; Sedikides, 1994, 1995), little is known about how mood affects negotiator cog- nition and bargaining strategies. These experiments seek to link cognitive theories of negotiation with research on affective influences on social judgments, in order to explore how moods may affect people's thoughts, plans, expectations, and bar- gaining strategies.
Strategies of Negotiation
Negotiation typically consists of a sequence of proposals and counterproposals that continue until agreement is reached or negotiations break down (Deutsch, 1973; Pruitt & Carnevale, 1993 ). Strategic moves within a negotiation can be conveniently classified into a few typical approaches. For example, Thomas
Journal of Personality and Social Psychology, 1998, VoL 14, No % 565-517 Copyright 1998 ljy the American Psychological Association, Inc. 0O22-3514/98/$3.0O
565
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
566 FORGAS
(1990) suggested that competition, collaboration, compromise, accommodation, and avoidance are examples of such basic ne- gotiating moves. In a similar vein, Pruitt and Carnevale (1993) identify concession making, contending, problem solving, inac- tion, and withdrawal as the five basic strategies available to negotiators. Others, such as Rahim (1983), take a more personal perspective in referring to dominating, obliging, integrating, compromising, and avoiding strategies. The critical feature in all such taxonomies is a distinction between cooperation and competition as fundamentally different psychological ap- proaches. Affect is strongly implicated in this dichotomy, ac- cording to Lanzetta (1989), who argued that * 'cooperative situa- tions involve shared, or empathetic emotional experiences and . . . competitive situations involve mutually exclusive, or coun- terempathetic emotional experiences" (p. 543). Thus, competi- tion seems linked to a heightened "sensitivity to differences and threats," whereas cooperation is associated with "similarities and common interests" (Deutsch, 1973, p. 2 9 ) . In addition to this cooperative-competitive dimension, more passive, interme- diate strategies have also been identified, such as "tit for tat" (a move-by-move reciprocation), and an indeterminate, random strategy (Rubin & Brown, 1975). The effects of mood on the planned and actual use of these four kinds of negotiating tactics are assessed here: cooperation, competition, tit for tat, and ran- dom moves.
Affect and Cognition
What are the mechanisms likely to mediate the influence of affect on negotiator cognition and bargaining strategies? Early theories of mood effects on thinking emphasized either condi- tioning principles, or psychoanalytic mechanisms, such as pro- jection (Feshbach & Singer, 1957). In contrast, contemporary explanations are predominantly cognitive (Bower, 1991; Fiedler, 1991; Forgas, 1995a, 1995b). Two kinds of mood effects on cognition have been identified: ( a ) informational effects occur when mood informs what people think through the greater avail- ability and use of affectively colored information (the content of cognition; Bower, 1991), and ( b ) processing effects occur when mood influences how people think (the process of cogni- tion). Evidence for the informational effects of mood comes from numerous studies that have shown a strong bias toward mood-consistent information in attention, memory, learning, and associative processes due to the operation of affect priming processes (R. A. Baron, 1990; Bower, 1991; Forgas & Bower, 1987; Mayer et al., 1992; Sedikides, 1995). Similar cognitive processes should also influence the way negotiators perceive and interpret a bargaining problem, and plan their negotiating strategies. Occasionally, heuristic processing can also produce mood-congruent judgments, which occur when people misattri- bute their mood as information or input in responding. However, such simple, heuristic strategies are unlikely to be relevant to formulating complex bargaining plans.
T h e Affect Infusion Model
The various informational and processing consequences of mood were recently integrated in a comprehensive, multiprocess Affect Infusion Model (AIM; K>rgas, 1992a, 1995a). The model suggests that affect infusion into thinking and judgments
is most likely when people adopt an open, constructive pro- cessing strategy, such as ( a ) systematic or substantive pro- cessing, or (b) heuristic processing. In contrast, affect infusion is unlikely when individuals use a targeted, predetermined infor- mation search strategy, such as (c) direct access of stored infor- mation or (d) motivated processing in the service of specific goals (Forgas, 1995a; Rusting, 1997). According to the AIM, the adoption of a processing style depends on a combination of factors such as the novelty, complexity, and salience of the task, and the personality, motivation, affective state, and cognitive capacity of the judge. The AIM can account for much of the available empirical evidence on mood effects on cognition and judgments to date, including some counterintuitive findings showing greater affect priming and affect infusion when more extensive, elaborate processing strategies are adopted in re- sponse to complex, demanding cognitive tasks (Fiedler, 1991; Forgas, 1992b, 1994, 1995b; Sedikides, 1995).
M o o d Effects on Negotiation
If one extrapolates from this research, several lines of evi- dence suggest that feeling good or feeling bad should also have important consequences for negotiator cognition and strategies (Lanzetta, 1989). Recent studies have shown that when one is faced with a complex task, such as selecting and evaluating a partner or a relationship, explaining a conflict, or allocating rewards to a group, good or bad mood has a significant priming effect on how people interpret the problem, the kind of informa- tion they use, and the kind of processes they adopt in con- structing a response (Forgas, 1992a, 1992b, 1995a, 1995b; For- gas & Fiedler, 1996). Further, people in a good mood also perform better on creative problem-solving tasks, demonstrate lower levels of anger and hostility, are more altruistic, optimistic, and flexible, and are also more inclined to be helpful (Rargas, Bower, & Moylan, 1990; Hertel & Fiedler, 1994; Isen, 1987). These mood effects should be especially marked when elaborate, substantive processing is required to deal with a complex, inde- terminate task, such as planning a bargaining encounter. It also appears that experiences of cooperation or competition selec- tively trigger positive, empathic, or negative, counterempathic feelings and evaluations of others (Lanzetta, 1989; Thompson, 1995).
Surprisingly, direct evidence for mood effects on bargaining strategies is far more limited. In one experiment, Carnevale and Isen (1986) found that in conditions requiring high achievement, individuals experiencing positive affect were less likely to adopt contentious negotiating strategies, and attained higher outcomes than did others. In another interesting study, R. A. Baron (1990) showed that individuals experiencing good moods due to the presence of a pleasant scent set higher goals, made more conces- sions, and were less likely to prefer avoidance or competition in face-to-face negotiations. Several important issues about the effects of mood on negotiation remain unanswered in the litera- ture, however. The role of negative moods on negotiation strate- gies and outcomes has received little attention in the past and is evaluated here. The processes mediating mood effects on negotiation also remain incompletely understood. Further, the additional role of the opposition's mood in facilitating or hinder- ing integrative bargaining strategies is also explored here (Ex- periment 3 ) .
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 567
Because of affect priming effects during substantive pro- cessing, it is assumed here that moods should exert a mood- congruent influence on negotiators' thoughts, expectations, and plans that will eventually influence their negotiating strategies and outcomes (Rubin et al., 1990; Thompson, 1990). Accord- ingly, mood effects on prenegotiation plans should play a sig- nificant role in mediating subsequent bargaining strategies and outcomes, a prediction specifically evaluated in Experiment 1. It is also unclear whether happy persons may be more effective in bargaining situations because they use more cooperative and integrative strategies, or because they are more confident and ambitious in their expectations and set themselves higher goals (R. A. Baron, 1990; Neale & Bazerman, 1991). Recent work by Hertel and Fiedler (1994) also suggested that elated mood may increase behavioral flexibility rather than positivity or co- operation per se. Thus, positive mood either may directly affect the level of cooperation, or may indirectly influence negotia- tion through generating increased cognitive and behavioral flexibility.
A i m s and Hypotheses
In summary, the objective of these experiments was to investi- gate the impact of both good and bad moods on prenegotiation plans, and on subsequent reports of bargaining strategies and outcomes in both interpersonal and intergroup negotiations. In terms of the AIM, people facing an unpredictable and uncertain negotiation task were expected to adopt a generative, substantive processing strategy when interpreting the bargaining encounter, allowing affect infusion to influence how they construe the task, the kind of expectations and goals they form, and the kind of bargaining strategies they plan to adopt. However, some individ- uals may approach the task from a more controlled, motivated perspective. Experiment 2 hypothesized that mood effects would be attenuated for individuals scoring high on need for approval and machiavellianism, who are more likely to adopt motivated rather than substantive processing strategies in thinking about the encounter. Consistent with the AIM, past studies have sug- gested such a reduction in affect infusion when targeted, moti- vated processing is used (Berkowitz, 1993; Forgas, 1991; Erber & Erber, 1994; Sedikides, 1994). Finally, the affective state of the opposition was also expected to exert a significant and additive mood-congruent effect on negotiating strategies and outcomes.
Experiment 1
The first experiment explored the role of mood in people's negotiating strategies and outcomes in an interpersonal and an intergroup task. Happy persons should plan and later report using more cooperative and more integrative negotiating strate- gies because of the greater availability of positively valenced thoughts and associations. In contrast, people experiencing a bad mood should underestimate the likelihood of success and the cooperativeness of the partner, thus leading them to adopt more competitive and ultimately less successful negotiating strategies (Rubin et al., 1990). A mediational analysis explores the effects of mood on prenegotiation plans, and on postnegotia- tion strategies and outcomes.
Method
Overview and participants. Individuals participated in what they believed were two unrelated experiments, (a) a study of verbal abilities (in fact, the mood induction) and (b) a study of interpersonal behavior (comprising the interpersonal and intergroup negotiation tasks). Mood was induced through manipulated feedback about performance on what was presented as a verbal abilities task. Participants were then randomly assigned into two teams, "students" and "faculty, " and engaged in (a) an informal, interindividual negotiation and (b) a formal, intergroup negotiation about an involving and realistic issue, their course prefer- ences in a revised curriculum. Planned and reported negotiation behav- iors were assessed both before and after each negotiation task. Partici- pants were 72 students (36 male, 36 female) who received course credit for participating and were randomly assigned to the experimental condi- tions. Because a preliminary analysis revealed no gender differences, data from men and women were combined.
Procedure. On arrival, participants were told that two separate ex- periments would be carried out to "save subject time." They next re- ceived instructions about the negotiation task, ostensibly while the first experiment on verbal abilities was set up. This allowed the negotiation task to commence immediately after the mood induction.
Mood manipulation. A previously validated false-feedback tech- nique was used to induce good or bad moods (Forgas, 1991). The task was described as a test of verbal abilities, asking participants to complete a series of 33 sentences of the kind "Car is to road as train is to . . . ." The sentences ranged from the easy to the difficult, with the last 8 items, although plausible, having no determinate answers (e.g., "Bread is to butter as river is to'. . . " ) . On average, 19 questions were completed in 5 min, with everybody completing at least 13 items in this time. In the negative condition, the cover sheet informed participants that the items were easy and most people completed the test in 5 min. In the positive condition, participants were informed that the test was difficult and few people completed more than 10 items in the time available.
On completion, participants were provided with a sheet containing feedback on the test. In the positive mood condition, completing between 7 and 13 questions was described as "average" and over 13 questions as "above average, indicating superior verbal ability and excellent mas- tery of verbal skills.'' For the negative mood group, average performance was described as 25-30 correct answers, with more than 30 correct answers as above average. Performance below 25 questions was labeled ' 'below average, indicating below average performance in this area of verbal ability and suggesting that these verbal skills have not been mas- tered." As in fact, all participants performed within the 14-23 range, everybody in the positive condition received above average marks and positive feedback, whereas those in the negative condition thought they performed poorly and received negative feedback. The control group completed the same verbal abilities test. However, they received no advance information about the expected difficulty, and were given no feedback on their performance. Instead, they were simply told that the test was still under development and were thanked for their help before progressing to the experimental negotiation task.
Negotiation task. The negotiation task concerned the selection of new courses for the psychology curriculum, a topic of direct interest to students in this study. At the beginning of the exercise, each person received an instruction sheet, informing them that they would act as one member of a three-person negotiating team, representing either students, or faculty. Allocations to the teams were randomized. Each participant was assigned a different course to bargain for (perception, learning, social psychology, etc.). Participants were told that their task was first to (a) negotiate within their team, trying to get the highest ranking for their course on their team's joint preference list, and (b) subsequently, to negotiate with the opposing team, trying to get the highest possible rank for their own team's preferred courses. Rules for the informal, interindividual negotiation task were left unspecified, allowing people
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
568 FORGAS
to use whatever strategies they wished. Rules for the formal, intergroup negotiation session stated that the bargaining was to take the form of six consecutive sessions, each designed to select one of the two compet- ing courses at the top of the priority lists of each team. Tf a course was not selected, it went into the next session as the team's highest priority choice. Each session was concluded with a vote. A majority vote was sufficient to place a course on the curriculum. Making deals, offering concessions, demanding concessions, and the breaking of these deals were all legitimate strategies.
Mood validation and debriefing. At the end of the task, participants completed a brief "postexperimental questionnaire" (in fact a mood manipulation check). Embedded among several distractor items (e.g., "Have you done similar tests before?"), they rated their postinduction mood on four 7-point bipolar scales: positive-negative, tense-relaxed, good-bad, and calm-aroused. A careful debriefing concluded the proce- dure, ensuring that no residual mood effects remained (Ross, Lepper, & Hubbard, 1975). There was no evidence of participant awareness of either the hypotheses, or the connection between the two parts of the study.
Dependent measures. Before and after the negotiation sessions, par- ticipants completed a brief questionnaire. Eight-point bipolar scales (1 = definitely no, 8 = definitely yes) were used to measure three sets of dependent variables, (a) Bargaining strategy: participants rated their planned (or actual) use each of four distinct bargaining strategies (coop- eration: support their choices and hope that they will support mine; competition: do not support their choices so that mine has a better chance; tit for tat: reciprocate whatever they do; and a random strategy: do whatever comes to mind), (b) Making deals: " D o you expect to make [have you made] deals within your team [with the opposing team]?" and "Do you expect to honor [have you honored] the deals made within your team [with the opposing team]?" (c) Expected and reported outcomes: " D o you expect to [Did you] get your way with members of your own team [the opposing team]?" and "What position do you expect [What position did] your preferred course [to] receive on your team's priority list?"
Results and Discussion
Mood validation. A principal-components analysis of the four mood validation scales showed that all four scales loaded on a single mood evaluation factor (Cronbach's a - .88). An analysis of variance of this combined measure showed a signifi- cant overall mood effect, F(2, 69) = 11.03, p < .01. As pre- dicted, participants in the positive condition rated their mood as significantly better, F(l, 70) - 7.58,/? < .01, whereas those in the negative mood condition rated their mood as worse than did controls (5.74 vs. 4.69 vs. 3.91), F(ly 70) = 4.89,p < .05. These results confirm that the mood induction was successful in producing different moods in the three experimental groups, as also found in previous research using this method (Forgas, 1995a).
The effects of mood on planned negotiation strategies. Coop- eration was overall preferred to competition in the informal, interpersonal negotiation, F ( l , 70) = 14.11, p < .01, and com- petition was preferred to cooperation in the more formal, in- tergroup encounter, F ( l , 70) = 5.93, p < .05. This seems to reflect the more antagonistic character of formal, intergroup bargaining encounters.
More important, mood had the predicted influence on partici- pants' plans and expectations. Plans to be cooperative, F(2,69) = 9.77, p < .05, competitive, F ( 2 , 69) = 7.43, p < .01, to make deals, F{2, 69) = 5.88,p < .05, and to honor deals, F ( 2 , 69) — 7.31, p < .01, were all significantly influenced by mood
before the first interpersonal negotiation task. Compared with controls, those in a happy mood were more likely, f ( l , 46) = 5.19, p < .05, and sad persons were significantly less likely, F( 1,46) = 4.76, p < .05, to plan to use a cooperative strategy. In contrast, plans to adopt a competitive bargaining strategy were significantly reduced by happy mood, F{\, 46) = 5.37, p < .05, and increased by sad mood compared with controls, F( 1, 46) - 4.16, p < .05, (Figure 1). Consistent with their more cooperative planned strategies, happy participants were also more likely to plan to make deals, F( 1, 46) = 5.44, p < .05, and to honor deals, F{ 1,94) - 4.58,p < .05, than were controls. Participants in a sad mood showed an opposite trend, although these differences failed to reach significance (Figure 1).
Mood also had a significant influence on participants' plans and expectations for the second, intergroup negotiation task, including their plans to be cooperative, F(2, 69) = 10.33, p < .01, as well as their plans to use competitive strategies, F(2, 69) - 6.01, p < .05, and to make, F ( 2 , 69) = 4.22, p < .05, and subsequently honor deals, F ( 2 , 69) = 6.35, p < .05. Happy persons planned to be more cooperative than were controls, F( 1, 46) = 4.35, p < .01, and sad persons expected to be less cooperative, F(l, 46) = 5.02, p < .05. Sad mood in turn in- creased intended competitiveness relative to controls, F{\, 46) - 4.14, p < .05, whereas happy persons were less competitive, F{ 1, 46) = 4.99, p < .05. There was also a nonsignificant trend (Figure 1) for happy persons to make and to honor deals more than did controls, and for sad participants to be less inclined to make and to honor deals than were controls.
The expectation by happy people to adopt more cooperative, integrative negotiating strategies indicates the predicted affect infusion into their bargaining plans. These results extend previ- ous mood research to the new domain of negotiation strategies. In terms of the AIM, the more optimistic assessment of the bargaining encounter by happy persons is consistent with the
PLANNED REPORTED
2.2 Happy Control Sad Happy Control Sad
MOOD STATE
Figure 1. The influence of positive, control, and negative mood on planned and reported negotiating strategies in an informal, interpersonal (top panel) and in a formal, intergroup (bottom panel) bargaining en- counter (COOP = cooperation; COMP = competition; DEALS = mak- ing deals; HONOR = honoring deals).
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 569
priming and infusion of affectively valenced information into their deliberations in the course of constructive, substantive pro- cessing (Fiedler, 1991; Forgas, 1995a). Given the constructive, elaborate nature of formulating negotiating plans, heuristic af- fect- as-information or affect-as-input mechanisms are unlikely to provide an alternative explanation for mood congruency in this case. To what extent did mood effects extend to reported negotiation strategies? We shall turn to this question next.
Mood effects on reported bargaining strategies. Postnegotia- tion reports confirmed that cooperation was overall preferred to competition in the informal, interpersonal negotiation, F(1,70) = 6.96, p < .05, whereas there were no differences in the use of these two strategies in the formal, intergroup encounter (Fig- ure 1). As predicted, mood also had a significant influence on the strategies used. After the informal, interpersonal encounter, mood influenced the use of cooperation, F ( 2 , 69) = 10.42, p
< .01, competition, F ( 2 , 69) = 5.97, p < .05, making deals, F ( 2 , 69) = 8.11, p < .01, and honoring deals, F ( 2 , 69) = 6.08, p < .05. Happy persons were more cooperative, F( 1, 46) = 7.27, p < .01, and made more deals, F ( l , 46) - 6.09, p < .05, than did controls, and sad individuals were somewhat more competitive than controls, F(l, 46) - 4.42, p < .05 (Figure I ) . Following the formal intergroup negotiation, mood had a similar influence on reported cooperation, F ( 2 , 69) = 13.33, p
< .01, competition, F ( 2 , 69) = 7.05, p < .01, and making deals, F ( 2 , 69) = 5.23, p < .05. Happy persons reported a tendency to be more cooperative, F ( l , 46) = 9.63, p < .01, and less competitive than did controls, F ( l , 46) = 5.82, p < .05. In contrast, negative mood resulted in less cooperation, F ( l , 94) = 4.52, p < .05, and less deal making, F ( l , 46) = 4,45, p < .05, compared with controls. These results appear to be remarkably robust, as indicated by the consistency of the findings across both the planned and reported strategies used, and across both the interpersonal and the intergroup negotiation tasks.
Mood effects on bargaining outcomes. Mood also had a significant influence on expected bargaining outcomes, F ( 2 , 6 9 ) = 11.36, p < .01: Happy persons predicted a higher overall rank for their preferred courses than did controls, F ( l , 46) - 7.01, p < .01, who in turn were more optimistic than sad persons (4.72 vs. 3.83 vs. 2.95), F ( l , 46) = 6.82,p < .05. It is interest- ing to note that these mood effects on bargaining self-efficacy were ultimately reflected in reported outcomes in both the infor- mal, F ( 2 , 69) = 6.73, p < .05, and the formal negotiating encounter, F ( 2 , 69) = 5.28, p < .05. Happy persons achieved better outcomes in the interpersonal negotiation than did con- trols, F(l, 46) = 5.22, p < .05, who did somewhat better than sad persons, although these differences were not significant (4.51 vs. 3.66 vs. 3.38). Good mood also resulted in better outcomes at the end of the intergroup bargaining session, F ( l , 46) = 6.55, p < .05.
These results show that happy people were not simply making more concessions but used more cooperative, integrative, and ultimately more successful bargaining strategies. Because bar- gaming outcomes in the intergroup negotiation are partly deter- mined by the ranking the course receives in the within-group negotiation, these two outcomes are not entirely independent events (Keenan & Camevale, 1989). Nevertheless, these find- ings present a coherent pattern, indicating that positive mood had a clear beneficial influence on participants' expected and
reported negotiating strategies and outcomes, both in the infor- mal, interpersonal and in the formal, intergroup negotiation tasks.
Mediational analyses. In the final step, a mediational analy- sis was performed using multiple regression procedures to test that it was mood effects on planned strategies that mediated subsequent differences in bargaining strategies and outcomes. Ttoo sets of analyses were performed, using the following four variables: the independent variable was mood (as rated by each participant on the mood validation scales); the mediating vari- able was planned strategies (average of planned use of coopera- tion, noncompetition, and making and honoring deals in the two encounters; Cronbach's a = .79); and the dependent variables were reported use of integrative strategies (as defined above, Cronbach's a = .76) and bargaining outcome (average rank achieved for preferred course).
To test the predicted pattern of mediation, three regression analyses were performed (R. M. Baron & Kenny, 1986). First, the independent variable (mood) was used to predict the media- tor (planned strategies). Second, the independent variable (mood) was used to predict the dependent variables (reported strategies and outcomes). Third, the independent variable (mood) and the mediator (planned strategies) were simultane- ously entered into a regression analysis to predict the dependent variables. On the basis of these analyses, standardized regres- sion coefficients were generated and tested for significance. To establish mediation, all three regression analyses should have yielded significant results (p < .05), as was indeed the case here (Figure 2 ) . Both planned strategies (/? - .53) as well as reported strategies (/? = .46) and outcomes (/? = .42) were significantly predicted by mood. Further, planned strategies sig- nificantly predicted both reported strategies and outcomes (/5 — .56 and .37). It is also critical to "examine not only the significance of the coefficients, but also their size" (R. M. Baron & Kenny, 1986, p. 1177). If mediation occurs, the effects of the independent variable on the dependent variables must be less in the third equation (when the mediator is also present) than in the second equation (when the mediator is absent). Such a pattern was obtained here (Figure 2 ) : The effects of mood on reported bargaining strategies and outcomes were markedly reduced when the mediator variable (planned strategies) was also included in the analysis (/? = .46 vs. .12, and 0 = .42 vs. .09).
These findings are consistent with the prediction that good mood enhanced and negative mood reduced integrative bar- gaining and outcomes through influencing prenegotiation plans and expectations. Postnegotiation reports confirmed that cooper- ation and integrative strategies (such as deal making) were more commonly used by those in a good mood. Seeking cooperative solutions is the optimum strategy producing superior outcomes in many social conflicts, as was also found here (Pruitt & Car- nevale, 1993; Thompson, 1995). As positive feedback on the mood induction task produced more cooperation rather than more competition here, these results do not support increased assertiveness or increased flexibility as the cause of bargaining differences. Could it be that mood influenced verbal reports rather than actual negotiating strategies? The fact that mood effects on planned and reported bargaining strategies were very similar even though mood must have been dissipated by the time postnegotiation reports were collected seems incompatible with
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
570 FORGAS
PREDICTOR MEDIATING PREDICTED Method
Figure 2. Mediational analysis showing significant (p < .05) links (with standardized regression coefficients in parentheses) between mood (predictor variable), planned negotiating strategies (mediating variable) and reported negotiating strategies and outcomes (dependent variables).
this interpretation. More critically, differences in bargaining out- comes as well as the mediational analyses support the existence of genuine mood effects on the negotiating strategies used.
Experiment 2
Experiment 1 showed that good or bad moods have a marked impact on how people think about, plan, and perform bargaining behaviors. However, more evidence is needed to firmly establish these effects. The first objective of Experiment 2 was to replicate and extend the pattern of findings obtained in Experiment 1. Experiment 2 also sought to explore the role of individual differ- ences such as need for approval or machiavellianism in mediat- ing mood effects on negotiation. There is growing consensus that an adequate psychological understanding of negotiation should consider personality and individual differences between negotiators (Thompson, 1990). Recent research on affect and cognition also suggests that individual differences can produce motivated processing and can mediate mood effects on social cognition and behavior (Kruglanski, 1989; Mayer & Salovey, 1988; Salovey & Mayer, 1990; Smith & Petty, 1995). The AIM (Forgas, 1995a) specifically predicts a reduction in affect infu- sion when information processing is dominated by a trait-based motivational objective. Two individual-difference variables were examined here: need for approval (Crowne & Marlowe, 1964) and machiavellianism (Christie, 1968). High scorers on these measures were expected to approach the bargaining task from a preexisting motivated perspective, such as the motivation to be successful manipulators (high machiavellians) or the motiva- tion to obtain approval (high social desirability scores), limiting the extent to which affect will infuse their construal of the bargaining encounter according to the AIM.
Overview and participants. The same method and procedure were used as in Experiment 1. Individuals were exposed to a false-feedback mood induction, followed by the two negotiation encounters. On a sepa- rate occasion some weeks after the negotiation task, participants com- pleted the Social Desirability Scale (Crowne & Marlowe, 1964) and the Machiavellianism Scale (Christie, 1968). Because Experiment 1 already established that both positive and negative moods produce sig- nificant differences compared with the control group, only positive and negative mood groups were used here. Planned and reported negotiation behaviors and outcomes were again assessed both before and after each negotiation task. Participants were 132 students of both genders ran- domly assigned to the different experimental conditions (63 positive, 69 negative mood), who participated in the study as part of their course requirements. As a preliminary analysis revealed no gender differences, data from men and women were analyzed joindy.
Procedure. The negotiation task and instructions were the same as described in Experiment 1. Participants were randomly assigned to three- person teams of either staff or students. They first negotiated within their team, followed by an intergroup negotiation aiming to get the best possible ranking for their preferences. The mood manipulation check asked participants to rate their current mood on 7-point positive-negative scales, embedded among several distractor items (e.g., "Have you done similar verbal tests before?"). A thorough debriefing concluded the procedure, designed to ensure that no residual mood effects remained (Ross et al., 1975). We found no evidence of any awareness of the hypotheses by participants.
Individual-differences data. At a separate session several weeks later, the same group of participants completed two individual-difference measures, the Machiavellianism V Scale (Christie, 1968) and the Social Desirability Scale (Crowne & Marlowe, 1964). Because of difficulties in contacting all individuals, individual-differences data were obtained from 103 of the original 132 participants, 51 of whom were in the negative mood condition and 52 of whom were in the positive mood condition originally. Machiavellianism and need for approval were se- lected for study here as these measures are likely to be directly relevant to how people think about and actually behave in strategic social encoun- ters such as a bargaining situation. The Machiavellianism V Scale con- tains 20 items, each consisting of three alternatives, with participants indicating which statement is most and which is least like their view of the world. This scale measures a person's view of others and their preferred strategies of dealing with people, assessing the extent to which a person is likely to have a motivated, detached, and even cynical view of human nature, and is likely to see others as manipulable in social situations.
The Social Desirability Scale consists of 33 items requiring true or false responses. Half the statements are culturally acceptable but likely to be untrue, and the other half are likely to be true but are socially undesirable. This scale measures the extent to which individuals are motivated to seek approval, by acting and describing themselves in terms of favorable, socially desirable characteristics (Crowne & Marlowe, 1964). High scores on these scales are likely to indicate a more targeted, motivated information-processing strategy when approaching a complex social encounter such as a negotiation. These scales were administered in a group testing session, using standard instructions, with about 20 participants assessed in each group.
Results and Discussion
Mood validation. An analysis of variance of participants' self-rated mood on the positive-negative scale indicated a sig- nificant difference between the happy and the sad conditions, F ( l , 130) - 7.95, p < .01. As expected, participants in the positive condition rated their mood as significantly better than
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 571
did those in the negative condition (5.93 vs. 4.37), confirming the effectiveness of the mood induction procedure.
Mood effects on expected negotiation strategies. As in Ex- periment 1, cooperation was preferred to competition in the interpersonal negotiation, F( 1, 130) = 8.23, p < .01, and com- petition was preferred to cooperation in the intergroup encoun- ter, F ( l , 130) = 5.99, p < .05, indicating more antagonistic approaches in intergroup bargaining. As predicted, mood sig- nificantly influenced prenegotiation plans for both the interper- sonal and the intergroup bargaining tasks. Before the interper- sonal negotiation, happy participants were more likely than sad participants to plan to use a cooperative strategy, F ( l , 130) = 5.74, p < .05. In contrast, sad participants expected to be more competitive than did happy participants, F ( l , 130) — 5.86, p < .05 (Figure 3 ) . A similar pattern emerged for the second, intergroup negotiating task, with happy participants planning to be more cooperative, F ( l , 130) — 7.07,/?< .01, and sad partici- pants planning to be more competitive, F ( 1, 130) = 7.91, p < .01 (Figure 3 ) . Participants in a happy mood were also more likely to anticipate making deals in their interpersonal negotia- tion, F ( l , 130) = 3.99, p < .05, and to plan to honor deals once made, F( 1,130) = 7 . 8 6 , p < .01. In the intergroup encoun- ter, there was also a greater tendency for happy participants to plan to make deals, F(l, 130) = 6.75, p < .05, and to subse- quently honor such deals, F{\, 130) = 12.58, p < .01, when compared with sad participants. These findings indicate signifi- cant affect infusion effects into prenegotiations plans and expectations.
Mood effects on the reported use of negotiation strategies. After the interpersonal negotiation, happy persons reported us- ing more cooperative strategies, F ( l , 130) = 9.14, p < .01, and sad people relied more on competitive strategies, F( 1, 130) = 3.94, p < .05. For the second, intergroup negotiation, positive mood again resulted in more cooperation F(\, 130) = 10.43, p < .01, whereas negative mood participants reported being more competitive, F( 1, 130) = 4.91, p < .05. Cooperation was overall preferred to competition in the informal, interpersonal negotiation, F(l, 130) - 7.76,p < .01, but there was no differ- ence in the use of these two strategies in the formal, intergroup negotiation task (Figure 4 ) . Deal making was also an important feature of both negotiation sessions. Happy persons made more cooperative deals both in the informal, interpersonal, F( 1, 130) = 10.70, p < .01, and in the formal, intergroup negotiation, F( 1, 130) = 5.03, p < .05, and were also more likely to honor such deals in the interpersonal encounter, F ( l , 130) = 7.76, p < .01. Because making a deal is essentially an integrative, cooperative strategy that presumes a positive evaluation of the opponent, these results further confirm the hypothesized benefi- cial effects of good mood on integrative bargaining.
Mood effects on bargaining outcomes. Induced mood had the predicted influence on people's expected and reported nego- tiating outcomes. Even before the first negotiation, happy per- sons expected a higher final rank for their preferred course than did sad individuals (4.38 vs. 3.31), F ( l , 130) = 8 . 1 8 , p < .01, indicating mood effects on bargaining self-efficacy judgments. Consistent with these heightened expectations, reported bar- gaining outcomes indicated greater success by happy persons. People in a positive mood achieved a significantly higher ranking for their courses within their group than did sad persons (3.79 vs. 2.93), F ( l , 130) - 7.95, p < .01, and their better reported
6.6
6
4.6
4
3.5
3
INTERPERSONAL INTERGROUP
CD HAPPY ^ SAD
COOPERATION COMPETITION COOPERATION COMPETITION
NEGOTIATING STRATEGY
0.5
5
4.5
4
3.6
3
2.6
2
INTERPERSONAL INTERGROUP
4.6
3.6
2.B
CD HAPPY
^ BAD 4.6
3.6
PLAN DEALS HONOR DEALS PLAN DEALS HONOR DEALS
NEGOTIATING STRATEGY
2.6
Figure 3. The effects of good and bad mood on expected negotiation strategies (cooperation, competition, making deals, and honoring deals) in interpersonal and intergroup negotiation, showing mean preference ratings for each strategy.
performance also endured in the intergroup negotiation task (3.69 vs. 2.81), F ( l , 130) - 9 . 4 1 , p < .05.
These results confirm and extend the results obtained in Ex- periment 1. Persons experiencing a temporary good mood again showed a general positivity bias in their expectation and their construal of a forthcoming bargaining encounter, indicating a greater inclination to be cooperative and a reduced tendency to be competitive. These results support the AIM, suggesting that affectively valenced information will infuse people's construc- tive deliberations prior to a complex, unpredictable social en- counter such as a negotiation. Mood effects were not limited to plans and expectations, but also influenced reported bargaining behaviors and outcomes. This confirmation of the beneficial effects of positive affect on people's ability to plan and use integrative bargaining strategies may have considerable implica- tions for many real-life negotiation tasks (Pruitt & Carnevale, 1993; Thompson, 1995).
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
572 FORGAS
5.5
4.5
3.5
2.5
INTERPERSONAL INTERQROUP
C 3 HAPPY ^ SAO
5.5
4.5
3.5
2.5
COOPERATION COMPETITION COOPERATION COMPETITION
NEGOTIATING STRATEGY
INTERPERSONAL INTERQROUP
4.6
3.5
2.6
C3 HAPPY
^ BAD 4.5
3.5
MAKE DEALS HONOR DEALS MAKE DEALS HONOR DEALS
NEGOTIATING STRATEGY
2.5
Figure 4. The effects of good and bad mood on reported negotiation strategies (cooperation, competition, making deals, and honoring deals) in interpersonal and intergroup negotiation, showing mean reported us- age for each strategy.
4.64 vs. 5.65), F ( l , 101) = 4.60, p < .05. As expected, there were also significant interactions between mood and machiavel- lianism on competitive bargaining, F( 1, 97) - 11.82, p < .01, and the use of the reciprocal, tit for tat strategy, F( 1,97) = 9.78, p < .01. Low machiavellians were significantly more inclined to adopt a competitive strategy in a bad mood than in a good mood (M = 4.65 vs. 3.03), F( 1,50) = 9.88, p < .01. Mood did not have such an influence on the bargaining strategies of high machiavellians (Figure 5 ) .
This interaction suggests that the effects of negative mood on priming a mood-congruent negotiating approach (in this case, competitiveness) were limited to low machiavellians. More cyn- ical, detached, and manipulative persons who were likely to adopt more motivated strategies when thinking about the bar- gaining task (high scorers) were less influenced by mood. Low machiavellians were also more likely to use a passive tit for tat strategy when in a positive than a negative mood (5.31 vs. 4.11), ^ ( 1 , 49) = 7.11, p < .01. In contrast, mood made little difference to high machiavellians' reported use of this strategy (Figure 5 ) . These findings confirm that individual differences can play a critical role in mediating mood effects on social judgments and behavior (Mayer & Salovey, 1988; Smith & Petty, 1995).
Need for approval. Consistent with their greater need for
6
6.5
5
4.6
4
3.6
3
2.5
COMPETITION
/
cr'
TIT FOR TAT
'"a
- • - HAPPY
•a- SAD
High Low High
MACHIAVELLIANISM
Low
6
5.5
5
4.5
4
3.6
3
2.5
Individual-differences data. Next, a series of two-way anal- yses of variance were carried out evaluating the effects of (a) high and low scores on the need for approval and machiavellian- ism scales (above or below the mean) and ( b ) mood on planned and reported negotiating strategies. For the purposes of this analysis, data about expected and reported bargaining strategies from the interpersonal and the intergroup negotiation sessions were averaged. Because the effects of mood on negotiation strat- egies were generally parallel across these conditions (Figures 3 and 4 ) , this procedure should not distort the results and will allow a more clear-cut analysis of the interaction between mood and individual differences.
Machiavellianism had two main effects on negotiating strate- gies. Not surprisingly, high machiavellians were less inclined to be cooperative than were low scorers {M = 3.42 vs. 3.96), F ( l , 101) = 4.01, p < .05. In a particularly interesting valida- tion of the construct, high machiavellians were also significantly less likely to honor deals than were low machiavellians (M =
COOPERATION COMPETITION MAKE DEALS
2.5
1.5 High Low High Low High Low
SOCIAL DESIRABILITY
Figure 5. The interaction of mood and individual differences (machia- vellianism and social desirability) on negotiation strategies.
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 573
approval, persons who scored high on the Social Desirability Scale were less likely to prefer a competitive strategy in their negotiations than were low scorers (3.16 vs. 3.72), F(l, 101) = 4.21, p < .05. These people were also more likely to make deals than did those scoring low on social desirability (5.15 vs. 4.22), F( 1, 101) = 8.79, p < .01. This suggests that low need for approval may be an important prerequisite for adopt- ing competitive, nonintegrative approaches in a bargaining encounter.
Need for approval also had the predicted interactive effect with mood on cooperative bargaining strategies, F(l, 97) = 9.41, p < .01. For those unconcerned with approval, positive mood increased and sad mood decreased their tendency to be cooperative (4.94 vs. 2.82), F ( l , 50) = 13.21, p < .01. In contrast, good or bad mood had no influence on the cooperative negotiation strategies of persons who scored high on need for approval (Figure 5 ) . A conceptually similar interaction between mood and need for approval was also found on competitive bargaining strategies, F ( l , 97) = 6.78, p < .05. Those uncon- cerned with approval were more likely to be competitive when in a negative than in a positive mood (4.48 vs. 3.19), F{\, 50) = 7.42, p < .05, whereas mood had little influence on those scoring high on social desirability (3.21 vs. 3.3). These findings were further supported by an interaction between need for ap- proval and mood on participants' tendency to make deals in their negotiations, F ( l , 97) = 8.93, p < .01. It appears that negative mood decreased the likelihood of making deals for low need for approval persons (3.43 vs. 5.01), F ( l , 49) = 8.02, p < .01, but again had little effect on high need for approval individuals (5.25 vs. 5.07).
These results are consistent with multiprocess models such as the AIM, suggesting that affect infusion is reduced when substantive processing is impaired and people tend to rely on preexisting motives to guide their appraisal of the bargaining encounter. Alternative theories that assume simple, heuristic pro- cessing as the universal vehicle for affect infusion (e.g., affect- as-information or affect-as-input models) seem unable to ac- count for these findings. Our results confirm that individual differences in task-relevant motivational tendencies are likely to play a critical role in mediating mood effects on cognition and on negotiation (Rusting, 1997).
Experiment 3
The first two experiments showed that mood states have a significant effect on people's planned and reported negotiating strategies and that individual differences can play a major role in mediating these effects. However, negotiation by definition involves contingent interaction with others, where the behav- ior—and mood state—of the opposition may also be critically important in determining the tone of the encounter and its even- tual outcome. Experiment 3 was designed to evaluate the influ- ence of matching as against nonmatching moods between oppos- ing teams on negotiating strategies and outcomes. Little work has been done on the role of similar and dissimilar moods on how people think and behave in interactive social situations (Argyle, 1991; Fbrgas, 1982). It seems that people generally prefer to interact with others experiencing matching affective states. Early work by Schachter (1959) showed that anxious people not only seek the company of others but specifically
prefer those in a similar affective state. More recently, Locke and Horowitz (1990) found that there may be a general tendency for people to prefer interacting with those experiencing a match- ing, rather than a dissimilar, affective state.
What are the likely implications of opposing teams experienc- ing matching rather than nonmatching moods for negotiating strategies and outcomes? Extrapolating from the previous exper- iments as well as prior research on negotiator cognition and affect (Bazerman & Carroll, 1987; Forgas, 1995a; Thompson, 1990), in Experiment 3 we predicted that the opposition's mood would have a significant additional impact on negotiating strate- gies. Specifically, we hypothesized that shared positive mood by both teams would result in more cooperative, integrative negotiating strategies and outcomes than would nonmatched positive mood. Conversely, a dysphoric mood shared by the opposition would have a greater negative impact than would unilateral negative mood. This prediction assumes that the oppo- sition's mood should have a separate and additive mood-congru- ent influence on the negotiating strategies adopted.
Method
Overview, design, and participant.?. This experiment compared the strategies and outcomes of negotiations in which members of two groups experienced matched positive, matched negative, or nonmatched moods. The procedure described in Experiment 1 was used. However, because the mood of the opposing team was the critical variable here, only the intergroup negotiation task was studied. Before and after the intergroup bargaining, participants completed a questionnaire, as described in Ex- periment 1, indicating the strategies they used and their negotiating outcomes. The experiment was based on a 2 X 2 between-groups design, with mood (happy, sad) and the opposition's mood (happy, sad) as the independent variables. Participants were 96 students of both genders who participated in the experiment as part of their course requirements. Thirty persons participated in negotiations in which both groups were in a positive mood, 36 in negotiations in which both groups were in a negative mood, and 30 in negotiations in which the mood of the two groups was inconsistent. In the absence of gender differences, data for men and women were combined.
Procedure. The same mood manipulation procedure was used as in the previous studies. Participants assigned to each of the three-person negotiating teams were either all happy or all sad, allowing the formation of matched or nonmatched negotiating encounters. The negotiation exer- cise was the same as used in Experiments 1 and 2, involving the selection of courses in psychology. Before and after the intergroup negotiation, participants again completed a questionnaire reporting on their planned and actual use of various negotiation strategies, deal-making behaviors, and negotiating outcomes. The experiment thus compared negotiating strategies and outcomes in encounters between matched happy, matched sad, and mismatched happy-sad negotiating teams.
Results and Discussion
Mood validation. Participants in the happy and the sad con- ditions rated their mood as significantly different on the positive- negative scale (5.66 vs. 4.25), F ( l , 94) = 7.38, p < .01, once again confirming the effectiveness of the mood induction procedure.
Mood effects on negotiation strategies. Mood again had a significant influence on planned bargaining strategies. Happy persons were more likely than sad persons to plan to use a cooperative strategy (5.21 vs. 3.67), F ( l , 94) = 9.27, p < .01, and sad individuals were more likely to use a competitive rather
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
574 FORGAS
than a cooperative strategy (5.52 vs. 4.13), F ( l , 94) = 5.96, p < .05. Consistent with their more cooperative stance, happy people were also more likely to plan, F(l, 94) = 6.88, p < .05, and subsequently to honor deals, F(1,94) = 7.78, p < .01, as part of their negotiating approach than did sad people {5.22 vs. 3.97, and 5.63 vs. 4.41). These results replicate the findings of Experiments 1 and 2 and confirm that good moods have a strong beneficial influence on the way people construct expecta- tions about a forthcoming negotiation.
Results also showed a clear mood-congruent influence on postnegotiation reports of bargaining behaviors. Happy persons reported using more cooperative strategies than did sad people, F ( l , 94) = 8 . 5 1 , p -O.Ol, whereas negative mood resulted in more competitive strategies, F{ 1, 94) = 5.67, p < .05. Making deals, F ( l , 94) = 7.39, p < .01, and honoring deals, F ( l , 94) = 7.75, p < .01, were significantly more common in good rather than bad moods (Figure 6 ) . These results replicate the findings of Experiments 1 and 2 and indicate a marked tendency for affect infusion in highly constructive social tasks such as negoti- ations, as predicted by the AIM.
Effects of the opposition's mood on negotiation strategies. The mood of the opposition also had a significant independent main effect on the reported use of cooperation, F ( l , 94) = 9.55, p < .01, and competition, F ( l , 94) = 6.69, p < .05, as well as making, F ( l , 94) = 7.40, p < .01, and honoring deals once made, F ( l , 94) = 4.72, p < .05. These results show a consistent pattern of an indirect mood-congruent bias in bar- gaining, with individuals facing a happy opposition reporting more cooperative and less competitive strategies and being more likely to make and to honor deals than people facing a dysphoric opposition (Figure 6 ) . There was no evidence for an interaction between participant mood and opposition mood on any of the dependent variables studied here, suggesting that the effects of partner mood were similar to and additive to the effects of people's own moods on negotiating strategies. These findings suggest that even when a negotiator is in a dysphoric mood, the
I I HAPPY-HAPPY
I B HAPPY-SAO
ESS SAD-HAPPY
• i SAD-SAD
COOPERATION COMPETITION MAKE DEALS HONOR DEALS
NEGOTIATION STRATEGIES
Figure 6. The effects of negotiator mood and opposition mood on four negotiating strategies (cooperation, competition, making deals, and honoring deals).
positive affective state of the opposition may still have a signifi- cant beneficial effect on reducing competitiveness and enhanc- ing cooperation in the bargaining encounter.
Mood effects on bargaining outcomes. As most participants were facing a team in a matching mood (N = 6 4 ) , in this experiment mood had no overall effect on bargaining outcomes. However, for the subset of individuals who faced an opposition team in a nonmatching mood (N = 36), a significant mood effect on bargaining outcomes was found. Happy persons re- ported greater success for their preferred courses than did dys- phoric persons (4.32 vs. 3.01), F ( l , 34) = 5.69, p < .05. These results show that positive moods conferred significant benefits in terms of bargaining outcomes, but only for those individuals facing an opposition in a bad mood.
The results of Experiment 3 confirm that the positive mood of self as well as the opposition can make a significant and independent contribution to the adoption of integrative negotia- tion strategies, in the form of greater cooperation, reduced com- petitiveness, and more constructive deal making. In this experi- ment, improved outcomes could be achieved only at the cost of reduced outcomes for the opposition, so the benefits of good mood on outcomes could be shown only where happy persons faced a dysphoric opposition, However, these results suggest that good mood may confer potential benefits on both sides in terms of finding integrative solutions in tasks that are not characterized by zero-sum outcomes. These findings establish that integrative bargaining strategies can be influenced not only directly, by the mood state of a person (as found in Experiments 1 and 2 ) , but also indirectly, by the affective state of the opposi- tion, suggesting something like a shared affective benefit in some interpersonal behaviors.
General Discussion and Conclusion
These three experiments provide solid cumulative evidence for the role of transient mood states in negotiator cognition and bargaining. Positive mood had a significant beneficial influence on people's expectations, strategies, and outcomes in both inter- personal and intergroup bargaining encounters. The main conse- quence of good mood seems to be a marked inclination to be more cooperative and less competitive in planned and reported bargaining strategies. These effects occurred both in informal, interpersonal and more formal, intergroup negotiation tasks, suggesting a degree of cross-situational generality in these find- ings. Results also showed a significant negative mood effect on negotiation here. The role of aversive affect in triggering competitive social behaviors has long been suspected (Berko- witz, 1993) but rarely demonstrated in the context of bargaining encounters. These results have several interesting theoretical and practical implications.
Theoretical Implications
What is the most likely explanation for such consistent mood effects on negotiation? It appears that mood effects on the cogni- tive processes and expectations of the negotiators when inter- preting the encounter played a central role in these results (Baz- erman & Carroll, 1987; Rubin et al., 1990; Thompson, 1990, 1995). Mood first influenced prebargaining plans and goals, leading to mood-consistent differences in negotiating strategies
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 575
and, ultimately, outcomes. The AIM offers a simple and parsimo- nious explanation of how such affect infusion occurs (Forgas, 1992a, 1995a). Given the complex and indeterminate nature of the negotiating task, participants' likely involvement in the encounter, and the personal relevance of the issue, the AIM predicts that a substantive, constructive processing strategy should be predominantly used in computing an assessment of the situation and formulating strategic plans. It is in the course of substantive processing that mood is most likely to selectively facilitate the recall, processing, and use of affectively congruent information, producing a significant mood-congruent influence on people's expectations and judgments.
In other words, good mood first resulted in more favorable expectations about the task and the opposition, leading to more cooperative and less competitive expected and reported bar- gaining strategies as a result of affect infusion processes, as confirmed by the mediational analyses in Experiment 1. Such positive, integrative strategies in turn produced more successful outcomes, as was also found in other studies (Pruitt & Carne- vale, 1993). These results indicate greater affect infusion in conditions requiring more substantive processing, as suggested by the AIM and as found in several recent experiments (Bower; 1991; Fiedler, 1991; Forgas, 1992b, 1995b; Mayer et al., 1992; Sedikides, 1995). Alternative theories assuming simple, heuris- tic processing strategies such as the affect-as-information and the affect-as-input models do not appear applicable to these results.
Could superior outcomes be explained in terms of good moods inducing greater flexibility and creativity, as some previ- ous studies suggest (Hertel & Fiedler, 1994)? The present re- sults do not support such an explanation. The mood effects obtained here appear clearly valenced, with positive mood selec- tively enhancing cooperation and negative mood facilitating competition. Together with the absence of mood effects on such flexible, indeterminate negotiatng strategies as tit for tat or ran- dom moves, these results do not indicate mood induced flexibil- ity in bargaining strategies.
A particularly interesting finding is the marked reduction of mood effects for individuals scoring high on need for approval and machiavellianism in Experiment 2. The AIM predicts that affect infusion should be constrained when people pursue a targeted, motivational goal that interferes with their ability to engage in open, substantive information processing, as is likely to be the case when high machiavellian or high need for approval individuals contemplate the prospect of a bargaining encounter (Rusting, 1997). These results again cannot be explained by theories that emphasize simplistic, heuristic processes as the sole mechanism of affect infusion, as do affect-as-information and affect-as-input models. More generally, these findings con- firm that trait variables can play a critical role in mediating mood effects on negotiator cognition (Mayer & Salovey, 1988; Thompson, 1990).
Practical Implications
Formal and informal negotiations of the kind studied here play an important role in everyday life. Despite widespread anecdotal and intuitive evidence for mood effects on negotiation, the practical consequences of this phenomenon remain incom- pletely understood. These results, together with earlier studies,
suggest a powerful tendency for people to adopt more coopera- tive and integrative strategies when feeling good, and more com- petitive strategies when feeling bad (R. A. Baron, 1990; Ber- kowitz, 1993; Carnevale & Isen, 1986). Such mood effects should have important practical implications for our understand- ing of conflict resolution both in interpersonal and in organiza- tional settings. Recent studies have found that moods can influ- ence the way people perceive and evaluate their intimate rela- tionships, attribute responsibility for their personal conflicts, and select conflict resolution strategies (Forgas, 1994; Forgas et al., 1990). The present experiments add to this evidence by showing that a marked mood-congruent effect may also operate in the course of strategic interpersonal negotiations.
The role of moods in organizational behavior is also receiving growing attention (R. A. Baron, 1990). Our results suggest that cooperation in good moods and competition in bad moods may be a more prevalent negotiating strategy even in formal, in- tergroup organizational encounters than is commonly realized. Experiment 3 in particular suggests that something like a shared affective benefit may well characterize such bargaining encoun- ters, with the mood of the opposition having an independent additional effect on the degree of cooperation adopted.
Limitations and Future Prospects
There are also some obvious limitations to these results. These studies attempted to create realistic and involving negotiating situations, with individuals discussing a topic of considerable personal interest (course preferences), creating a relevant and familiar bargaining context. Nevertheless, the generality of our findings must be limited by the degree of realism we were able to achieve in these bargaining encounters, as is inevitably the case with controlled experiments on negotiation. We should also note that the success-failure mood induction procedures used here (as is the case with all such procedures) may have produced additional, unintended effects on self-assertiveness or self-es- teem in some participants. However, the finding of increased cooperation rather than competition after positive feedback con- tradicts assertiveness effects as an alternative explanation. In any case, to the extent that our results are consistent with various mood-congruent outcomes reported in other experiments using different mood induction procedures, it is unlikely that mood manipulation confounds had a significant impact on our findings (cf. Forgas, 1995a).
The experimental procedure of looking at two consecutive negotiation tasks may also have introduced carryover effects, with cooperation and competition in the interpersonal bargaining encounter a potential influence on the subsequent intergroup negotiation (Keenan & Carnevale, 1989). However, given the similar mood effects on prenegotiation plans and reported bar- gaining strategies in both encounters, it seems unlikely that interactive experiences in the first task were a major influence on subsequent behavior. Another issue is that the bargaining strategies used were assessed in terms of post negotiation self- reports here. Could it be that mood influenced these verbal reports rather than actual bargaining behaviors? The results of the mediational analyses in Experiment 1 and the fact that bar- gaining outcomes were also systematically influenced by mood indicate that there must have been genuine behavioral differ- ences in bargaining consistent with postnegotiation reports.
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
576 FORGAS
Another issue concerns the quality of the moods induced here. In common with much of the earlier research on mood effects on cognition and judgments, these studies examined the effects of nonspecific, undifferentiated, mild good or bad moods on negotia- tion. Of course, more intense and specific affective states may well have different effects. The influence of specific emotions such as anger on negotiator cognition and bargaining strategies deserves serious attention in future research (Berkowitz, 1993).
Finally, the proposed explanation of these results in terms of mood induced influences on cognitive strategies as predicted by the AIM would benefit from further confirmation that more di- rectly assesses actual processing strategies. It is also possible, indeed likely, that additional features of the person, the task, and the opposition not considered here may play a critical role in cueing alternative processing strategies and thus different mood effects. Future research needs to consider the role of such contex- tual variables in recruiting different processing strategies and, thus, in mediating mood effects on negotiation performance.
Negotiation is among the most complex cognitive tasks people face in everyday life. Despite clear evidence for the role of affect in many social judgments and behaviors, little has been known about how feelings impact on people's negotiating strate- gies and outcomes. These studies showed that both good and bad moods have a marked mood-congruent influence on planned and reported bargaining strategies and outcomes. In contrast, people with a predetermined motivation (such as high need for approval or high machiavellianism) were less likely to show mood effects, consistent with their greater use of motivated processing strategies. A multiprocess framework (Fiedler, 1991; Forgas, 1995a) appears particularly suitable to understanding these subtle effects. Further research on the role of affect in negotiation should be of considerable theoretical as well as applied interest to our understanding of mood effects on cogni- tion and of the dynamics of interpersonal and intergroup behav- ior in particular.
References
Argyle, M. (1991). A critique of cognitive approaches to social interac- tion and social judgments. In J.P. Forgas (Ed.), Emotion and social judgments. Oxford, England: Pergamon Press.
Baron, R. A. (1990). Environmentally induced positive affect: Its impact on self-efficacy, task performance, negotiation, and conflict. Journal of Applied Social Psychology, 20, 368-384.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research. Journal of Personality and Social Psychology, 51, 1173-1182.
Bazerman, M. H., & Carroll, J. S. (1987). Negotiator cognition. Re- search in Organizational Behavior, 9, 247-288.
Berkowitz, L. (1993). Towards a general theory of anger and emotional aggression. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition (Vol. 6, pp. 1-46). Hillsdale, NJ: Erlbaum.
Bower, G. H. (1991). Mood congruity of social judgments. In J.P. Forgas (Ed.), Emotion and social judgments (pp. 31-53). Oxford: Pergamon Press.
Carnevale, P. J. D., & Isen, A . M . (1986). The influence of positive affect and visual access on the discovery of integrative solutions in bilateral negotiation. Organizational Behavior and Human Decision Processes, 37, 1-13.
Christie, R. (1968). The machiavellianism scale. Unpublished manu- script, Columbia University.
Crowne, D., & Marlowe, D. (1964). The approval motive. New York: Wiley.
Deutsch, M. (1973). The resolution of conflict. New Haven, CT: Yale University Press.
Deutsch, M. (1994). Constructive conflict resolution: Principles, train- ing and research. Journal of Social Issues, 50, 13-32.
Erber, R., & Erber, M. W. (1994). Beyond mood and social judgment: Mood incongruent recall and mood regulation. European Journal of Social Psychology, 24, 79-88.
Eeshbach, S., & Singer, R. D. (1957). The effects of fear arousal and suppression of fear upon social perception. Journal of Abnormal and Social Psychology, 55, 283-288.
Fiedler, K. (1991). On the task, the measures and the mood in research on affect and social cognition. In J.P. Forgas (Ed.), Emotion and social judgments (pp. 83-104). Oxford: Pergamon Press.
Forgas, J. P. (1982). Episode cognition: Internal representations of inter- action routines. In L, Berkowitz (Ed.), Advances in experimental social psychology (Vol. 15, pp. 59-100). New York: Academic Press.
Forgas, J. P. (1991). Mood effects on partner choice: Role of affect in social decisions. Journal of Personality and Social Psychology, 61, 708-720.
Forgas, J. P. (1992a). Affect in social judgments and decisions: A multi- process model. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 25, pp. 227-275). New 'fork: Academic Press.
Forgas, J. P. (1992b). On bad mood and peculiar people: Affect and person typicality in impression formation. Journal of Personality and Social Psychology, 62, 863-875.
Forgas, J. P. (1994). Sad and guilty? Affective influences on the explana- tion of conflict episodes. Journal of Personality and Social Psychol- ogy, 66, 56-68.
Forgas, J. P. {1995a). Mood and judgment: The Affect Infusion Model (AIM). Psychological Bulletin, 116, 39-66.
Forgas, J. P. (1995b). Strange couples: Mood effects on judgments and memory about prototypical and atypical targets. Personality and So- cial Psychology Bulletin, 21, 747-765.
Forgas, J. P., & Bower, G. H . ( 1987). Mood effects on person perception judgements. Journal of Personality and Social Psychology, 53, 53- 60.
Forgas, J. P., Bower, G. H., & Moylan, S. J. (1990). Praise or blame? Affective influences on attributions for achievement. Journal of Per- sonality and Social Psychology, 59, 809-818.
Forgas, J. P., & Fiedler, K. (1996). Us and them: Mood effects on intergroup discrimination. Journal of Personality and Social Psychol- ogy, 70, 28-40.
Hertel, G., & Fiedler, K. (1994). Affective and cognitive influences in a social dilemma game. European Journal of Social Psychology, 24, 131-146.
Isen. A. (1987). Positive affect, cognitive processes and social behaviour. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 20, pp. 203-253). New York: Academic Press.
Keenan, P. A., & Carnevale, P. J. (1989). Positive effects of within- group cooperation on between-group negotiation. Journal of Applied Social Psychology, 19, 977-992.
Kruglanski, A. (1989). Lay epistemics and human knowledge: Cognitive and motivational bases. New York: Plenum.
Lanzetta, J. T. (1989). Expectations of cooperation and competition and their effects on observers' vicarious emotional responses. Journal of Personality and Social Psychology, 56, 543-554.
Locke, K. D., & Horowitz, L. M. (1990). Satisfaction in interpersonal interactions as a function of similarity in level of dysphoria. Journal of Personality and Social Psychology, 58, 823-831.
March, J. G., & Simon, H. A. (1958). Organizations. New York: Wiley. Mayer, J. D., Gaschke, Y. N., Braverman, D. L., & Evans, T. W. (1992).
Mood congruent judgment is a general effect. Journal of Personality and Social Psychology, 63, 119-132.
Mayer, J. D., & Salovey, P. (1988). Personality moderates the interaction of mood and cognition. In K. Fiedler & J. P. Forgas (Eds.), Affect,
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
MOOD AND NEGOTIATION 577
cognition and social behaviour (pp. 87-99). Toronto, Canada: Hogrefe.
Neale, M.'A., & Bazerman, M. H. (1991). Cognition and rationality in negotiation. New Ifork: Free Press.
Pruitt, J. D., & Carnevale, P. J. D. (1993). Negotiation in social conflict. Buckingham, England: Open University Press.
Rahim, M. A. (1983). A measure of styles of handling interpersonal conflict. Academy of Management Journal, 26, 368-376.
Ross, L., Lepper, M , & Hubbard, M. (1975). Perseverance in self- perception and social perception: Biased attributional processes in the debriefing paradigm. Journal of Personality and Social Psychology, 32, 880-892.
Rubin, J. Z., & Brown, B. R. (1975). The social psychology of bar- gaining and negotiation. New Yoxkr. Academic Press.
Rubin, J. Z., Kim, S. H., & Peretz, N. M. (1990). Expectancy effects and negotiation. Journal of Social Issues, 46, 125-139.
Rusting, C. L. (1997). Personality, mood, and cognitive processing of emotional information: Three alternative models. Manuscript submit- ted for publication.
Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9, 185-211.
Schachter, S. {1959). The psychology of affiliation. Stanford, CA: Stan- ford University Press.
Sedikides, C. (1994). Incongruent effects of sad mood on self-concep- tion valence: It's a matter of time. European Journal of Social Psy- chology, 24, 161-172.
Sedikides, C. (1995). Central and peripheral self-conceptions are differ- entially influenced by mood: Tests of the differential sensitivity hy- pothesis. Journal of Personality and Social Psychology, 69, 759-777.
Smith, S. M., & Petty, R. E. (1995). Personality moderators of mood congruence effects on cognition: The role of self-esteem and negative mood regulation. Journal of Personality and Social Psychology, 68, 1092-1107.
Thomas, K. W. (1990). Conflict and negotiation processes in organiza- tions. In M. D. Dunnette (Ed.), Handbook of industrial and organiza- tional psychology (2nd ed.). Chicago: Rand McNally.
Thompson, L. (1990). Negotiation behavior and outcomes: Empirical evidence and theoretical issues. Psychological Bulletin, 108, 515-532.
Thompson, L. (1995). They saw a negotiation: Partisanship and involve- ment. Journal of Personality and Social Psychology, 68, 839-853.
Received October 4, 1996 Revision received May 1, 1997
Accepted May 5, 1997 •
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
