APPLIED PSYCHOLOGY: AN INTERNATIONAL REVIEW, 2006, 55 (1), 73–90

Thinking Deeply May Sometimes Help:

SCHEI
O
COGNITIVE
Blackwell
Oxford,
Applied
APPS
©
0269-994X
January
1
55riginalET
Article
International
UKAL.MOTIVATION
Psychology:
2006
Publishing
Association
an and NEGOTIATION
LtdInternational
for Applied ROLES
Psychology, 2005

Cognitive Motivation and Role Effects in

Negotiation
Vidar Schei,* Jørn K. Rognes and Solfrid Mykland
Norwegian School of Economics and Business Administration, Norway

On étudie dans cet article la façon dont la motivation cognitive et le rôle de

négociateur interagissent pour prédire les comportements et les résultats des
négociations duelles. La motivation cognitive de 64 participants a été
mesurée antérieurement à une simulation de négociation achatvente. Les
résultats montrent que l’impact de la motivation cognitive sur la négociation
dépend du rôle joué. Comme prévu, les paires pourvues de vendeur à la
motivation cognitive élevée parvenaient bien plus facilement à un accord que
celles dont le vendeur avait une motivation cognitive basse. Par contre, la
motivation cognitive des acheteurs n’avait pas de retombées sur le déroule-
ment de la négociation. L’analyse des transcriptions des négociations indiqua
que l’information fournie avait une fonction d’interface entre la motivation
cognitive des vendeurs et l’accord. On discute des conséquences théoriques et
pratiques.

This article examines how cognitive motivation and negotiation role interact
to predict behavior and outcomes in integrative dyadic negotiations. The
cognitive motivation of 64 participants was measured prior to a simulated
buyer–seller negotiation. The results suggest that the effect of cognitive motiva-
tion in negotiation depends upon the role of the negotiator. As predicted,
dyads with sellers who were high on cognitive motivation achieved much
better joint outcomes than dyads with sellers low on cognitive motivation.
By contrast, the cognitive motivation of the buyers did not affect negotiation
outcomes. Mediational analyses of negotiation transcripts illustrated that the
giving of information was a mediating factor between the cognitive motiva-
tion of the sellers and joint outcomes. Theoretical and practical implications
are discussed.

* Address for correspondence: Vidar Schei, Norwegian School of Economics and Business
Administration, Department of Strategy and Management, Breiviksveien 40, N-5045 Bergen,
Norway. Email: [email protected]
We wish to thank Associate Editor Michele J. Gelfand and two anonymous reviewers for
their constructive comments during the review process.
Parts of this research were presented at the annual meeting of the International Association
for Conflict Management in Salt Lake City, Utah, June 2002.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ,
UK and 350 Main Street, Malden, MA 02148, USA.
74 SCHEI ET AL.

INTRODUCTION

Negotiation is part of everyday life, and many of the negotiation situations

we encounter have integrative potential. This implies that it is possible to
achieve joint outcomes higher than those that are achieved by splitting the
differences and settling for compromises. Unfortunately, it seems that
neither the concept of integrative negotiations nor the process of how to
achieve high joint outcomes comes intuitively to negotiators (Brett, 2001).
Therefore, negotiators often fail to identify and achieve high joint outcomes
in integrative negotiations (see Lewicki, Barry, Saunders, & Minton, 2003,
for a review). Failure to realise the win-win potential in the negotiations can
lead to impasses, escalated conflict level between the parties, dissatisfied
negotiators, and unstable agreements (Pruitt, 1981). Research on negoti-
ation has therefore focused intensively on how negotiators can overcome the
barriers to high joint outcomes (Bazerman, Curhan, Moore, & Valley, 2000).
A consistent finding is that a negotiation process that facilitates the
exchange and processing of information is critical to capitalise on win-win
potential (Pruitt & Carnevale, 1993).
The purpose of the present article is to investigate whether an individual
difference variable, cognitive motivation, helps to overcome barriers in
negotiation to create win-win agreements. Cognitive motivation, or need for
cognition, is a stable individual difference in the tendency to engage in
arduous, analytical thinking (Chatterjee, Heath, Milberg, & France, 2000)
and has not previously been studied in research on integrative negotiations.
In research on individual problem solving, cognitive motivation has consistently
been found to affect the search for, and use of, information (see Cacioppo,
Petty, Feinstein, & Jarvis, 1996, for a review). For example, individuals high on
cognitive motivation have a tendency to eagerly seek out and use relevant
information, and base their judgments on rational considerations (Berzonsky
& Sullivan, 1992; Leary, Sheppard, McNeil, Jenkins, & Barnes, 1986).
Conversely, individuals low on cognitive motivation have a tendency to rely
more on cognitive heuristics (Cacioppo et al., 1996) and are influenced by
framing biases in decision making (Smith & Levin, 1996). Given the import-
ance of gathering and combining information across negotiation issues and
negotiators in order to identify high joint outcomes, we expect that cognitive
motivation will be a potent predictor of negotiation success.
The effects of cognitive motivation on performance have typically been
studied with the individual as the unit of analysis (Cacioppo et al., 1996).
Negotiation is, of course, an interactive task where outcomes depend on the
interaction between two (or more) individuals occupying different roles, e.g.
buyer and seller. Previous negotiation research suggests strong role effects
(e.g. Cai, Wilson, & Drake, 2000). Negotiation roles have effects on both
pre-negotiation expectations and on the negotiators’ influence on process

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 75
and outcomes. As discussed below, sellers have been found to be more import-
ant for the negotiation process and outcomes than buyers (Cai et al., 2000;
Weingart, Thompson, Bazerman, & Carroll, 1990). Thus, it is likely that
effects of cognitive motivation on negotiation depend on role. In what
follows, we first elaborate upon the basic challenges that are associated with
integrative negotiation. We then discuss how cognitive motivation and role
are expected to impact outcomes in integrative negotiation. Finally, we
discuss how the negotiation process mediates the effects of cognitive
motivation and role on outcomes.

COGNITIVE MOTIVATION AND ROLE DIFFERENCES IN

INTEGRATIVE NEGOTIATION
Integrative negotiation is a challenging task. In order to identify solutions
that imply higher joint outcomes than simple compromises, negotiators
have to gather and combine information cues in new and non-obvious ways.
Typically in negotiations, the parties have a better understanding of their
own priorities and goals than of the other party’s preference structure.
Therefore, the private information held by each of the negotiators must,
through the negotiation process, be combined into more complex informa-
tion and knowledge about how the parties’ preferences relate to each other.
In dyadic negotiations, the knowledge of two individuals must be transformed
into new collective information and knowledge. Thus, the major tool for
achieving high joint outcomes is understanding how own preferences are
related to the preferences of the other party. Usually in negotiations, this
understanding is achieved through explicit information exchange about
priorities and interests, or by implicit information exchange through
the energetic exchange of numerous multiple-issue offers (cf. Tutzauer &
Roloff, 1988). The ability to generate collective information is hindered by
positional argumentations and by an over-focus on single-issue offers that
ignores the complexity of the negotiations (Weingart et al., 1990).
We expect that, depending on negotiation role, individuals high on
cognitive motivation may potentially contribute to the active exchange of
information and thereby to high joint outcomes in negotiations. Previous
research suggests why this might be the case. For example, high cognitive
motivation contributes to an individual’s ability to recall more information
cues (Kassin, Reddy, & Tulloch, 1990), especially when the information is
inconsistent with previously held expectations (Lassiter, Briggs, & Slaw,
1991). Furthermore, these individuals focus more energetically on cognitive
tasks (Osberg, 1987), base their judgment more on empirical information
(Leary et al., 1986), and actively search for and utilise relevant information
(Berzonsky & Sullivan, 1992). They are also curious (Olson, Camp, & Fuller,
1984), and oriented towards maximising information gain (Sorrentino,

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
76 SCHEI ET AL.

Bobocel, Gitta, Olson, & Hewitt, 1988). They are therefore more effective
decision makers in badly structured tasks than individuals with low cogni-
tive motivation (Nair & Ramnarayan, 2000). In contrast, individuals low on
cognitive motivation generally are easily stressed by complex cognitive tasks
(Gulgos, 2001). Taken together, the findings discussed above suggest that
negotiators with high cognitive motivation may be active and focused, and
seek out new information from their counterparts. They do so, not because
they are more intelligent than those low on cognitive motivation (Cacioppo
et al., 1996), but because of their enjoyment of challenging cognitive tasks.
In negotiation, however, negotiators’ role may influence the expectations
about the task and the influence the negotiator has on process and out-
comes. Negotiation research has consistently found characteristics of the
seller to be more important for process and outcomes than characteristics
of the buyer. Weingart et al. (1990), for example, found a positive relation-
ship between providing information and joint outcomes, but only when the
information was provided by the sellers. Similarly, Cai et al. (2000) found
that the sellers’ behavior influenced outcomes more than the behavior of the
buyers. Moreover, they found that seller characteristics (level of collectiv-
ism) influenced the negotiation process and the joint outcomes, whereas
buyer characteristics did not impact process and outcomes.
Differential role effects in negotiation are likely due to norms and role
expectations that are associated with buyers and sellers. In any given nego-
tiation, expectations direct attention and constrain behavior (Neale, Huber,
& Northcraft, 1987), and thereby influence role enactments. Buyers are
often perceived as more powerful than sellers (Brett, 2001; Drake, 2001).
This affects the negotiators’ behavior, even if objectively there may be no
power differences. Furthermore, buyers and sellers seem to frame the nego-
tiation problem differently (cf. Neale & Bazerman, 1985). Typically, buyers
are negatively framed since they have to pay when agreements are reached.
Therefore, they are not very concessionary, they are stubborn, and willing
to risk stalemates. Sellers, on the other hand, are being paid for the
commodity they are offering. A deal is highly valued, they will not risk
stalemates, and will actively pursue an agreement.
The differences in learned role expectations may result in differential
effects of cognitive motivation across roles. Sellers will actively take the
initiative in negotiations in their efforts to secure an agreement (Neale et al.,
1987). Buyers will be passive to a larger extent, refuse to concede, and
expect the sellers to come up with acceptable propositions. Thus, buyers will
only participate in active information exchange, and search for creative
solutions, if the process is initiated by the sellers, and if they believe they
have nothing to lose from participation in the process. Therefore, sellers
high on cognitive motivation may trigger an active process of information
exchange, which enables them to identify and propose solutions that

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 77
improve joint outcomes. Sellers low on cognitive motivation are lazier prob-
lem solvers and do not trigger a process that helps them to identify creative
solutions that integrate preferences constructively.
While sellers must engage in the challenging task of persuading buyers in
order to get high joint outcomes, to a large extent buyers expect to be
offered an acceptable deal. Thus, their primary role is not to initiate exten-
sive information search and processing. Rather, they may focus on own
preferences and accept an agreement when these preferences are met in
acceptable ways (cf. Cai et al., 2000). Given that the role expectation of
buyers does not include active persuasion, their degree of cognitive motiva-
tion matters less for the negotiation process and outcomes. Thus, we put
forth the following hypotheses:

H1a: In buyer–seller dyads where the sellers have high cognitive motivation
the joint outcomes for the dyad and the individual outcome for the sellers
will be higher than in dyads where the sellers have low cognitive motivation.
H1b: In buyer–seller dyads, the cognitive motivation of the buyers will have
no influence on the individual and the joint outcomes.

We suggest that the energetic approach taken by cognitively motivated

sellers influences the negotiation process and thereby outcomes. In negotia-
tion research, one typically distinguishes between integrative and distribu-
tive behaviors and processes. Integrative behaviors include the explicit and
implicit exchange of information about priorities across negotiation issues.
Explicit information exchange includes asking for and giving information
about priorities. Implicit information exchange implies that the information
is embedded in multi-issue offers. The negotiators then gradually uncover
how their preferences are interrelated through a trial-and-error process. Dis-
tributive behaviors imply a focus on one issue at a time through positional
statements and arguments. As expected, integrative behaviors foster high
joint outcomes, while distributive behaviors hinder such outcomes.
Based on the reasoning leading to Hypotheses 1a and 1b, we expect dyads
where sellers have high cognitive motivation to display more integrative and
less distributive behavior than dyads where sellers have low cognitive motiva-
tion. They do so by actively searching for and using information that
helps the dyad identify integrative potential, and by avoiding distributive
behavior that leads to low joint outcomes. Sellers low on cognitive motiva-
tion will focus to a lesser extent on generating new information, and they
depend more on buyer behavior and own pre-negotiation expectations and
less on the development of problem definition when negotiating.
While prior research does not guide us in how the process in dyads with
cognitively motivated sellers develops, we propose that the approach taken
to the negotiation by the cognitively motivated sellers may also trigger

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
78 SCHEI ET AL.

integrative behavior among buyers. There is a strong tendency for reciprocity

in negotiations (Brett, Shapiro, & Lytle, 1998; Gouldner, 1960), and the
sellers’ eager search for an understanding of the negotiation problem may
trigger similar behavior from their opponents. Neale et al. (1987) suggested
that sellers must typically initiate action and the buyers will respond.
Cognitively motivated sellers may thus facilitate high joint outcomes by
stimulating the information exchange, by making multiple-issue offers, and
by avoiding using distributive behavior such as single-issue offers.
Furthermore, cognitively motivated sellers may unilaterally search for
and develop solutions even if their counterparts are reluctant to get involved
in integrative behavior. Thus through trial-and-error, they may be able to
find ways of satisfying their counterparts and at the same time increase their
own outcomes. This route to high joint outcomes implies that the sellers
display more integrative behavior than buyers, or that seller behavior is
more influential in affecting outcomes. The latter seems to be the case in the
research by Cai et al. (2000) on collectivism, roles, and outcomes in negoti-
ation, and the study by Weingart et al. (1990) on negotiation tactics. In sum,
we suggest that sellers with high cognitive motivation use integrative behavior,
trigger integrative behavior from the buyers, and thereby improve the joint
outcomes.

H2: The effect of sellers’ cognitive motivation on joint outcomes will be

mediated by information sharing and offer type (i.e. single- vs. multiple-issue
offers).

METHOD

Participants and Procedures

Sixty-four business students enrolled in negotiation courses participated in
this study. Their average age was 25 years, and women comprised 42 per cent
of the sample. We conducted the negotiation simulations during the first
meeting of classes. A seller and a buyer negotiated the delivery of television
sets (cf. Pruitt & Lewis, 1975). They negotiated three issues: delivery time,
product variations, and financing terms. Each issue had nine alternative
settlement points. We chose this simulation because it is a commonly used
variable-sum negotiation that allows for integrative outcomes through
logrolling (Pruitt, 1981). That is, the parties can achieve high joint outcomes
by exchanging concessions on their low-profit issues.
Prior to the negotiation exercise, the students were asked to answer a
questionnaire containing several measures, including a measure of cognitive
motivation, after which the negotiation exercise started. Each student
received confidential background information and a profit schedule. The

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 79
background information contained information about the three negotiation
issues and some arguments to further their case. The profit schedule showed
their individual earnings for each possible alternative on each of the three
issues. The context of the negotiation was role-neutral as it did not stimulate
any role (seller or buyer) to take the initiative or to be more active in the
negotiation. The subjects had 10 minutes for preparation before they were
randomly assigned to dyads (a seller and a buyer) and led to separate rooms.
The dyads were given 30 minutes to negotiate. Finally, the students were
debriefed.

Measurement and Validation

Cognitive Motivation. We used the short version of “the need for cogni-
tion scale” (Cacioppo, Petty, & Kao, 1984) to measure cognitive motivation
(CM). The scale consisted of statements such as “I prefer complex rather
than simple problems”, “I enjoy having responsibility for situations that
require a lot of thinking”, and “I appreciate tasks which imply finding new
solutions to problems”. Higher scores reflected higher CM (scales ranged
from 1 to 5). In their extensive review, Cacioppo et al. (1996) reported high
reliability and validity for scales using subsets of the 18 items. Consistent
with previous research, we used the items that loaded on the dominant
factor in a principal component analysis. The factor explained 24.6 per cent
of the variance, and the coefficient alpha was .79 for the eight items. One
dyad was excluded from further analyses because of extremely low buyer
CM. Furthermore, in order to reduce multi-collinearity when main effects
and interaction effects were analysed, we recoded the CM variable so it was
centered on zero.

Outcomes. At the individual level, outcomes were measured as the total

profit achieved by the negotiator across the three issues. The possible indi-
vidual outcome ranged from 0 to 8000. At the dyadic level, the negotiators’
joint outcomes were measured as the sum of the individual outcomes
achieved by a seller and a buyer in a dyad. We also measured Pareto
efficiency. We developed an index based on Tripp and Sondak (1992), where
we measured the number of possible agreements that were superior to the
solution chosen by each dyad. However, as the correlation between Pareto
efficiency (reversed) and joint outcomes was very high (r = .90, n = 31), we
decided to use only one of the measures. We used joint outcomes, as this is
the most widely used measure in dyadic negotiations (Neale & Northcraft,
1991). In fully integrative outcomes, the joint outcomes will be 10,400.
Integrative outcomes are possible because the importance of each single
issue varies across the negotiators. Thus, by making trade-offs across
the issues (i.e. give on the issue of low value in order to get concessions on

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
80 SCHEI ET AL.

the issue of high value) the negotiators can get higher joint outcomes
than if they make compromises sequentially on each single issue. If the
negotiators did not exchange concessions on their low-profit issues, but
rather compromised on each of the issues, they would only reach 8000 in
joint outcomes.

Process Variables. The verbal interactions of 17 randomly chosen

dyads were tape-recorded. The dyads that were tape-recorded were ran-
domly chosen from all classes, and the tape-recorded dyads did not differ
significantly from the other dyads on any of the variables included in the
study. Transcripts of the tape-recorded dyads were coded for information
sharing and offer by two independent coders, using a procedure similar to
that of Cai et al. (2000). The unit of analysis was the uninterrupted talking
turn. Only turns including information sharing or offers were coded.
Information sharing included giving information and asking for informa-
tion and were coded separately for sellers and buyers. Giving information
was defined as statements about how the negotiator prioritised the issues
(e.g. “Date of delivery is more important to me than financial terms”).
Asking for information was defined as questions about how the opposite
negotiator prioritised the issues (e.g. “What is the most important issue for
you?”). First, the two coders independently identified the units of informa-
tion sharing. One coder identified 45 turns and the other coder identified 43.
Forty-one of these talking turns were common for the two coders, and these
turns were used in the subsequent analyses. The coders then independently
classified the units as either giving information or asking for information.
Cohen’s kappa was .98 for the two coders. Disagreements in classification
were resolved through discussion. Information-sharing variables were
positively skewed. Consequently, we transformed the information variables
by adding a constant of 1 and then calculating the square root twice
(cf. Tabachnick & Fidell, 1996). This resulted in relatively low skewness
(< 1.26), except for the buyers’ asking for information (which only occurred
two times in total).
Offers included both single-issue offers and multiple-issue offers and
were coded separately for sellers and buyers. Single-issue offers were defined
as propositions on one of the three issues (e.g. “I claim eight product
variations”). Multiple-issue offers were defined as propositions which
combined at least two of the issues (e.g. “If I get eight product variations,
I can give you 2–3 weeks on delivery time”). Using the same procedure
as for information sharing, the two independent coders agreed on 216
talking turns (out of 226 for both coders). The coders then categorised each
of the 216 turns as either single-issue or multiple-issue. Cohen’s kappa was
.93 for offers. Coders resolved disagreements in classification through
discussion.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 81
RESULTS
All dyads reached an agreement. Descriptive statistics and correlations are
shown in Table 1.

Outcomes
We used hierarchical regression analyses to test how the sellers’ CM and
buyers’ CM impacted outcomes (Hypotheses 1a and 1b). We did three
regression analyses. Dependent variables in the three regressions were
sellers’ outcome, buyers’ outcome, and joint outcomes, respectively. The
independent variables in all three regressions were the cognitive motivation
of the sellers (sellers’ CM), the cognitive motivation of the buyers (buyers’
CM), and the interaction of the cognitive motivation of the sellers and the
buyers. Sellers’ CM and buyers’ CM were entered in Step 1, and the inter-
action was entered in Step 2. The results from the three regression analyses
are summarised in Table 2.
First, we examined how the CM of the sellers and the buyers impacted
joint outcomes. The results are displayed in the right hand columns of Table 2.
The model was statistically significant, F(2, 27) = 6.35, p < .01, R2 = .32
(Step 1). Sellers’ CM had a significant positive effect on joint outcomes
(β = .56, p < .01), but buyers’ CM did not affect joint outcomes (β = −.17,
ns). The product term added in Step 2 yielded no significant interaction
effect. Second, we examined the effects of the sellers’ and the buyers’ CM
on the sellers’ outcome. The results are shown in the left hand columns
of Table 2. The model was statistically significant, F (2, 27) = 5.33, p < .05,
R2 = .28 (Step 1). Sellers’ CM had a significant positive impact on own
outcome (β = .53, p < .01), while buyers’ CM did not affect the sellers’
outcome (β = −.15, ns). The interaction effect was not significant. Third, we
examined the effects of the sellers’ and the buyers’ CM on the buyers’
outcome. The results are summarised in the middle columns of Table 2. The
regression was not significant, F (2, 27) = 0.00, ns (Step 1), and neither the
buyers’ own CM, nor the sellers’ CM, nor the interaction, had any impact
on buyers’ outcome. These results support Hypotheses 1a and 1b. Sellers’
CM (but not the buyers’ CM) has a positive impact on own individual
outcome and joint outcomes.

Process
We used the procedure recommended by Baron and Kenny (1986) to test
for mediation (Hypothesis 2). A variable functions as a mediator when
(i) the independent variable affects the mediator, (ii) the mediator affects the
dependent variable, and (iii) when the effects in (i) and (ii) are controlled

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
Psychology.
© 2006 The Authors. Journal compilation © 2006 International Association for Applied

82
TABLE 1
Descriptive Statistics and Correlations for Study Variables

SCHEI ET AL.
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Sellers
1. Individual outcome 4694 764
2. Cognitive motivation 3.56 0.60 .52**
3. Give information 0.88 1.36 .38 .49*
4. Ask information 0.53 0.72 .22 .41 .28
5. Single offers 2.18 2.04 −.34 −.62** −.21 −.17
6. Multiple offers 4.65 2.74 −.08 .10 .11 −.25 .11

Buyers
7. Individual outcome 4700 695 −.44* .03 .43 .15 −.14 −.19
8. Cognitive motivation 3.46 0.46 −.08 .13 .03 −.20 .28 .14 −.01
9. Give information 0.88 1.11 .49* .31 .69** .18 −.20 −.27 .13 −.16
10. Ask information 0.12 0.49 .15 .02 .42 −.20 −.28 .13 .19 −.53* .59*
11. Single offers 2.06 1.71 −.57* −.73** −.51* −.28 .55* −.29 .09 .29 −.32 −.31
12. Multiple offers 3.82 2.43 −.09 .15 −.01 −.18 .03 .69** −.11 .12 −.08 .02 −.16

Dyads
13. Joint outcomes 9394 773 .59** .54** .70** .33 −.45 −.22 .46** −.09 .59* .29 −.50* −.17
14. Cognitive motivation 3.51 0.40 .37* .83** .40 .21 −.33 .15 .03 .67** .15 −.27 −.41 .18 .39*
15. Give information 1.76 2.25 .49* .44 .91** .27 −.22 −.12 .30 −.03 .92** .50* −.43 −.06 .71** .33
16. Ask information 0.64 0.79 .29 .40 .52* .81** −.32 −.16 .26 −.50* .51* .41 −.44 −.16 .48* .04 .55*
17. Single offers 4.24 3.31 −.50* −.76** −.39 −.25 .90** −.08 −.05 .32 −.29 −.33 .86** −.06 −.54* −.41 −.36 −.43
18. Multiple offers 8.47 4.75 −.09 .14 .05 −.24 .08 .93** −.16 .14 −.19 .08 −.25 .91** −.21 .18 −.10 −.18 −.08

Note: N = 31 dyads for outcomes and cognitive motivation, n = 17 (tape-recorded dyads) for the process variables. Descriptive statistics reported are based on variables
before recoding and transformations, while correlations are based on recoded and transformed variables. Cognitive motivation of the dyads is based on the average of the
sellers’ and the buyers’ cognitive motivation.
* p < .05; ** p < .01.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 83
TABLE 2
Regression Analyses on Outcomes for Sellers, Buyers, and Dyads

Sellers’ outcome Buyers’ outcome Joint outcomes

Cognitive motivation (CM) Step 1 Step 2 Step 1 Step 2 Step 1 Step 2

Sellers’ CM .53** .53** .01 .03 .56** .57**

Buyers’ CM −.15 −.16 −.01 .02 −.17 −.14
Sellers’ CM × Buyers’ CM −.05 −.23 −.18
R2 .28 .28 .00 .05 .32 .35
F for R2 5.33* 3.46* 0.00 0.49 6.35** 4.70**
∆R2 .00 .05 .03
F for ∆R2 0.07 1.46 1.27

Note: N = 31. Standardised beta coefficients are shown.

* p < .05; ** p < .01.

for, a significant relation between the independent and the dependent

variable is no longer significant. Therefore, we next present the results
for the relationship between (i) CM and process, (ii) process and joint
outcomes, and (iii) CM and joint outcomes when the above effects are
controlled for.

Cognitive Motivation and Process. We first examined how CM affected

the negotiation process. Sellers’ CM and buyers’ CM were correlated with
each of the four process variables (giving information, asking for informa-
tion, single-issue offers, and multiple-issue offers). These correlations were
calculated separately for the sellers, the buyers, and the dyads (see Table 1).
For all correlations reported in this section: n = 17. The correlations for
giving information show that only the sellers’ CM had a positive effect on
information giving. Sellers’ CM significantly affected the sellers’ giving
information (r = .49, p < .05), and significantly (marginally) affected the
total information giving in the dyads (r = .44, p < .10). Buyers’ CM did
not affect the information giving in the negotiation. Regarding asking
for information, sellers’ CM had a positive (but not significant) effect on
the sellers’ asking for information, and on the total asking for information
in the dyad. Buyers’ CM, on the other hand, had a significant negative
impact on the buyers’ asking for information (r = −.53, p < .05), and on
the dyads’ total asking for information (r = −.50, p < .05). However, recall
that buyers hardly asked for information. Regarding single-issue offers,
sellers’ CM had a significantly negative effect on how often single-issue
offers were used by the sellers themselves (r = −.62, p < .01), the buyers (r
= −.73, p < .001), and the dyads (r = −.76, p < .001). Conversely, buyers’ CM
did not affect the use of single-issue offers by the sellers, the buyers, or the

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
84 SCHEI ET AL.

dyads. Finally, the correlations for multiple-issue offers were not significan-
tly correlated with CM. Neither the sellers’ CM nor the buyers’ CM could
explain the use of multiple-issue offers by sellers, buyers, or dyads.

Process and Joint Outcomes. Next, we examined how the process vari-
ables were related to the joint outcomes (see Table 1). Giving information
was positively and significantly related to joint outcomes, regardless of who
gave the information. Joint outcomes were enhanced by the information
giving of the sellers (r = .70, p < .01), the buyers (r = .59, p < .05), and the
total information giving in the dyads (r = .71, p < .001). Asking for informa-
tion was also positively and significantly related to joint outcomes when
we examined the aggregated level of asking for information in each dyad
(r = .48, p < .05). However, asking for information done by either the seller
or the buyer alone did not influence the joint outcomes significantly. Single-
issue offers were negatively related to joint outcomes. The number of single-
issue offers from sellers (r = −.45, p < .10), from buyers (r = −.50, p < .05),
and the total number of single-issue offers in the dyad (r = −.54, p < .05)
were all negatively related to joint outcomes. Finally, multiple-issue offers
from sellers, from buyers, or in the dyad as a whole were not significantly
correlated with joint outcomes.

Mediation. The analyses above show that two of the four dyadic pro-
cess variables are potentially mediators. Giving information and single-issue
offers are related both to the independent variable sellers’ CM (although
only marginally so for giving information at the dyadic level), and to the
dependent variable joint outcomes. In order to demonstrate mediation, we
needed to regress the dependent variable on both the independent variable
and on the mediators. We did a hierarchical regression analysis on joint
outcomes, entering sellers’ CM in Step 1, and giving information and single-
issue offers in Step 2. The variables entered in Step 2 mediate if they had a
significant effect on the dependent variable, and if the independent variable
entered in Step 1 dropped from significance in Step 1 to non-significance in
Step 2. The result from the regression analysis is shown in Table 3. The
effect of sellers’ CM on joint outcomes dropped from significance (β = .54,
p < .05) in Step 1, to non-significance (β = .07, ns) when the proposed
mediators were entered in Step 2. Furthermore, giving information had a
significantly positive effect on joint outcomes (β = .58, p < .01), while single-
issue offers were negatively, but not significantly, related to joint outcomes
(β = −.28, ns). The explained variance increased significantly when the anti-
cipated mediators were entered, F(2, 13) = 5.22, p < .05, ∆R2 = .32. The
results largely support Hypothesis 2, since information giving mediated
between sellers’ CM and joint outcomes.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 85
TABLE 3
Regression Analyses for Mediating Effects on Joint Outcomes

Variables Step 1 Step 2

Sellers’ cognitive motivation .54* .07

Give information (dyad) .58**
Single offers (dyad) −.28
R2 .29 .61
F for R2 6.11* 6.66**
∆R2 .32
F for ∆R2 5.22*

Note: n = 17 (tape-recorded dyads). Standardised beta coefficients are shown.

* p < .05; ** p < .01.

DISCUSSION
The results indicate that cognitive motivation and role interact to affect
processes and outcomes in integrative negotiations. High cognitive motiva-
tion triggers an integrative negotiation process that is characterised by
information giving that results in high joint outcomes. However, cognitive
motivation does not influence negotiation in a simple and uniform way.
Our results suggest strong role effects. We found that joint outcomes were
enhanced by high cognitive motivation only among sellers. Sellers with high
cognitive motivation compared to sellers with low cognitive motivation
were more integrative (giving information) and less distributive (avoiding
single-issue offers). Furthermore, they succeeded in reducing the use of
single-issue offers in the dyads. However, the process that was triggered by
highly cognitively motivated sellers did not affect individual outcomes for
the buyers. Thus, the increase in joint outcomes came as a direct conse-
quence of the increase in the individual outcome for the sellers. Cognitive
motivation among buyers did not affect the negotiation outcomes.
In general, our results support an information processing perspective on
negotiation (De Dreu, Weingart, & Kwon, 2000). In order to exploit the
integrative potential in negotiations, the parties must be able to integrate the
private information about preferences held by each negotiator. The new
knowledge structure that is generated includes information about the rela-
tionship between the preferences of the negotiators, and helps the negoti-
ators to uncover possible agreements that produce joint outcomes better than
compromises would. A motivated information processing perspective on
negotiation (De Dreu, 2003) implies that individuals vary in the degree to
which they apply themselves to active information search and use. Previous
research has shown that both situational and dispositional factors influence

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
86 SCHEI ET AL.

information processing. For example, accountability seems to increase

information processing (De Dreu, Koole, & Steinel, 2000), while time pres-
sure inhibits motivation to search for and use information in negotiation
(De Dreu, 2003). Furthermore, social motives, as induced by social cues or
by individual dispositions, also influence motivated information search. In
general, negotiators with prosocial motives seem to be more motivated to
collect and use information from their counterparts than negotiators with
egoistic motives (De Dreu, Weingart, & Kwon, 2000). Finally, need for
cognitive closure, which is found to be negatively, but weakly (r = −.26)
correlated with cognitive motivation (Cacioppo et al., 1996), inhibits mot-
ivated information processing (De Dreu, Koole, & Oldersma, 1999). The
focus on cognitive motivation in our study complements previous research
that motivated information processing is an important element in integrative
negotiation.
Our results suggest a strong person–situation interaction regarding
the effects of cognitive motivation in negotiation. This is consistent with
dispositional research in organisational behavior which has in general moved
towards person–situation interactions (e.g. George, 1992) and towards
cross-level research (e.g. Goodman, 2000). Moreover, the present results
complement the findings of other negotiation scholars who have studied
personality (e.g. Barry & Friedman, 1998). Although historically, personality
has been seen as not very predictive in negotiations (Thompson, 1990), our
research suggests that personality does matter in negotiation; yet it interacts
with situational factors. Future negotiation research will benefit from exam-
ining person (i.e. cognitive motivation)–situation (i.e. roles) interactions.
Future research is also needed to identify the underlying relationships
between roles and cognitive motivation in negotiation. We posited that the
reasons for the differential effects of cognitive motivation across roles are
found in different role expectations for buyers and sellers. Buyers may per-
ceive themselves as more powerful than sellers and frame the deals in terms
of losses. This may lead buyers not to be overly affected by the negotiation
process, and to be more influenced by their pre-negotiation expectations.
Thus, their prime task is not to gather and use information (i.e. to identify
the preferences of the sellers and to relate them to their own). Rather,
buyers state their demands, hold out, and risk impasses (Brett, 2001). The
sellers, however, perceiving themselves as less powerful than the buyers and
being positively framed, may be more eager to create agreements through
the negotiation process. When they have low cognitive motivation they
may do this more through concessions and less through information
processing. On the other hand, when the sellers are highly cognitively
motivated their motivated information processing leads them to identify
agreements that increase their own outcome, without reducing the outcome
for the buyers.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 87

Limitations
Although the current study offers a potential of new insight on individual
differences in negotiations, its contributions must be evaluated with the
study limitations in mind. First, as always, it is necessary to be cautious
when generalising from data on students in a simulation. Professional nego-
tiators have been found to reach higher joint outcomes than students in
negotiation simulations (Neale & Northcraft, 1986). However, the study by
Neale and Northcraft indicated that a cognitive bias such as framing affects
professionals and students in similar ways. Thus, cognitive factors may have
similar effects on professionals and students. Another related potential
limitation is that the simulation only had limited personal consequences for
the participants. Although observations and discussions with the participants
indicated strong involvement, we cannot rule out the possibility that the
effects of cognitive motivation will change when more could be won or lost.
For example, cognitive motivation may become less important when the
personal involvement in the task is very high (e.g. personal involvement may
substitute for cognitive motivation). Further research can address this by
using, for example, a monetary reward in order to explore whether the
effects of cognitive motivation are the same on different levels of personal
involvement.
Furthermore, in the present study, cognitive motivation was measured.
Thus, we cannot completely discard that our findings may be caused by
other individual difference factors that co-vary with cognitive motivation.
Nevertheless, the design allowed us to test for the effects of naturally occur-
ring individual differences in cognitive motivation in a controlled setting.
Finally, as we measured cognitive motivation just before the negotiation
simulation, priming effects might be a problem. However, in our study the
students answered a questionnaire with several scales before they received
information about the negotiation simulation. In addition, they were informed
that their answers on the various scales would form the basis for personal
feedback on individual dispositions later in the course.

Practical Implications
Negotiation is part of everyday life. We negotiate in work settings and
privately. Unfortunately, we often fail to realise the integrative potential
that exists in many negotiation situations. The present research contributes
to the knowledge about how to improve the negotiation process and out-
comes. First, the results in this study affirm that negotiators must actively
focus on exchanging information. More specifically, they should actively
link the negotiation issues and try to understand how their own preferences
relate to those of the other negotiator. Thereby, they can uncover solutions

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
88 SCHEI ET AL.

that are better than simple compromises. Previous research has shown that
negotiators have poor understanding of the other party’s preferences, and
therefore do not recognise the integrative potential (Neale & Bazerman,
1991). Our research suggests that this problem can be overcome by the
energetic giving and asking for information that are linked to high cognitive
motivation. Second, our research suggests that, for sellers, high cognitive
motivation helps in overcoming barriers to achieve high joint outcomes in
negotiations. Certainly, the direct implication is that in integrative negotiations
one should select sellers that are high on cognitive motivation. However,
this may not always be a practical solution. Therefore, one may want to
identify situational factors that can help individuals, when they are not
inclined by disposition, to become cognitively motivated. Situational factors
that are likely to induce cognitive motivation include accountability and low
time pressure (cf. De Dreu, 2003). Finally, for the buyers, our results indicate
that increased cognitive motivation on its own does not help in integrative
negotiations. Perhaps informing buyers about role expectations and educat-
ing them about integrative negotiation could remove the obstacles that
hinder the positive effects of cognitive motivation on outcomes.

REFERENCES
Baron, R.M., & Kenny, D.A. (1986). The moderator–mediator variable distinction
in social psychological research: Conceptual, strategic and statistical considera-
tions. Journal of Personality and Social Psychology, 51, 1173–1182.
Barry, B., & Friedman, R. (1998). Bargainer characteristics in distributive and inte-
grative negotiation. Journal of Personality and Social Psychology, 74, 345–359.
Bazerman, M.H., Curhan, J.R., Moore, D.A., & Valley, K.L. (2000). Negotiation.
Annual Review of Psychology, 51, 279 –314.
Berzonsky, M.D., & Sullivan, C. (1992). Social-cognitive aspects of identity style:
Need for cognition, experimental openness and introspection. Journal of Adoles-
cent Research, 7, 140 –155.
Brett, J.M. (2001). Negotiating globally. San Francisco, CA: Jossey-Bass.
Brett, J.M., Shapiro, D.L., & Lytle, A. (1998). Breaking the bonds of reciprocity in
negotiations. Academy of Management Journal, 41, 410–424.
Cacioppo, J.T., Petty, R.E., Feinstein, J.A., & Jarvis, W.B. (1996). Dispositional
differences in cognitive motivation: The life and times of individuals varying in
need for cognition. Psychological Bulletin, 119, 197–253.
Cacioppo, J.T., Petty, R.E., & Kao, C.F. (1984). The efficient assessment of need
for cognition. Journal of Personality Assessment, 48, 306–307.
Cai, D.A., Wilson, S.R., & Drake, L.E. (2000). Culture in the context of intercul-
tural negotiation: Individualism–collectivism and paths to integrative agreements.
Human Communication Research, 26, 591– 617.
Chatterjee, S., Heath, T.B., Milberg, S.J., & France, K.R. (2000). The differential
processing of price in gains and losses: The effects of frame and need for cogni-
tion. Journal of Behavioral Decision Making, 13, 61–75.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
 COGNITIVE MOTIVATION AND NEGOTIATION ROLES 89
De Dreu, C.K. (2003). Time pressure and closing of the mind in negotiation. Organ-
izational Behavior and Human Decision Processes, 91, 280–295.
De Dreu, C.K., Koole, S.L., & Oldersma, F.L. (1999). On the seizing and freezing
of negotiator inferences: Need for cognitive closure moderates the use of heuris-
tics in negotiation. Personality and Social Psychology Bulletin, 25, 348–362.
De Dreu, C.K., 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., Weingart, L.R., & Kwon, S. (2000). Influence of social motives on
integrative negotiation: A meta-analytic review and test of two theories. Journal
of Personality and Social Psychology, 78, 889 –905.
Drake, L.E. (2001). The culture–negotiation link: Integrative and distributive bar-
gaining through an intercultural communication lens. Human Communication
Research, 27, 317–349.
George, J.M. (1992). The role of personality in organizational life—issues and evid-
ence. Journal of Management, 18, 185 –213.
Goodman. P. (2000). Missing organizational linkages: Tools for cross level research.
Thousand Oaks, CA: Sage.
Gouldner, A.W. (1960). The norm of reciprocity: A preliminary statement. American
Sociological Review, 25, 161–178.
Gulgos, S. (2001). Need for cognition and cognitive performance from a cross-
cultural perspective: Examples of academic success and solving anagrams. The
Journal of Psychology: Interdisciplinary and Applied, 135, 100–112.
Kassin, S.M., Reddy, M.E., & Tulloch, W.F. (1990). Juror interpretations of ambi-
guous evidence—the need for cognition, presentation order, and persuasion. Law
and Human Behavior, 14, 43 –55.
Lassiter, G.D., Briggs, M.A., & Slaw, R.D. (1991). Need for cognition, causal
processing, and memory for behavior. Personality and Social Psychology Bulletin,
17, 694 –700.
Leary, M.R., Sheppard, J.A., McNeil, M.S., Jenkins, T.B., & Barnes, B.D. (1986).
Objectivism in information utilization: Theory and measurement. Journal of Per-
sonality Assessment, 50, 32 – 43.
Lewicki, R.J., Barry, B., Saunders, D.M., & Minton, J.W. (2003). Negotiation. New
York: McGraw-Hill.
Nair, K.U., & Ramnarayan, S. (2000). Individual differences in need for cognition
and complex problem solving. Journal of Research in Personality, 34, 305–328.
Neale, M.A., & Bazerman, M.H. (1985). The effects of framing and negotiator
overconfidence on bargaining behaviors and outcomes. Academy of Management
Journal, 28, 34 – 49.
Neale, M.A., & Bazerman, M.H. (1991). Cognition and rationality in negotiation.
New York: Free Press.
Neale, M.A., Huber, V., & Northcraft, G.B. (1987). The framing of negotiation:
Contextual vs. task frames. Organizational Behavior and Human Decision Pro-
cesses, 39, 157–175.
Neale, M.A., & Northcraft, G.B. (1986). Experts, amateurs and refrigerators: Com-
paring experts and amateur negotiators in a novel task. Organizational Behavior
and Human Decision Processes, 38, 305 –317.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.
90 SCHEI ET AL.

Neale, M.A., & Northcraft, G.B. (1991). Behavioral negotiation theory: A frame-
work for conceptualizing dyadic bargaining. Research in Organizational Behavior,
13, 147–190.
Olson, K., Camp, C., & Fuller, D. (1984). Curiosity and need for cognition. Psycho-
logical Reports, 54, 71–74.
Osberg, T.M. (1987). The convergent and discriminant validity of the need for cog-
nition scale. Journal of Personality Assessment, 51, 441–450.
Pruitt, D.G. (1981). Negotiation behavior. New York: Academic Press.
Pruitt, D.G., & Carnevale, P.J. (1993). Negotiation in social conflict. Buckingham:
Open University Press.
Pruitt, D.G., & Lewis, S.A. (1975). Development of integrative solutions in bilateral
negotiation. Journal of Personality and Social Psychology, 31, 621–633.
Smith, S.M., & Levin, I.P. (1996). Need for cognition and choice framing effects.
Journal of Behavioral Decision Making, 9, 283 –290.
Sorrentino, R.M., Bobocel, D.R., Gitta, M.Z., Olson, J.M., & Hewitt, E.C. (1988).
Uncertainty orientation and persuasion: Individual differences in the effects
of personal relevance on social judgements. Journal of Personality and Social
Psychology, 55, 357–371.
Tabachnick, B.G., & Fidell, L.S. (1996). Using multivariate statistics. New York:
HarperCollins.
Thompson, L. (1990). Negotiation behavior and outcomes: Empirical evidence and
theoretical issues. Psychological Bulletin, 108, 515–532.
Tripp, T.M., & Sondak, H. (1992). An evaluation of dependent variables in experi-
mental negotiation studies: Impasse rates and Pareto efficiency. Organizational
Behavior and Human Decision Processes, 51, 272–295.
Tutzauer, F., & Roloff, M.E. (1988). Communication processes leading to integra-
tive agreements: Three paths to joint benefit. Communication Research, 15, 360–
380.
Weingart, L.R., Thompson, L., Bazerman, M.H., & Carroll, J.S. (1990). Tactical
behavior and negotiation outcomes. International Journal of Conflict Manage-
ment, 1, 7–31.

© 2006 The Authors. Journal compilation © 2006 International Association for Applied
Psychology.