Multiparty Negotiation Support

MULTIPARTY NEGOTIATION SUPPORT 129

Journal of Management Information Systems / Summer 2002, Vol. 19, No. 1, pp. 129–150.

© 2002 M.E. Sharpe, Inc.

0742–1222 / 2002 $9.50 + 0.00.

Multiparty Negotiation Support:The Role of Visualization’s Influence on theDevelopment of Shared Mental Models

RODERICK I. SWAAB, TOM POSTMES, PETER NEIJENS,MARIUS H. KIERS, AND ADRIE C.M. DUMAY

RODERICK I. SWAAB is a doctoral candidate at the Amsterdam School of Communica-tions Research. He received his master’s degree in communication and organizationsat the University of Amsterdam. His research focuses on the role of information andcommunication in negotiation processes.

TOM POSTMES is a senior lecturer at the School of Psychology, University of Exeter.He holds a doctorate in social psychology from the University of Amsterdam. He isaffiliated with the Department of Communication Science, University of Amsterdam,and holds a fellowship of the Royal Netherlands Academy of Arts and Sciences(KNAW). He is currently an Associate Editor of the British Journal of Social Psy-chology. His research and teaching is concerned with computer-mediated communi-cation, social influence in groups, and collective behavior.

PETER NEIJENS is a senior professor in communications at the University of Amsterdamand scientific director of the Amsterdam School of Communications Research. Heholds a doctorate in methodology and statistics from the University of Amsterdam.His research focuses on corporate communication, decision-making, communicationcampaigns and media effects, and public opinion.

MARIUS H. KIERS is a senior consultant at TNO building and construction research.He earned his master’s degree in economics at the Free University Amsterdam. Hespecializes in innovation and technology management, as well as in spatial econom-ics and regional development.

ADRIE C.M. DUMAY is the head of medical informatics of TNO Prevention and Health.He holds a doctorate in information and communication theory from Delft Univer-sity of Technology. The research was carried out while he was head of the VirtualityLab of TNO. His present work focuses on simulation and information systems inmedicine.

ABSTRACT: The study examines a method for supporting multiparty negotiations bymeans of a Negotiation Support System (NSS). More specifically, this study investi-gated the effect of visualization support on the development of shared mental modelsamong negotiators who resolved a spatial planning dispute. The objective of this studyis to determine how to support the development of shared mental models in order tostimulate more productive negotiations. A further goal is to provide guidelines for thedesign of NSS. Compared with a control condition, visualization improved three

130 SWAAB ET AL.

aspects of negotiations: visualization support aided negotiators’ convergence of per-ceptions of reality and had positive socio-emotional consequences in terms of in-creasing cohesiveness and entitativity. As a result, groups with visualization supportreached consensus more easily and were more satisfied with the process. In sum, thecurrent study provides support for the idea that presenting negotiators with unam-biguous information helps negotiators develop shared mental models.

KEY WORDS AND PHRASES: multiparty negotiation, negotiation support systems, pro-social climate, shared mental model, visualization of information.

THE PAST DECADE HAS SEEN A RISING INTEREST in what was called the “group mind”at the end of the nineteenth century. Unfortunately, the research in this area of collec-tive sense-making has not yet lead to consensus about the conceptualization of suchforms of collective cognition. So, several terms are used to describe the processes bywhich multiple individuals make collective sense of their surroundings. Socialsharedness [45], mental models [40], collective minds [48], shared situation models[35], socially shared cognition [38], and shared mental models [23] are all conceptsthat refer to how dyads, groups, and larger collectives gain an understanding of phe-nomena that is not merely grounded in individual cognition but also consensual amonggroup members.

However, there are also commonalties across definitions. All views recognize theexistence of knowledge structures at a supra-individual level, for example, relating toaspects of the group’s context or task. In addition, they all stress the potential produc-tivity that such socially shared knowledge structures have for the group. For example,it has been argued that a shared understanding helps group members to coordinatetheir collaboration [7, 24, 34, 36]. The concept of “shared mental models” is one ofthe most widely used terms for referring to these socially shared knowledge struc-tures [23, 40]. A mental model is a mental representation of a system that can be usedto describe, explain, and predict future system states [40]. Mental models allow indi-viduals to understand phenomena, make inferences, and experience events by proxy[18]. They refer to a general class of cognitive constructs that have been invoked toexplain how knowledge and information are presented in the mind [23]. Several sortsof shared mental models may be important for group functioning: models about thegroup, models about group members, and models about work or the task at hand [24,34, 36, 43].

In this paper we shall focus especially on the last type of shared mental model:models about the task at hand (in this case a negotiating task). The idea is that ashared mental model of the task serves to organize the beliefs of individuals about thenature of a problem and possible solutions in a given situation [31, 47]. Thus, in thepresent research we define a shared mental model as the amount of overlap in knowl-edge among different negotiators about the problem or task they face, and its possiblesolutions.


Shared Mental Models in Negotiation

WE ARGUE THAT NEGOTIATION OUTCOMES are likely to improve to the extent thatmental models are more shared. It is often the case within negotiations that differentparties see what they expect to see due to different knowledge they possess a priori.Thereby, the motives they have during the negotiation also influence their percep-tions. So, it is very likely that negotiators have different conceptualizations of theproblem situation, and inconsistencies between their cognitions about both the prob-lem and potential solutions may foster further disagreement during negotiations overand above any other disagreements that may exist between them. If, on the otherhand, negotiators have similar knowledge of the problem, understand its nature insimilar ways, and agree about the potential solutions, one could envisage there arefewer obstacles that stand in the way of settlement. In addition, parties may recognizethe common ground among them, which could form the basis for convergence.

We expect that shared mental models have two consequences in negotiations. In thefirst place, if mental models are shared, negotiators will interpret any informationabout the problem from a common viewpoint, and this is likely to stimulate settle-ment or the achievement of consensus about what are acceptable or less acceptablesolutions [20, 43, 45].

Second, we expect that due to the existence of a shared mental model, negotiatorswill recognize that they and their opponents have similar views on certain issues. Therecognition of similarity is generally a good predictor of social attraction and is likelyto foster the development of a common identity of some form [16]. Both liking andcommon identity reflect a pro-social climate [11, 14, 15, 25]. This pro-social climateis a strong precursor to mutually satisfactory negotiation and decision-making [5, 9,37]. Although we do not attempt to influence the pro-social climate directly in thisstudy, we do think that it could be affected by shared mental models, such that moreoverlap in cognitions has a positive affective outcome for a group of negotiators.

In sum, we argue that the influence of shared mental models on negotiation out-comes is positive for two reasons. On one hand, a shared mental model influencesnegotiation outcomes directly by providing common views of the problem at handand potential solutions, thereby facilitating consensus. On the other hand, we specu-late that shared mental models might affect the pro-social climate positively, which inturn should have a positive impact on negotiation outcomes.

Supporting the Development of Shared Mental Models

NOW THAT WE HAVE ILLUSTRATED THE IMPORTANCE of shared mental models fornegotiation outcomes, a further question is how one could stimulate the developmentof shared mental models during negotiations.

The overall idea is that the development of shared mental models could be facili-tated by the presentation of information that is available prior to and during the nego-tiation process. This information should be designed and presented in such a way thatit leads to a convergence of perceptions, for example, by aiding negotiators to define

132 SWAAB ET AL.

a problem unambiguously. Thus, we argue that controlling the presentation and avail-ability of information is an important tool for aligning the interpretations and percep-tion of the problem, and thus for the development of a shared mental model.

Group information processing models may inform us about how to provide nego-tiators with information [17]. According to such models of information processing,people process different activities. First, group members perceive the informationthey are exposed to. After this attention phase, group members encode the informa-tion by structuring, evaluating, interpreting, and transforming the information into amental representation. Hereafter, the information will be stored in memory to be re-trieved later in the negotiation process.

Because encoding involves negotiators to structure and interpret the information inorder to transform it into knowledge, it functions as a key element in producing cog-nitions of the problem and possible solutions. Therefore, if negotiators structure andinterpret the information in similar ways, knowledge about the problem and possiblesolutions will also be more shared. On the other hand, if negotiators interpret thesituation differently, cognitions are likely to diverge, which may hinder the negotia-tion process. Thus, the challenge is to stimulate similar encoding processes amongnegotiators in order to develop shared knowledge toward the problem and possiblesolutions. Because the encoding phase is preceded and influenced by the attention toavailable information, we argue that one possibility for stimulating convergence be-tween negotiators could be to intervene within this phase. Therefore, we suggest thatpresenting them with unambiguous information stimulates similar structuring andrepresentation processes and thereby supports the development of shared mentalmodels.

One possibility to provide information unambiguously is to use visual representa-tion aids, which are known to have a strong impact on the problem perception ingroups [3, 31]. These visual representation aids provide ways of presenting and ma-nipulating a problem situation by means of images, charts, maps, and so on [49]. Theunderlying idea behind the powerful impact of such representational aids on cogni-tions is that they can be comprehended quite easily and that they tend to minimize thescope for divergent interpretation compared with textual and verbal information ex-change [39]. The graphical forms can provide a common framework that is a basis fornegotiators in interacting with each other and communicating proposals and preferredsolutions [31]. Thereby, graphical representations can also diminish the time it takesnegotiators to verbalize their perceptions of the problem. So, attention should befocused on perceiving the visual representation, rather than relying solely on logicalinferences based on interpretations of language [26]. The idea is that discussion evokedby the representational aids will lead negotiators to realize which information is ab-sent and important and which is present and unimportant [1].

In sum, by unambiguous visualization of information regarding the problem andpossible solutions, negotiators are able to encode the information more similarly. Asa result, mental models will be more shared and we propose that this will be associ-ated with other negotiation outcomes.


Summarizing the ideas above in the form of hypotheses to be tested, we expect thatvisualization support affects four distinct classes of negotiation outcomes. The firstoutcome is the shared mental model of negotiators. Based on the theoretical analysisabove, a main objective of negotiation support could be to foster a common under-standing among negotiators about the nature of problem, and possible solutions [31,43, 45]. This common understanding could be triggered by unambiguous visualiza-tion of a problem, thereby providing negotiators with similar perceptions of the prob-lem at hand and of possible solutions. Due to well-designed visual aids, shared mentalmodels should arise more easily.

H1: Visualization support fosters the development of a shared mental model amongnegotiating parties, compared with negotiations without visualization support.

As a result of the like-mindedness created by shared task mental models, common-alities among negotiators will become more salient and this perceived similarity islikely to foster a pro-social climate. To measure the pro-social climate, we distin-guished two different variables. One variable related to the pro-social climate is thedegree to which negotiating parties are bonded to each other. Negotiation supportshould increase feelings of attraction for specific group members and for theoverarching group itself, and increase the degree to which negotiators coordinatetheir efforts to achieve common goals. These properties are commonly known ascohesion [8, 12].

H2a: Visualization support increases cohesion among negotiating parties com-pared with negotiations without visualization support.

A second variable closely related to the pro-social climate is the perception of acommon identity. In this study, common identity was operationalized with a measureof entitativity [6, 13]. This concept captures the degree of perceived unity amongnegotiators, such that strong entitativity implies that differences between negotiatorsare small. As such, entitativity can be seen as a measure of common in-group identity[32]. Parallel to the reasoning for the previous hypothesis, support should increaseentitativity.

H2b: Visualization support increases entitativity among negotiating parties com-pared with negotiations without visualization support.

We argue that both variables (shared mental model and a pro-social climate) havepositive effects on the negotiation outcome. One key outcome of any negotiation isthe reduction of interparty conflict, thereby increasing possibilities for an agreement.Typical for conflict situations is the existence of conflicting interpretations of theproblem and especially its solutions. Due to potential effects of both shared mentalmodel and a pro-social climate, we expect that visualization support will have posi-tive effects on this negotiation outcome. The underlying idea is that visualizationhelps the creation of a shared mental model, which together with a pro-social climate,builds the foundation for a common agreement in the form of interparty consensus.

134 SWAAB ET AL.

H3: Visualization support increases consensus among negotiating parties com-pared with negotiations without visualization support.

To examine a negotiator’s evaluation of the method of visualization support, threedifferent satisfaction variables are measured. Because satisfaction is closely relatedto people’s appreciation of the quality of negotiations as a social process, we expectthat satisfaction will be greater when there is visualization support. We distinguishbetween satisfaction with the outcome, the process, and the facilitator’s role. Thesethree may be interrelated, but measure conceptually different aspects the support.

H4a: Visualization support increases satisfaction with the outcome of negotia-tion among negotiating parties compared with negotiations without visualiza-tion support.

H4b: Visualization support increases satisfaction with the process of negotia-tion among negotiating parties compared with negotiations without visualiza-tion support.

H4c: Visualization support increases satisfaction with the facilitator who sup-ports negotiation among negotiating parties compared with negotiations withoutvisualization support.

Method

Participants and Design

PARTICIPANTS WERE RECRUITED from different social science departments in the Neth-erlands. Fifty-six third year students (33 men and 23 women) participated in the ex-periment. There were seven groups in every condition, with four persons per group.The average age was 24 years. Participants received no financial reward for theirparticipation. They were given a guided tour through the Virtuality laboratory. Par-ticipants were randomly assigned to experimental conditions. Although some groupmembers knew each other, most did not. None of the group members had a pasthistory of working together within their particular group. The experiment had oneindependent variable, the type of negotiation support: groups were either supportedby visualization or they were not. Groups in both conditions performed the sametasks and communicated verbally.

Negotiation Task

Participants performed a negotiation task in which each individual represented a dif-ferent party with conflicting interests. The purpose of the task was for each party toexplore the possibility of reaching a joint agreement that would maximize their re-turns. The negotiation task was a simulated business game based on the real-life ne-gotiations surrounding environmental planning and water management in the harbor


of Rotterdam, a negotiation in which several parties with conflicts of interest neededto reach consensus. In the game, four of these parties negotiated whether joint invest-ments should be made in buildings and factories, new roads, or a network of pipe-lines. Every participant played a different role, and they had to maximize the resultfor their party. Two participants played representatives of an industrial organization,which we shall call X and Y, both oil refineries. A third participant negotiated onbehalf of the Ministry of Transport, Public Works, and Water Management, and thefourth participant represented an environmental protection organization.

All the parties had plans within this area, but settlement depended on the ministryto subsidize (or sometimes to not penalize) these plans. Plans were displayed in thesimulation model in terms of their financial and environmental returns. Althoughboth industrial organizations wanted to expand, the nature of their desired expansiondiffered strongly. For example, the small industrial organization wanted to build newroads because their plant was isolated from the rest of the area. The other industrialorganization, a big multinational refinery wanted to expand their buildings and facto-ries. So, interests of industrial organizations conflicted in the sense that they bothwanted something else. Further, they differed in their core business as the smallerorganization was a shipping company and the latter an oil refinery organization. Al-though industrial policies would have financial benefits, both had negative environ-mental consequences. The objective of the small industrial organization was to persuadethe ministry to subsidize part of their costs because they did not have sufficient finan-cial means to realize their plans. However, the large industrial party did have enoughfinancial resources, it depended on the ministry because they needed building li-censes. The third party involved—an environmental protection organization—hadextremely different plans and wanted to realize a network of pipelines for efficientheat and energy reuse. In terms of financial and environmental returns, this plan wasenvironment-friendly, but financially less attractive for the two industrial organiza-tions. The problem for the environmental protection organization was that they hadno budget available for the realization of their plans, and were thus dependent oninvestments of the ministry and of the industrial organizations. Their objective was topersuade them.

The ministry played a different role than the other three actors. The ministry’s mainobjective was not to stimulate investment in this or that, but to develop a sharedpolicy in the harbor area. This policy should meet two requirements: it should im-prove the economy of the region, and contribute to a cleaner and healthier environ-ment. The ministry’s representative had no concrete objectives, such as the otherthree organizations, but had the power to subsidize plans that would improve thesituation in the region. Moreover, although the ministry did not have veto power, theycould penalize plans that would be harmful to the environment, by imposing the so-called ecotax. This is extra taxation on environmental pollution, and would decreasethe financial performance of the industrial organizations. The idea underlying thetask that parties with interdependent outcomes negotiated about an issue, but thatconflicts of interests hindered the required settlement, and the “best” solution wouldbe a win-win solution, based on the different objectives.

136 SWAAB ET AL.

Procedure

At the moment of arrival, participants were asked to join the negotiating table. Eachof the participants were asked to read the instructions, explaining in detail their rolein the game, giving them the background of their organization, and the objective oftheir party. Within this background information, participants were given private in-formation about the costs and benefits of negotiation outcomes for them. Also, theywere given directives for their negotiation strategy. After reading the instructions,participants were given the full description of the problem at hand, introducing theissue. After that, they filled out a short pretest questionnaire. This took 15 minutes.After the introduction, the facilitator informed negotiators that their objective was toexplore the possibilities of a shared policy, and presented three possible scenarios.These three scenarios represented the optimal plans of three of the parties involved:refinery X, refinery Y, and the environmental protection organization. This introduc-tion to the parties’ optimal scenarios took 20 minutes.

The participants were then asked by the facilitator to comment on these scenarios.Subsequently, they negotiated and explored the possibilities for a shared policy. Themaximum time they could use for the negotiations was one hour. After 60 minutes ofnegotiating, participants were asked if they wanted to make a decision now, or con-tinue with the negotiations in the future (one of the groups in the unsupported condi-tion didn’t want to continue future negotiations at all). The negotiations were thenterminated, and participants were asked to fill in a posttest questionnaire. The facili-tator was present at all times to structure the meeting and, if necessary, as an instiga-tor of the discussion.

Independent Variable: Type of Negotiation Support

Negotiators were supported with a visualization system, the Virtual Business System(VBS), developed by the Virtuality Group at the Netherlands Organization for Ap-plied Scientific Research (TNO). The system uses simulation models that synchro-nize with the visualization model. So, negotiators could say exactly where and towhat extent they wanted to penalize their plans. For example, the big industrial orga-nization was able to visually demonstrate his plans within the area of discussion.Therefore, he could draw new factories on a big screen, whereas environmental andfinancial profits were rendered immediately. All parties involved could see these ef-fects, and were able to comment on the scenarios in a structured way. Participantswere abled to form their own representations before discussing the problem alto-gether within the group (see also [31] for a similar procedure). In the second place,interpersonal interaction in both experimental conditions went via face-to-face dis-cussion, accounting for a “rich” communication climate, required for negotiation sup-port [29]. The visualization could be in a two-dimensional, three-dimensional, orfour-dimensional (three-dimensional graphics with modifications over time) format.

The manipulation of visualization support consisted of letting negotiators use theVBS in the support condition, or not in the control condition. Practically, this meant


that although all information was presented on one screen (visible to all) in the sup-port condition, multiple sources of information were used in the control condition.

Players in the visualization support condition were supported by projections on abig semicircular, 150-degree screen with presentations depicting the environment,alterations made to it, and the economic and environmental consequences. They satwith each other around a table, facing the screen, and were supported by an indepen-dent facilitator, a whiteboard, and flip-over (see Figure 1). Participants in the no-support control condition met each other without any automated aids for visualizationsupport, and sat around a big table, also supported by the independent facilitator, awhiteboard, and flip-over. In this configuration, an overhead projector was present.On the wall, behind the facilitator, a two-dimensional map of the whole area waspresent. The participants also had a little map on their desks, with a two-dimensionalview of the harbor area.

A further difference between conditions was that players in the support conditioncould interact with the VBS system. Their proposals for alterations (such as buildingroads or investments made) were passed to the system by the facilitator, and immedi-ately visualized on the three-dimensional map of the environment (Figure 2).

Another section of the screen was reserved for simulations. When participants de-cided to invest, for example in new buildings, financial and environmental effectscould be visualized in this display (Figure 2). In contrast, in the control condition,participants were not able to make the adjustments to the shared display. Instead,each negotiator had a small-copied map on which they could make notes. In addition,an overhead display and flip-over were available to share the information about vari-ous simulations of economic and environmental effects.

In sum, in the support condition, all relevant information was displayed on onescreen, and players had the opportunity to visualize their proposals and gain someinsight into the consequences of various settlements. In contrast, although in the con-trol condition the exact same information was provided, the displays were static, alltwo-dimensional, and different types of information were displayed on different sources

Figure 1. Configuration Players, Facilitator, and Display in the Visualization SupportCondition

138 SWAAB ET AL.

Figure 2. Visualization and Negotiation Support: Three-dimensional visualization of theharbor area (top) and corresponding financial/environmental consequences (bottom) of acertain investment


(sheets of flip-over, maps, and a whiteboard), some of which were shared and someof which were handed out to individual negotiators.

Dependent Measures

We distinguish between four types of dependent measures: three types of negotiationoutcomes, and some additional outcome variables. The wording of questions of thescales are presented in the Appendix. All responses were recorded on five-point Likert-type scales ranging from 1 (very negative/strongly disagree) to 5 (very positive/stronglyagree), unless indicated otherwise. Scale means were computed by averaging acrossitems. The variables and their Cronbach’s alpha are displayed in Table 1, reliabilitieswere acceptable.

The measure of shared mental models was included in both pretest and posttest.This measure captured the amount of overlap in cognition about the issue at hand asreflected in participants’ assessment of the importance of a shared policy and sharedobjectives for themselves, the environment, and for the economic development of theregion [24]. So, these were all questions regarding an individual’s knowledge of theproblem at hand and its possible solutions. It should be pointed out that, to our knowl-edge, this operationalization of shared mental models is a novel method in negotia-tion support research.

With regard to the changes in structure of the pro-social climate over the course ofthe negotiation, measures of cohesion [27] were included in the pretest and posttest.Entitativity was only measured in the posttest [41] (see Table 1 for details).

The level of consensus was measured with a measure of differentiation within thegroup, proposed by Linville et al. [30]. This measure of differentiation can rangefrom 0 (total agreement) to 1 (maximum disagreement), and is less susceptible to theinfluence of “lone deviants’’ than other measures of variance.1 The degree of differ-entiation between parties was computed for the responses to items of the shared men-tal model scale, and then averaged. Thus, it was a measure of consensus about theproblem at hand.

Table 1. Scale Reliabilities

Measured in

Variable name Pretest Alpha Posttest Alpha

Shared mental model Yes 0.69 Yes 0.80Cohesion Yes 0.75 Yes 0.76Entitativity No — Yes 0.61Consensus Yes — Yes —Satisfaction process No — Yes 0.89Satisfaction outcome No — Yes 0.80Satisfaction facilitation No — Yes 0.74

Note: dashes indicate that reliabilities could not be computed.

140 SWAAB ET AL.

Further, subjective assessment of typical group research variables were recordedwith scales of satisfaction with the outcome, satisfaction of the process, and satisfac-tion with the facilitation [4].

Analytic Strategy

Because of the interdependence among negotiators, the group was the level of analy-sis. Although we had seven groups within each condition, analyses of variance(ANOVA) were conducted to test for between-condition differences. Some depen-dent variables were collected both prior to and after the negotiation. These variableswere analyzed in a repeated-measures ANOVA with experimental condition as a be-tween-participants variable, and time as a within-participants variable. This analysisyields three types of effects (Table 2). One is the between-groups main effect of sup-port condition. This indicates whether support across pre- and posttest has any effecton the dependent variables. The second is the pre-post difference main effect. Thisindicates whether there are any changes from pre- to posttest across support condi-tions. Finally, there is the pre-post times support condition interaction. This indicateswhether changes from pre- to posttest are different for the support versus no supportcondition. Since we predict that support should yield better outcomes than no sup-port, we predict that this interaction should be significant. In sum, hypotheses weretested by looking at the differences between pre- and posttest for each condition.

Results

Shared Mental Model

REGARDING SHARED MENTAL MODELS, there was no difference between the supportcondition (M = 3.97; SD = 0.22) and the control condition (M = 3.92; SD = 0.33) inthe pretest, F (1,12), n.s. (not significant). In the posttest however, the supportedcondition developed more shared mental models (M = 4.43; SD = 0.20) than thecontrol condition (M = 4.04; SD = 0.49). Changes over time differed reliably betweenconditions, F (1,12) = 7.20, p < 0.05. This provides support for our hypothesis thatvisualization support develops shared mental models (H1).

Pro-Social Climate

At the pretest, the level of cohesion in the support condition (M = 3.02; SD = 0.29)was similar to the control condition (M = 3.00; SD = 0.57). However, at the posttest,supported groups were more cohesive (M = 3.42; SD = 0.50) than control groups(M = 2.71; SD = 0.42). The significant interaction effect between the two conditionsover time, F (1,12) = 7.52, p < 0.05, supports our hypothesis that visualization sup-port increases cohesion (H2a).


Tabl

e 2.

Mea

ns p

er C

ondi

tion

(Sta

ndar

d D

evia

tions

Bet

wee

n B

rack

ets)

and

Ana

lysi

s of

Var

ianc

e

Pret

est

Post

test

Supp

ort

Pre-

post

Inte

ract

ion

Supp

ort

No

supp

ort

Supp

ort

No

supp

ort

F (1

,12)

F (1

,12)

F (1

,12)

Sha

red

men

tal m

odel

3.97

a3.

92a

4.43

bc4.

04ac

1.82

21.5

1***

7.20

*(0

.22)

(0.3

3)(0

.20)

(0.4

9)C

ohes

ion

3.02

a3.

00a

3.42

b2.

71a

2.99

0.21

7.52

*(0

.29)

(0.5

7)(0

.50)

(0.4

2)E

ntita

tivity

——

3.70

a3.

03b

10.8

6**

——

——

(0.3

5)(0

.39)

Con

sens

us0.

72a

0.66

a0.

52b

0.66

a0.

977.

87*

7.66

*(0

.07)

(0.1

2)(0

.12)

(0.1

1)S

atis

fact

ion

with

pro

cess

——

4.04

a3.

36b

16.4

0**

——

——

(0.2

9)(0

.34)

Sat

isfa

ctio

n w

ith o

utco

me

——

3.62

a3.

30a

1.61

——

——

(0.3

8)(0

.56)

Sat

isfa

ctio

n w

ith fa

cilit

atio

n—

—4.

26a

3.82

b6.

70*

——

——

(0.4

0)(0

.21)

Not

es:

Mea

ns i

n a

row

with

a d

iffe

rent

let

ter

(a, b

, or

c) d

iffe

r re

liabl

y at

p <

0.0

5; d

ashe

s in

dica

te t

hat

the

effe

ct c

anno

t be

com

pute

d be

caus

e m

easu

res

wer

e on

ly c

ompl

eted

at t

he p

ostte

st; *

p <

0.0

5; *

* p

< 0

.01;

***

p <

0.0

01.

142 SWAAB ET AL.

Predictions were also confirmed for entitativity. Although the control conditionscored neutrally (M = 3.03; SD = 0.39), groups in the support condition (M = 3.70;SD = 0.35) scored significantly higher on entitativity after negotiations, F (1,12) =10.86, p < 0.01. Groups with visualization support are thus more entitative thannonsupported groups, which supports our hypothesis H2b.

Negotiation Outcome

Results for consensus indicated that at the pretest, there was no reliable differencebetween the support condition (M = 0.72; SD = 0.07) and the control condition(M = 0.66; SD = 0.12), F (1,12) = 1.12, n.s. Over time, however, the level of consen-sus increased in the support condition (M = 0.52; SD = 0.12), but it did not increase inthe control condition (M = 0.66; SD = 0.11), F (1,12) = 7.66, p < 0.05. Thus, H3 wasconfirmed: visualization support increases the level of consensus among negotiators.

Outcome-Related Variables

Results for satisfaction with the outcome show that groups in the support condition(M = 3.62; SD = 0.38) were also somewhat more satisfied on this dimension of thenegotiation than in the control condition (M = 3.30; SD = 0.56). This difference wasnot significant, however, F (1,12) = 1.61, n.s. This leads us to reject our hypothesisthat VBS exposure leads to more satisfaction with the negotiation outcome (H4a).

After negotiation, supported groups were more satisfied with the process (M = 4.04;SD = 0.29) than control groups (M = 3.36; SD = 0.34), F (1,12) = 16.40, p < 0.01.Visualization support thus leads to more satisfaction with the negotiation process, insupport of our hypothesis (H4b).

For satisfaction with the facilitation, a reliable difference between the support con-dition (M = 4.26; SD = 0.40) and the control condition (M = 3.82; SD = 0.21) wasfound, F (1,12) = 6.70, p < 0.05. This leads us to accept our hypothesis that groupswith visualization support are more satisfied with the facilitation (H4c).

Discussion

RESULTS SHOWED A NUMBER OF RELIABLE DIFFERENCES in negotiation outcomes whencomparing negotiations with visualization support with unsupported negotiations.The pretest did not show reliable differences between conditions, which implies thatchanges were caused by experimental treatments. In particular, visualization supportimproves the outcomes of negotiations on a number of central negotiation variables,in comparison with a control condition. Most important, sharedness of the mentalmodel increases significantly in the support condition, as opposed to the control con-dition. Thus, the prediction was confirmed that visualization support may aid thefocusing of groups in such a way that shared mental models emerge. In view of thisfinding, it is not surprising that groups with visualization support reached consensusmore easily than control groups did.


Visualization support also improves the pro-social climate. Supported groups weresignificantly more cohesive when compared with unsupported groups. The resultsalso confirmed our predictions for negotiating groups’ entitativity: once more, sup-ported groups were more entitative, indicating that the perceived unity was greater.Thus, the results on socio-emotional variables tend to support the prediction thatvisualization support fosters a positive climate among negotiating parties.

No significant effects were found for satisfaction with the outcome. Because sup-ported groups reached significantly more consensus, this result could be seen as some-what surprising. A likely explanation is that only two out of seven supported groupsreached a final agreement, against zero in the control groups. This means that most ofthe groups did not have a concrete “outcome’’: they had converged, but still did notagree completely yet. However, the results do confirm that visualization support leadsto significantly more satisfaction with the negotiation process, in support of our hy-pothesis. Finally, although groups in both conditions seemed satisfied with the facili-tation of the meeting, groups with visualization support were more satisfied, in supportof our hypothesis.

In sum, visualization appears to improve three aspects of negotiations: (1) visual-ization support aids negotiators’ convergence of perceptions of reality, (2) it has posi-tive socio-emotional consequences in terms of increasing cohesiveness and entitativity,and (3) it stimulates consensus formation (and hence diminishes conflict). In the study,then, visualization appeared to have contributed positively to the negotiation process,and functioned as a strong cue for cognitions about the problem [2, 46]. We thereforeconclude that negotiations about a spatial planning issue are successfully supportedby means of visual representation aids.

Establishing that visualization has three different effects raises the question in whatway these effects are related to each other. As mentioned briefly in the introduction,we were working on the a priori assumption that shared mental models are the media-tor, and that positive climate and consensus are the outcomes. However, it is alsopossible that a positive pro-social climate is the mediator, or even that consensus is thefundamental process driving the other effects. It is hard if not impossible to make anydefinite claims about causation on the basis of the present data: ideally, one wouldmanipulate each of the three variables involved experimentally in order to establishtheir impact on the others. However, we did conduct a post hoc analysis of our theo-retical assumption by means of a structural equation model.2 This explorative analysisprovided tentative support for our assumption that shared mental models mediate theother effects. Nonetheless, in order to overcome the problems of inferring causationfrom correlation—especially with samples of this size—future research should be de-voted to assessing the central role of shared mental models and the processes involved.

Theoretical Implications

There are several theoretical implications of this study. First, it sheds light on theoperationalization of shared mental models. To our best knowledge, the method tomeasure the shared mental models we developed is a novel one, although it does not

144 SWAAB ET AL.

cover all aspects that such models comprise. Future research could extend these mea-sures to also cover shared knowledge of procedures and credibility [34, 36]. In anycase, the information processing perspective proved to be helpful in determining whenand how information should be provided in order to increase the sharedness of men-tal models. We inferred that providing unambiguous information (in this case by vi-sualization) in the encoding stage is essential in attaining shared mental models. Indeed,it appeared that discrepancies in knowledge and perceptual distortions among thenegotiators were reduced by the use of visualization support [28, 44]. This is alsoillustrated by participants’ comments such as “visualization made the whole issueless complex and more transparent” and “I was able to see causalities on an abstractlevel, which I would not have seen without the visualization tool.” An interestingquestion for future research could be to examine what types of information supportthe development of shared mental models in negotiation contexts other than spatialplanning conflict.

Furthermore, this study has implications for Negotiation Support Systems (NSS)theory. Much of the research that has been conducted to examine the utility of NSSshas been based on Lim and Benbasat’s [29] theory of NSS. This theory recognizes theseparate but important roles played by two components of NSS—decision supportsystems and electronic communication support. As the authors propose, the decisionsupport systems component of NSSs should provide negotiators with an increasedcapacity for information processing. In the current study, one might argue that weprovided the decision support component by presenting negotiators with unambigu-ous information, which stimulated similar encoding processes among negotiators andled to the development of shared mental models.

In this regard, our findings support the theoretical proposition by Lim and Benbasat[29]. However, our results also show that the provision of visual information had aneffect on the pro-social climate. That is, groups with negotiation support developed amore pro-social climate than those who were unsupported. Although ultimately anempirical question requiring further attention in future research, the present findingsdo suggest that the social climate may be an important factor involved in negotiations[9]. This suggests that future theorizing about NSS development should also considerwhen and how NSSs can foster a pro-social climate.

Moreover, designing NSSs with an electronic communication component mighthave effects that are detrimental to a pro-social climate: Lim and Benbasat [29], forexample, argue this elicits task-oriented behavior [see also 33]. More generally, muchresearch on NSS has focused on how, when, or why to deliver the right informationwith certain NSSs, but these studies tend to de-emphasize the relational demands thatnegotiations place on the communication environment [21, 22, 42, 50]. We argue thatNSSs should cater for a pro-social climate, since this factor has been shown to be agood predictor of negotiation settlement with and without the use of computer-medi-ated communication techniques [9, 19]. Thus, we argue that one prerequisite of NSSsis that the communication support they provide allows for “rich” interaction [5]. Inthe current study, we did so by providing both conditions the possibility to negotiateface-to-face.


In sum, although we found support for existing theories of negotiation support andthe design of NSS, our results also suggest that prior research tends to underestimatethe importance of the relational aspects within negotiation.

Limitations

What is remarkable about the present study, is that the control condition providedessentially similar information to negotiators, but via different channels. Thus, par-ticipants in the control condition also received maps, charts, and all the relevant back-ground information. Three aspects were different, however. First, the relevantinformation in the VBS was displayed simultaneously and in one location (on a largesemicircular screen), which forced negotiators to focus on the same objects through-out their negotiations. Second, the VBS offered the possibility to make interactivechanges to the display, which heightened participants’ awareness of the impact offuture scenario’s while focusing them on similar outcomes. Third, the VBS environ-ment presented an impressive display of technology, and forced participants to focuson the displays, which were colorful, dynamic, and entertaining. We believe that thesethree aspects mutually caused the differences between conditions, and that they canaccount for the emergence of shared mental models and consensus. Future researchcould attempt to isolate the respective influence of each of these three aspects of theNSS used in this study.

Although the results support the conclusion that visualization in negotiations canbe productive, the use of the experimental sample (student groups) may limit thestudy’s ecological validity [10]. To some extent, this limitation is reduced by the factthat NSS used in this study is also successful in “real-world” negotiations, whennegotiators are actual organizations and companies. For example, the VBS recentlyplayed a significant role during a negotiation in which stakes were quite high forthose involved, and where managers and policy-makers from different parties (gov-ernment, oil industries, and interest groups) discussed water management in the Dutchmainport of Rotterdam. Although a final agreement was not reached during the initialnegotiations when the VBS was used, negotiators had high expectations of futuresuccess, and post-negotiation interviews showed that they gained considerable in-sight into each other’s interests, problems, and in the possible solutions they couldreach. The development of a shared policy in this area is momentarily advancingrapidly. Thus, case studies tend to confirm that visualization may be beneficial forreal-world negotiation support.

It should be pointed out that the generalizability of this study is potentially limitedto spatial planning issues only. Visualization is especially appropriate to this type ofnegotiation issue. Indeed, we believe it is highly likely that the appropriateness ofvisualization as a support instrument varies across different negotiation issues. How-ever, this caveat does not affect the more general conclusion that we tentatively drawfrom our findings, namely that the provision of unambiguous information aids nego-tiation settlement through the development of a shared mental model and a positivenegotiation climate.

146 SWAAB ET AL.

Conclusion

On theoretical grounds, we proposed that during multiparty negotiations, shared mentalmodels are likely to be a crucial variable that NSSs should seek to maximize. Theresults showed that in a spatial planning dispute, visualization is an effective meansof improving the shared mental model of a group of negotiators, fostering a pro-social climate among negotiating parties, and thereby stimulating high levels of con-sensus. Thus, visualization supported the attainment of shared mental models, whichcould be the nucleus of shared solutions despite existing conflicts of interest. Resultsalso suggest that negotiation support could seek to improve the pro-social climate.

Acknowledgments: The authors thank Barbara van Knippenberg, Rick Watson, Ron Rice, MarkAakhus, Vladimir Zwass, and two anonymous reviewers for their helpful comments on earlierversions of this paper.

NOTES

1. Nevertheless, within-negotiation variance is strongly related to differentiation: in the presentstudy, their correlation was r (14) = 0.77, p < 0.01.

2. A model tested the theoretical assumption that visualization impacts on shared mentalmodels. Shared mental models were subsequently hypothesized to influence both pro-socialclimate and consensus. The model finally specified that pro-social climate would have a posi-tive effect on consensus. The model was tested with the structural equations program EQS [3],allowing tests of the significance of hypothesized paths from the covariance matrix, as well asthe degree of fit of the whole model. The fit of the model as tested was good, χ2(7) = 7.91,p = 0.34, n.s., CFI = 0.99, NFI = 0.90. Moreover, the hypothesized paths were all highlyreliable.

REFERENCES

1. Alabastro, M.; Beckman, G.; Gifford, G.; Massey, A.P.; and Wallace, W.A. The use ofvisual modeling in designing a manufacturing process for advanced composite structures. IEEETransactions on Engineering Management, 42, 3 (1995), 233–242.

2. Angehrn, A.A., and Luthi, H.-J. Intelligent decision support systems: A visual interactiveapproach. Interfaces, 20, 6 (1990), 17–28.

3. Boland, J., R.J.; Ramkrishnan, V.T.; and Te’eni, D. Designing information technology tosupport distributed cognition. In J.R. Meindl, C. Stubbart, and J.F. Porac (eds.), CognitionWithin and Between Organizations. London: Sage Publications, 1996, pp. 245–280.

4. Briggs, R.O., and Vreede, G.J. de. Measuring satisfaction in GSS meetings. In K. Kumarand J.I. DeGross (eds.), Proceedings of the Eighteenth International Conference on Informa-tion Systems. New York: ACM, 1997, pp. 483–484.

5. Burgoon, J.K.; Bonito, J.A.; Bengtsson, B.; Ramirez, A.; Dunbar, N.E.; and Miczo, N.Testing the interactivity model: Communication processes, partner assessments, and the qual-ity of collaborative work. Journal of Management Information Systems, 16, 3 (Winter 1999–2000), 33–56.

6. Campbell, D.T. Common fate, similarity, and other indices of the status aggregates ofpersons as social entities. Behavioral Science, 3, 1 (1958), 14–25.

7. Cannon-Bowers, J.A.; Salas, E.; and Converse, S. Shared mental models in expert teamdecision making. In N.J. Castellan Jr. (ed.), Individual and Group Decision Making. Mahwah,NJ: Lawrence Erlbaum, 1993, pp. 221–246.


8. Chidambaram, L. Relational development in computer supported groups. MIS Quar-terly, 20, 2 (1996), 143–163.

9. De Dreu, C.K.W.; Weingart, L.R.; and Kwon, S. Influence of social motives on integra-tive negotiation: A meta-analytic review and test of two theories. Interpersonal Relations andGroup Processes, 78, 5 (2000), 889–905.

10. Dennis, A.R.; Nunamaker, J.F.; and Vogel, D.R. A comparison of laboratory and fieldresearch in the study of electronic meeting systems. Journal of Management Information Sys-tems, 7, 3 (Winter 1991–1992), 107–135.

11. Eggins, R.A.; Haslam, S.A.; and Reynolds, K.J. Social identity and negotiation: Sub-group representation and superordinate consensus. Personality and Social Psychology Bulletin(in press).

12. Forsyth, F. Group Dynamics. London: Brooks/Cole Wadsworth, 1999.13. Gaertner, L., and Schopler, J. Perceived ingroup entitativity and intergroup bias: An

interconnection of the self and others. European Journal of Social Psychology, 28, 6 (1998),963–980.

14. Gaertner, S.L.; Mann, J.; Murell, A.; and Dovidio, J.F. Reducing intergroup bias: Thebenefits of recategorization. Journal of Personality and Social Psychology, 57, 2, (1989),239–249.

15. Gaertner, S.L.; Dovidio, J.F.; Anastasio, P.A.; Bachman, B.A.; and Rust, M.C. The com-mon ingroup identity model: Recategorization and the reduction of intergroup bias. In W.Stroebe and M. Hewstone (eds.), European Review of Social Psychology, vol. 4. London:Wiley, 1993, pp. 1–26.

16. Haslam, S.A. Psychology in Organizations: The Social Identity Approach. London: SagePublications, 2000.

17. Hinsz, V.B.; Tindale, R.S.; and Vollrath, D.A. The emerging conceptualization of groupsas information processors. Psychological Bulletin, 121, 1 (1997), 43–64.

18. Johnson-Laird, P.N. Mental Models: Towards a Cognitive Science of Language, Infer-ence, and Consciousness. Cambridge, MA: Harvard University Press, 1983.

19. Kahai, S.S., and Cooper, R.B. The effect of computer-mediated communication on agree-ment and acceptance. Journal of Management Information Systems, 16, 1 (Summer 1999),165–188.

20. Kameda, T.; Tindale, R.S.; and Davis, J.H. Cognitions, preferences, and social sharedness:Past, present, and future directions in group decision-making. In S.L. Schneider and J. Shanteau(eds.), Emerging Perspectives on Judgment and Decision Research. Cambridge: CambridgeUniversity Press (in press).

21. Kersten, G.E. Support for group decisions and negotiations: An overview. In J. Climaco(ed.), Multicriteria Analysis, Heidelberg: Springer-Verlag, 1997, pp. 332–346.

22. Kersten, G.E., and Michalowski, W. Negotiation modelling and support: Expanding theDSS paradigm. In C. Holsapple and A.B. Whinston (eds.), Recent Developments in DecisionSupport Systems. Berlin: Springer-Verlag, 1993, pp. 101–131.

23. Klimoski, R., and Mohammed, S. Team mental model: Construct or metaphor. Journalof Management, 20, 2 (1994), 403–437.

24. Kraiger, K., and Wenzel, L.H. Conceptual development and empirical evaluation of mea-sures of shared mental models as indicators of team effectiveness. In M.T. Brannick, E. Salas,and C. Prince (eds.), Team Performance Assessment and Measurement: Theory, Methods, andApplications. Mahwah, NJ: Lawrence Erlbaum, 1997, pp. 63–84.

25. Kramer, R.M., and Brewer, M.B. Effects of group identity on resource use in a simulatedcommons dilemma. Journal of Personality and Social Psychology, 46, 5 (1984), 1044–1057.

26. Larkin, J., and Simon, H. Why a diagram is (sometimes) worth 10,000 words. CognitiveScience, 11, 1 (1987), 65–100.

27. Lea, M.; Spears, R.; and de Groot, D. Knowing me, knowing you: Effects of visualanonymity on stereotyping and attraction in computer-mediated groups. Personality and So-cial Psychology Bulletin, 27, 5 (2001), 526–537.

28. Lewicki, R.J.; Litterer, J.A.; Minton, J.W.; and Saunders, D.M. Negotiation, 2d ed. NewYork: McGraw-Hill, 1994.

29. Lim, L.-H., and Benbasat, I. A theoretical perspective of negotiation support systems.Journal of Management Information Systems, 9, 3 (Winter 1993–1994), 27–44.

148 SWAAB ET AL.

30. Linville, P.W.; Salovey, P.; and Fischer, G.W. Perceived distributions of the characteris-tics of in-group and out-group members: Empirical evidence and a computer simulation. Jour-nal of Personality and Social Psychology, 57, 2 (1989), 165–188.

31. Massey, A.P., and Wallace, W.A. Understanding and facilitating group problem structur-ing and formulation: Mental representations, interaction, and representation aids. DecisionSupport Systems, 17, 4 (1996), 253–274.

32. McGarty, C.; Haslam, S.A.; Hutchinson, K.J.; and Grace, D.M. Determinants of per-ceived consistency: The relationship between group entitativity and the meaningfulness ofcategories. British Journal of Social Psychology, 34, 3 (1995), 237–256.

33. Miranda, M.M., and Bostrom, R.P. The impact of group support systems on group con-flict and conflict management. Journal of Management Information Systems, 10, 3 (Winter1994–1995), 63–95.

34. Moreland, R.L.; Argote, L.; and Krishnan, R. Socially shared cognition at work:Transactive memory and group performance. In J.L. Nye and A.M. Brower (eds.), What’sSocial About Social Cognition? Research on Socially Shared Cognition in Small Groups. Thou-sand Oaks, CA: Sage Publications, 1996, pp. 57–84.

35. Orasanu, J.M. Decision-making in the cockpit. In E.L. Wiener and B.G. Kanki (eds.),Cockpit Resource Management. San Diego, CA: Academic Press, 1993, pp. 137–172.

36. Peterson, E.; Mitchell, T.; Thompson, L.; and Burr, R. Collective efficacy and aspects ofshared mental models as predictors of performance over time in work groups. Group Processesand Intergroup Relations, 3, 3 (2000), 296–316.

37. Rahim, M.A. A measure of styles of handling interpersonal conflict. Academy of Man-agement Journal, 26, 2 (1983), 368–376.

38. Resnick, L.B.; Levine, J.M.; and Teasley, S.D. (eds.). Perspectives on Socially SharedCognition. Washington, DC: American Psychological Association, 1991.

39. Rosenhead, J. Rational Analysis for a Problematic World. New York: Wiley, 1989.40. Rouse, W.B., and Morris, N.M. On looking into the black box: Prospects and limits in

the search for mental models. Psychological Bulletin, 100, 3 (1986), 349–363.41. Sassenberg, K. The two faces of anonymity: Effects on group decision making. In T.

Postmes, R. Spears, M. Lea, and S. Reicher (eds.), SIDE Issues Centre Stage: Recent Develop-ments in Studies of De-Individuation in Groups. Amsterdam: KNAW, 2000, pp. 93–106.

42. Thiessen, E.M.; Loucks, D.P.; and Stedinger, J.R. Computer-assisted negotiations ofwater resources conflicts. Group Decision and Negotiation, 7, 2 (1998), 109–129.

43. Thompson, L., and Fine, G.A. Socially shared cognition, affect, and behavior: A reviewand integration. Personality and Social Psychology Review, 3, 4 (1999), 278–302.

44. Thompson, L.; Peterson, E.; and Kray, L. Social context in negotiation: An informationprocessing perspective. In R.M. Kramer and D.M. Messick (eds.), Negotiation as a SocialProcess. Thousand Oaks, CA: Sage Publications, 1995, pp. 5–36.

45. Tindale, R.S., and Kameda, T. “Social sharedness” as a unifying theme for informationprocessing in groups. Group Processes & Intergroup Relations, 3, 2 (2000), 123–140.

46. Van Voris, P.; Millard, W.D.; Thomas, J.; and Urban, D. TERRA-Vision—The integra-tion of scientific analysis into the decision making process. International Journal of Geo-graphic Information Systems, 7, 2 (1993), 143–164.

47. Weick, K., and Bougon, M. Organizations as cognitive maps. In H. Sims and D. Gioia(ed.), The Thinking Organization. San Francisco: Jossey-Bass, 1986, pp. 103–135.

48. Weick, K.E., and Roberts, K.H. Collective mind in organizations: Heedful interrelatingon flight decks. Administrative Science Quarterly, 38, 3 (1993), 357–381.

49. Zachary, W. A cognitively based functional taxonomy of decision support techniques.Human Computer Interaction, 2, 1 (1986), 25–63.

50. Zigurs, I.; Reitsma, R.; Lewis, C.; Hübscher, R.; and Hayes, C. Accessibility of com-puter-based simulation models in inherently conflict-laden negotiations. Group Decision andNegotiation, 8, 6 (1999), 511–533.


Appendix

Constructs and Measurements

Scale Questions

Shared mental model A shared policy within the harbor area is positive foryour organization

A shared policy within the harbor area is positive forthe environment

The placement of pipelines is positive for yourorganization

The placement of pipelines is positive for theenvironment

In the long term, investments in pipelines have positiveeconomical consequences

In the long term, investments in pipelines have positiveenvironmental consequences

Cohesion The cooperation in the group was efficientThere is a sense of unity in this groupThe members of this group form a unityDespite the differences between the members, the

group develops its own identity Entitativity Select which of the figures below fits your idea of the

group best

❏ ❏ ❏ ❏ ❏

❏ ❏ ❏ ❏ ❏

❏ ❏ ❏ ❏ ❏

❏ ❏ ❏ ❏ ❏

150 SWAAB ET AL.

Satisfaction with the I am very satisfied with the method that has been usedprocess during the meeting

My overall judgment of the method that has been usedduring the meeting is positive/negative

The influence of the method that has been used duringthe meeting on the decision-making process waspositive/negative

The influence of the method that has been used duringthe meeting on the quality of the deliberation in thegroup was positive/negative

The influence of the method that has been used duringthe meeting on the atmosphere in the group waspositive/negative

I rate the quality of the method that has been used duringthe meeting with the number 1–10

Satisfaction with the I accept the outcome of this meetingoutcome The decision that has been made during this meeting

is the right oneI am satisfied with the results of this meetingI rate the result of this negotiation with the number 1–10

Satisfaction with the The facilitator’s knowledge was sufficient to guide thefacilitation content of the process

The process of the meeting has been guided effectivelyThe facilitator’s influence on the process of the meeting

was positive/negative

Documents

Multiparty Negotiation Support