See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/232601618

Looking and competing: Accountability and visual access in

integrative bargaining

Article  in  Journal of Personality and Social Psychology · January 1981

DOI: 10.1037/0022-3514.40.1.111

CITATIONS READS
131 535

3 authors, including:

Peter Carnevale Dean Pruitt

University of Southern California George Mason University
97 PUBLICATIONS   5,120 CITATIONS    121 PUBLICATIONS   6,268 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The Selection of Mediation Tactics in Public Sector Disputes: A Contingency Analysis View project

Readiness Theory Project View project

All content following this page was uploaded by Peter Carnevale on 13 May 2015.

The user has requested enhancement of the downloaded file.

Journal of Personality and Social Psychology Copyright 1981 by the American Psychological Association, Inc.
1981, Vol. 40, No. 1, 111-120 0022-3514/81/4001-0111500.75

Looking and Competing: Accountability and Visual Access

in Integrative Bargaining
Peter J. D. Carnevale, Dean G. Pruitt, and Steven D. Seilheimer
State University of New York at Buffalo

This research was designed to reconcile conflicting findings about the impact of
negotiator accountability on bargaining process and outcome. Some studies have
found that high accountability produces more pressure (dominance) tactics and
lower joint benefit than low accountability. Other studies have found no ac-
countability effects. Results of the present research suggest that negotiators'
visual accessibility is a moderator of the effects of accountability. Accountability
only produced lower joint benefit when the negotiators were face to face. Co-
operative behavior and reports of a cooperative atmosphere were also diminished
when negotiators were facing each other rather than talking across a barrier.
These results were interpreted on the assumption that negotiation involves a
competitive definition of the situation, especially under high accountability. In
such a context, staring is likely to be employed to dominate and is likely to be
interpreted as domineering behavior. When there is no visual access, staring is
not possible; hence, there is less temptation to try to dominate and less reason
to view the other party as making a similar effort. The result is more cooperation
and greater joint benefit.

This research dealt with the effects of ne- ible-rigidity" hypothesis about integrative
gotiator accountability on joint benefit in bargaining. This hypothesis postulates that
bilateral negotiation. Negotiators are "ac- agreements involving high joint benefit will
countable" to the extent that their constit- result when both negotiators have high, rel-
uents can reward or punish them on the basis atively rigid aspirations in conjunction with
of their performance. "Joint benefit" means a flexible, problem-solving orientation. A
the collective gain achieved by the two ne- problem-solving orientation involves the goal
gotiators in the final agreement. The study of finding an agreement that is mutually
was part of a larger series of investigations acceptable to both parties. It can be con-
of the antecedents of high joint benefit. Pro- trasted with a win/lose orientation, which
cesses that lead to high joint benefit have involves the goal of profiting at the other
sometimes been called "integrative bargain- party's expense. The rationale for this hy-
ing" (Walton & McKersie, 1965) because pothesis is as follows: High aspirations pro-
the negotiators' separate values must be in- vide the motivation necessary to seek out the
tegrated (i.e., reconciled) to achieve this out- nonobvious options that must so often be
come. found if high joint benefit is to be achieved.
An earlier study in this series (Pruitt The tactics adopted under a problem-solving
& Lewis, 1975) provides support for orientation make it possible to locate such
what Pruitt and Lewis (1977) call a "flex- options.
Pruitt and his associates (see Pruitt &
The research reported here was supported by Na- Carnevale, 1980; Pruitt & Lewis, 1977) have
tional Science Foundation Grant BNS76-10963. The described several classes of negotiation tac-
authors wish to acknowledge the helpful comments of
Alice M. Isen and Brenda Major. Portions of this article tics. Two of these are encouraged by a prob-
were presented at the 50th annual meeting of the East- lem-solving orientation:
ern Psychological Association, April 1979. 1. Exchange of information about values
Requests for reprints should be sent to Peter J. D.
Carnevale, Department of Psychology, State University
and priorities. Receiving this kind of infor-
of New York at Buffalo, 4230 Ridge Lea Road, Buffalo, mation allows a negotiator to think simul-
New York 14226. taneously about both parties' welfare, which

in
112 P. CARNEVALE, D. PRUITT, AND S. SEILHEIMER

is necessary for achieving an insightful so- In such constant-sum situations, a gain for
lution to the common problem. one party is a loss for the other.
2. Heuristic trial and error, in which each Accountability has been studied exten-
party makes an orderly series of proposals sively in experiments involving unidimen-
reflecting his/her own values and priorities sional negotiation tasks. The majority (eight)
in an effort to find one that also appeals to of these studies show that accountability
the other party. Four intercorrelated tactics makes bargainers reluctant to concede (Bar-
have been identified in this latter class (Kim- tunek, Benton, & Keys, 1975; Benton, 1972;
mel, Pruitt, Magenau, Konar-Goldband, & Benton & Druckman, 1973, 1974; Druck-
Carnevale, 1980; Pruitt et al., 1978): (a) fre- man, Solomon, & Zechmeister, 1972; Kli-
quently making new and different proposals, moski, 1972; Klimoski & Ash, 1974; Kogan,
(b) making concessions on low-priority is- Lamm, & Trommsdorff, 1972).2 A minority
sues while holding firm on high-priority is- (three) of these studies found no account-
sues, (c) proposing all the options one can ability effects (Frey & Adams, 1972; Cru-
devise at a given level of benefit to oneself der, 1971; Cruder & Rosen, 1971).3 The
before conceding to a lower level of benefit,1 majority finding is ambiguous with respect
and (d) requesting the other's reaction to to integrative bargaining. If the reluctance
each proposal. The first of these tactics is to concede results from the maintenance of
simple trial and error; the second and third high aspirations, accountability should en-
are heuristics for generating new proposals courage agreements that involve high joint
that suit one's own needs; the fourth is a
manifestation of the problem-solving orien- 1
This third, "systematic-concession" tactic was orig-
tation that gives rise to this class of tactics. inally described by Kelley and Schenitzki (1972), who
A third class of tactics, pressure tactics, showed that it produced high joint benefit in negotiation
is a manifestation of the win/lose orienta- sessions involving note passing.
2
tion. The following four intercorrelated tac- In four of these studies (Bartunek et al., 1975; Ben-
tics are examples of this class: use of threats, ton & Druckman, 1973, 1974; Druckman et al., 1972),
accountability was confounded with the presence or ab-
positional commitments, arguments aimed sence of a constituent. But the results were identical to
at persuading the other to concede, and ef- the three in which pure accountability was manipulated
forts to raise one's status in the other party's (Benton, 1972; Klimoski, 1972; Klimoski & Ash, 1974),
eyes. All are aimed at dominating the other suggesting that accountability was the important vari-
able in the first four studies. The experiment by Kogan
party. et al. (1972) is not strictly a study of accountability in
Prior experiments (Carnevale, Pruitt, & that the constituents' capacity to reward or punish the
Britton, 1979; Kimmel et al., 1980; Lewis negotiator was not manipulated. But a close cousin of
& Fry, 1977; Pruitt et al., 1978; Pruitt & accountability was manipulated—the constituent's power
Lewis, 1975; Schulz & Pruitt, 1978) have to make decisions about how to conduct the negotiation;
the results were similar to those achieved with account-
shown that under certain conditions infor- ability.
mation exchange and/or heuristic trial and It should also be noted in these studies that the ac-
error are antecedents of high joint benefit, countability effect was stronger under some conditions
whereas pressure tactics are an antecedent than others. It was weak or nonexistent when (a) the
bargainers believed that their constituents wanted them
of low joint benefit. to behave cooperatively toward the other negotiator
The negotiation task employed in this re- (Benton & Druckman, 1974), (b) the bargainers did not
search (which will be described later) has find membership in their groups attractive (Klimoski,
integrative potential in that there are a num- 1972), (c) a mediator suggested a solution or encouraged
ber of possible agreements with differing lev- role-reversal behavior (Bartunek, Benton, & Keys,
1975), and (d) the bargainers were school girls (Druck-
els of joint profit. However, not all negoti- man et al., 1972). However it should be noted that a
ation situations have this potential. In full accountability effect was obtained when the subjects
particular, integrative potential is small or were college women (Benton & Druckman, 1973, 1974).
3
nonexistent in "unidimensional" tasks, where One other study (Druckman, 1967) that failed to
achieve an accountability effect is not included in this
the options are arrayed along a single di- list because unlike the other 11 studies, there were no
mension that has opposing value for the two constituents in the laboratory with the subject. Instead,
parties, such as a single price or wage rate. a weak, role-playing manipulation was employed.
 LOOKING AND COMPETING 113

profit. But if it is due to a win/lose orien- orientation" means a goal of enhancing one's
tation, accountability should discourage the own benefit without reference to the other's
development of such agreements. gains.) All negotiators played the role of
To resolve this ambiguity, Pruitt et al. accountable representatives, though constit-
(1978) manipulated accountability in a sit- uents were not actually present. Significant
uation with integrative potential. Under high interactions were found for both pressure
accountability, there were more pressure tactics and joint benefit, with the largest
tactics and lower joint profit than under low amount of pressure tactics and the least joint
accountability, suggesting that accountabil- profit in the condition where the negotiators
ity encourages a win/lose orientation. Fur- had an individualistic orientation and talked
thermore, accountability was found to have face to face. An analogy can be made be-
no effect on several measures of level of as- tween the individualistic orientation in the
piration. In an unpublished follow-up study, Lewis and Fry study and high accountability
however, the present authors were unable to in our studies, in that both seem likely to
replicate these findings. There were only impart a competitive definition of the situ-
trends in the direction of higher account- ation. Hence, this finding suggests that a
ability producing more pressure tactics and combination of high accountability and vi-
lower joint benefit. sual access produces especially strong pres-
A comparison of the procedures used in sure tactics and especially low joint benefit.
these two studies and a review of the liter- Lewis and Fry (1977) provide a possible
ature suggested the hypothesis tested in the theoretical rationale for this hypothesis,
present experiment. In the Pruitt et al. based on the assumption that face-to-face
(1978) study, negotiations were conducted interaction permits bargainers to gaze at
face to face, whereas in the follow-up study each other. Under a competitive definition
negotiations were conducted over an inter- of the situation, the other party's gaze may
com. Furthermore, a review of the prior, be interpreted as an effort to dominate,
unidimensional-task literature revealed that which must be defended against by pressure
in the eight studies that achieved account- tactics. These pressure tactics in turn dis-
ability effects, negotiations were conducted courage the development of high joint ben-
face to face, whereas in the three that did efit. In support of this interpretation, these
not obtain such effects, negotiation was by authors found that in the face-to-face con-
means of note passing. Hence, we hypothe- dition, dyads whose members stared more
sized that accountability enhances pressure at each other also employed more pressure
tactics and diminishes joint benefit when tactics and achieved lower joint benefit. A
negotiators can see and hear each other, but related possible interpretation is that verbal
not when they can only hear each other. and nonverbal dominance behaviors are tied
This hypothesis is superficially inconsis- together psychologically, so that the former
tent with prior research by Milgram (1974) will be attenuated if the latter cannot be
and Gahagan (1970), who found that phys- enacted. If so, under a competitive definition
ical proximity makes it harder for people to of the situation, bargainers will make pres-
impose costs on one another, and by Morley sure statements only if they can accompany
and Stephenson (1977), who found that ne- them with efforts to stare the other down.
gotiators are more concerned with achieving If staring is ruled out by a barrier, joint
equal outcomes when face to face than when profit will be larger because it will not be so
talking over the telephone. However, the easy for the bargainers to try to dominate
hypothesized interaction very much resem- each other verbally.
bles one found by Lewis and Fry (1977) in Assuming that high accountability creates
a task with integrative potential. These au- a competitive definition of the situation, sim-
thors manipulated orientation (problem- ilar arguments can be employed in support
solving vs. individualistic) and visual access of the hypothesis tested in the present re-
(face to face vs. conversation across a bar- search.
rier) in a 2 X 2 design. ("Individualistic In summary, we hypothesized an inter-
114 P. CARNEVALE, D. PRUITT, AND S. SEILHEIMER

action between negotiator accountability Procedure

and visual access so that accountability will
enhance the use of pressure tactics and will In the high and low accountability conditions, two
confederates posing as subjects were in the room with
diminish joint benefit when negotiators can the subjects. Two teams were formed, each consisting
see and hear each other, but not when they of a subject and a confederate. The task was described
can only hear each other. To test this hy- as involving "negotiation between two simulated com-
pothesis, we employed a 3 X 2 factorial de- panies." Team members were seated beside each other
sign. The three levels of the first variable and introduced. One was then randomly designated as
the "company owner" and the other as the "company
were high, low, and no accountability, with negotiator." This was a bogus randomization procedure
the latter involving no constituent. On the in that confederates always played the role of company
basis of findings by Benton (1972), we ex- owner. Care was taken to ensure that neither teammates
pected the low- and no-accountability con- nor opposing negotiators were previously acquainted or
had interacted while waiting for the experiment to begin.
ditions to have similar effects. The two levels The teams were seated across a table from one an-
of the second variable consisted of visual other, separated by a large barrier that prevented them
access, where the negotiators talked face to from seeing one another. They were given printed copies
face, and no visual access, where they talked of the instructions, which were presented by tape re-
across a barrier. corder as they followed the text. The experimenter was
present and periodically stopped the tape recorder to
answer questions. Each team was given its own profit
schedule, and the schedules were explained in such a
Method way that neither team could tell what numbers appeared
in the other's schedule. The participants were told that
Subjects the negotiators would be able to talk freely to each other
about anything they wished, including the profits in their
The negotiators were 132 males recruited by adver- schedules, but could not show their schedules to one
tisement on the campus of the State University of New another.
York at Buffalo. They were promised $3 for their par- The negotiators were told that the profits made in the
ticipation plus any benefits they might accrue. Their negotiation would be "converted into real money and
ages ranged from 18 to 40 years. Twenty-two other par- divided between you and the owner." They were shown
ticipants (11 negotiator pairs) were discarded from the a printed form, the Settlement Report, that was to be
design because of prior knowledge of the task (5), in- used to report to the company owner how much money
ability to understand the task (3), or prior acquaintance had been made in the negotiation. A brief quiz was
with the other participants (3). The subjects discarded administered to be sure each participant understood his
were evenly distributed across the design. Two subjects profit schedule and how to compute the cash proceeds
were scheduled at a time and appeared with two un- (the negotiator and owner answered the quiz sepa-
dergraduate confederates who served as constituents. rately). The accountability instructions were then pre-
Eleven pairs of subjects were randomly assigned to each sented.
of the six cells in the design. The owners were next sent to a separate room, to
write a message that would be sent to their negotiator
to indicate how they wanted the negotiator to act and
that would also reiterate part of the instructions. It was
Negotiation Task said that the owners would be busy filling out question-
naires during the negotiation. Thus, they would not be
The task was very similar to that employed in our able to see the actual negotiation but would receive a
previous research on integrative bargaining (Kimmel et report about how much money their respective negoti-
al., 1980; Pruitt et al., 1978; Pruitt & Lewis, 1975). It ator had made. Their role was minimal after this point.
required pairs of participants, taking the roles of buyer The negotiators were then given a mimeographed check-
and seller in a wholesale market, to agree on prices for list reminding them of the most important parts of the
three appliances: television sets, vacuum cleaners, and procedure. This list was read to them as they followed
typewriters. The buyer was said to represent a large the text, and it was retained by them during the nego-
department store, and the seller a manufacturer of small tiation. Just after this, a handwritten note was delivered
appliances. to the negotiators by their respective company owners.
Each negotiator had a profit schedule that could not This note was designed to reinforce the accountability
be shown to the other. In the buyer's schedule, TVs had manipulation.
the highest and typewriters the lowest potential for The negotiation then began. A time limit of 30 min.
profit. These priorities were reversed for the seller. was imposed, during which the negotiators were to reach
Hence, the task had integrative (logrolling) potential in agreement. If agreement was not reached after 25 min.,
the sense that high joint profits could be achieved if the the experimenter administered a 5-min. warning with
negotiators exchanged concessions on their low-profit the statement, "It is important that you reach agree-
items. ment, since otherwise you will both make nothing from
 LOOKING AND COMPETING 115

the negotiation." In all conditions the experimenter sat divide it between us any way you want." No mention
in a room behind a one-way mirror during the negoti- was made of an evaluation in the low accountability
ation, and the participants knew he was there. After the condition.
negotiation, questionnaires were filled out and the par-
ticipants were interviewed, debriefed, and finally paid
the amount they had earned. Manipulation of Visual Access
In the no-accountability condition, the procedure was Under visual access, the barrier that had separated
identical except that there were no company owners and the teams during the instructions was removed just be-
the subject was told he would keep whatever money he fore the start of the negotiation, allowing the negotiators
earned. to see one another. Under no visual access, the barrier
remained in place for the negotiation. Thus, visual in-
Manipulation of Accountability teraction was eliminated, but verbal communication was
not hindered.
In the high accountability condition, the negotiators
were told that the money each made would be sent to
his company owner who would "decide how to divide
Results
it between himself and youself and would also "write Manipulation Checks
a brief evaluation of how well you have done in the
negotiation, so that you can see the basis for his decision In comparison to low accountability ne-
about dividing up the money." In the low accountability
condition, the negotiators were told that they would re-
gotiators, high accountability negotiators
ceive the money and could "decide how to divide it be- reported feeling more accountable to their
tween yourself and the company owner." In this con- company owner, F(l, 40) = 5.70, p < .05;
dition, nothing was said about an evaluation by the more concerned with satisfying their com-
company owner. Under no accountability, at no time pany owner, F(\, 40) = 7.40, p < .01; and
was any mention made of a company owner.
The manipulation was reinforced by material in- less powerful than their company owner,
cluded in the checklist and by the Settlement Report. F( \, 40) = 13.32, p < .001. These differences
Under high accountability, the checklist said, "Your suggest that the accountability manipulation
company owner will be responsible for dividing any real was successful.
money earned in the negotiation. (Your company owner
will also write a report on your performance.)" The
Settlement Report had a space for the negotiator to fill Outcome of the Negotiation
in his net profit, spaces for the company owner to in-
dicate how this profit should be divided, and a space for All pairs of negotiators reached agree-
the owner to write an "evaluation of negotiator perfor- ment, as might be expected, since this was
mance." On completion of the negotiation, this report the only way to make money in the negoti-
was to go from the negotiator to the owner and then
back to the negotiator. Under low accountability, the ation task. Table 1 presents means for the
checklist said, "You, the negotiator, will be responsible outcomes of negotiation.
for dividing any real money earned in the negotiation." Two measures of joint benefit were em-
The Settlement Report had spaces for the negotiator to ployed, both drawn from Sen (1970): total
fill in his net profit and to divide this profit between the profit made by the dyad as a whole and the
owner and himself. No mention was made of an eval-
uation, and this report was to go only from the negotiator profit made by the less successful negotiator.
to the owner. Also in the checklist, negotiators under The latter measure is sensitive to both the
high accountability were told they would meet with their overall dyadic outcome and the equality of
company owner and not with the other negotiator after the outcomes achieved by the two negotia-
the negotiation. Negotiators under low accountability
were told they would not meet with their company owner tors. Two other outcome measures are also
but would meet with the other negotiator after the ne- presented. As can be seen in rows 1 and 2
gotiation. of Table 1, high accountability led to di-
The high and low accountability conditions also dif- minished utility of agreement only when ne-
fered in the message each negotiator received from his gotiators had visual access to one another.
respective company owner. The message reiterated part
of the instructions and told each negotiator to "make Significant interactions were obtained for
as much as you can for our company." In the high ac- both joint profit, F(2, 60) = 3.82, p < .05,
countability condition, the message said, "I understand and the profit for the less successful nego-
that I'll be given any money you make and that I'll tiator, F(2, 60) = 3.81, p < .05. Thus, the
divide it between us any way I want. I'll also write an
evaluation of how well I think you did." In the low part of our hypothesis about joint profit was
accountability condition, it said, "I understand that you supported. No significant main effects were
will be given any money you make and that you will obtained for these two variables. No signif-
116 P. CARNEVALE, D. PRUITT, AND S. SEILHEIMER

Table 1
Outcome Measures as a Function of Visual Access and Accountability
Visual access No visual access

Outcome measure HiA LoA NoA HiA LoA NoA

1. Total dyad profit 8,845 9,936 9,964 9,909 9,800 9,827
2. Profit of less successful
negotiator 4,027 4,782 4,791 4,673 4,555 4,655
3. Profit of more successful
negotiator 4,818 5,154 5,173 5,236 5,245 5,172
4. Row 3 minus row 2 791 372 382 563 690 517

Note. HiA = high accountability; LoA = low accountability; NoA = no accountability.

icant main effects or interactions were ob- prediction that accountability would have
tained for the profit of the more successful a larger impact on pressure tactics in face-
negotiator or the differences between the to-face negotiations than in negotiations
more successful and less successful negoti- without visual interaction was borne out by
ators. a significant interaction for this index, F(2,
60) = 3.23, p < .05. As hypothesized, the
Negotiation Process most pressure tactics by far were exhibited
in the high accountability-visual access cell.
A number of process measures were de- Rows 2 and 3 of Table 2 provide means
rived from transcripts of the negotiation ses- from indices of information exchange and
sions. Some involved counting the number heuristic trial and error. Our index of infor-
of times various themes were mentioned in mation exchange involved the proportion of
the negotiation and dividing by the total statements in which truthful information
number of statements made. These measures was given about the numbers in the speaker's
had interrater reliability correlation coeffi- profit schedule. A significant main effect for
cients of .79 or above. Others involved re- visual access was found on this variable, with
cording or counting various features of the negotiators providing one another more in-
sequence of offers made by the negotiators. formation when not able to see one another
Still another measure involved counting the than when able to see one another, F(l,
number of first-person singular and plural 60) = 4.34, p < .05. No other significant ef-
pronouns employed in the negotiation. In our fects were found for this variable. Our index
experience, the latter two types of codes can of heuristic trial and error involved the mean
be extracted with practically no error; hence, of the standard scores of four measures: the
reliability coefficients were not calculated number of different offers made, the extent
for them. Table 2 presents five indices de- of concessions on low-priority appliances,
rived from these measures and two indices the number of offers made before a sub-
based on items from the postquestionnaire. stantial concession, and the proportion of
All tests of significance on the means in this statements in which the listener was asked
table were done with dyad scores. to react to the speaker's most recent offer.
Means from the index of pressure tactics No significant effects were obtained for this
are shown in row 1 of Table 2. This index variable.
consisted of the sum of equally weighted A measure of negotiator aspirations was
measures of the use of threats, positional obtained by averaging standard scores of
commitments, status slurs, and arguments three intercorrelated variables: value of the
that the other should concede for reasons first offer, value of the second offer, and
that relate to the commercial setting hypo- number of offers made after one negotiator
thetically surrounding the negotiation. Our had mentioned an offer that was equal to or
 LOOKING AND COMPETING 117

Table 2
Process Measures as a Function of Visual Access and Accountability
Visual access No visual access

Process measure HiA LoA NoA HiA LoA NoA

1. Pressure tactics .24 .06 .03 .08 .04 .05

2. Information exchange .03 .07 .08 .11 .17 .12
3. Heuristic trial and
error -.19 .11 .83 -.25 -.01 -.83
4. Aspiration level -.03 1.61 -.57 .03 -.34 -.43
5. Understands other's
priorities 7.00 10.55 11.91 12.55 12.73 10.64
6. We/I ratio .31 .40 .29 .62 .43 .48
7. Cooperative
atmosphere 2.96 4.09 4.46 4.36 4.82 4.77

Note. HiA = high accountability; LoA = low accountability; NoA = no accountability.

better for both of them than $4,000.4 The are given in rows 6 and 7 of Table 2. The
results are shown in row 4. No significant same result emerged from both measures:
effects were obtained for this measure. There was a more positive atmosphere when
In order for negotiators to gain insight into the negotiators were not able to see one an-
which solutions are most mutually benefi- other than when they were able to see one
cial, they must develop an understanding of another. The we/1 ratio was greater under
one another's values and priorities. A mea- no visual access than under visual access,
sure of such understanding was constructed F( 1, 60) = 4.70, p < .05, and the atmosphere
from postquestionnaire items regarding was rated as more cooperative under no vi-
which commodities offered the other nego- sual access than under visual access, F(l,
tiator the most and least profits. The means 60) = 7.19, p < .01. There was also a sig-
for this index are shown in row 5 of Table nificant main effect for accountability on the
2. These results were similar in form to those measure of perceived cooperativeness, F(2,
for our measures of joint benefit. The ne- 64) = 3.74, p < .05. A less cooperative at-
gotiators had less understanding of one an- mosphere was perceived under high account-
other's priorities in the high-accountability- ability than under low or no accountability.
visual access condition than in any of the Again there were no significant differ-
other conditions. Again the interaction was ences between the low and no-accountabil-
statistically significant, F(2, 60) = 3.62, ity conditions on any of these measures.
p < .05.
Two measures of interpersonal atmo- Correlational Findings
sphere in the negotiation were also em-
ployed. One was a "we/I ratio," consisting Within-cell correlations were computed
of the number of first-person-plural pro- among the variables shown in Tables 1 and
nouns (we, us, our, ours) referring to the two 2 in an effort to shed further light on how
negotiators divided by the number of first- high joint benefit is achieved. These corre-
person-singular pronouns (I, me, my, mine). lations had 59 degrees of freedom: one for
We view this ratio as a measure of the extent
to which the negotiators have a sense of com- 4
$4,000 represents the amount that can be achieved
mon identity or unit relationship with one in this negotiation task by adopting an obvious compro-
another. The other was a 7-point postques- mise that involves no exchange of concessions. This is
the lowest score ordinarily made by a negotiator. For
tionnaire rating scale concerning how co- both parties to improve on this score, it is necessary to
operative the atmosphere had been in the exchange concessions on low-priority issues and thus
negotiation. The means for these measures exploit the integrative potential in the situation.
118 P. CARNEVALE, D. PRUITT, AND S. SEILHEIMER

each of the 66 dyad scores, minus one for predictors of joint benefit were significantly
each of the six cell means, minus one for the correlated with total profit under high ac-
line of regression. The following are among countability. For pressure tactics, r(20) =
the highlights of these results. -.49, p < .05; for information exchange,
There were positive correlations among r(20) = .47, p < .05; for heuristic trial and
the three measures that were higher under error, r(20) = .37 p < .10. Pressure tactics
no visual access than under visual access, as were again inversely related to understand-
follows: information exchange—we/I ratio, ing the other's priorities, r(20) = -.45,
r = .35, p < .01; information exchange—co- p < .05, and understanding was again pre-
operative atmosphere, r = .47, p < .01; we/ dictive of total profit, r(20) = .72, p < .01,
I ratio—cooperative atmosphere, r — .47, in this subset of cells.
p < .01. This suggests that these measures
dealt with much the same thing, perhaps Discussion
what Deutsch (1973) calls a "cooperative
process." The results of this study show that the
The best predictor of total dyad profit was impact of accountability on the conduct and
the understanding negotiators had about one outcome of negotiation is moderated by vi-
another's priority structure (r = .54, p < sual access. As hypothesized, in face-to-face
.01). None of the three traditional predictors negotiation, high accountability produced
of joint benefit—pressure tactics, informa- more pressure tactics and lower joint out-
tion exchange, or heuristic trial and error— come than low or no accountability. But ac-
was significantly related to total profit (nor countability had no effect when negotiators
were these measures related to one another). talked across a barrier. A similar conclusion
However, the perception of a cooperative can be reached by comparing earlier studies
atmosphere was so related (r = .33, p < .01), of the effect of accountability in unidimen-
suggesting that some as yet unidentified sional negotiation settings. In studies involv-
problem-solving process was at work in gen- ing face-to-face interaction, accountability
erating high joint profit. produced diminished concession rate,
In line with the flexible-rigidity hypoth- whereas in studies involving note passing, no
esis, our measure of negotiator aspirations accountability effects were found. Assuming
was positively correlated with total profit that slow concession making in unidimen-
(r = .33, p < .05). This index was also pos- sional tasks is a form of pressure tactics akin
itively related to heuristic trial and error to positional commitment, this would appear
(r = .40, p < .01) and inversely related to to be the same effect as that found in the
pressure tactics (r = —.33, p < .05). present study.
The pressure tactics index was inversely Another way of describing the same re-
related to insight (r = -.30, p < .05). sults is that under high accountability, bar-
Information exchange was predictive of gainers who were face to face exhibited more
the profit of the less (though not the more) pressure tactics and achieved lower joint
successful bargainer (r = .34, p < .01). Our benefit than those who talked across a bar-
interpretation of this finding is that infor- rier. A broader generalization of this finding
mation exchange produces a mutual under- is possible in the light of Lewis and Fry's
standing of the two parties' priority struc- (1977) discovery of similar results in their
tures, making it possible to identify an individualistic-orientation condition. One
alternative that provides equally high out- can argue that when there is a competitive
comes to the two parties. This interpretation definition of the situation, visual access pro-
is supported by the finding of a significant duces pressure tactics and low joint benefit.
inverse correlation (r = —.35, p < .01) be- Visual access was found to diminish the
tween information exchange and the abso- cooperativeness of the interaction between
lute difference between the two parties' out- the bargainers under all levels of the ac-
comes. countability variable. When face to face, the
Though unrelated to total profit across the bargainers exchanged less information about
entire set of conditions, the three traditional values, employed more I's and fewer we's,
 LOOKING AND COMPETING 119

and indicated that the atmosphere of the types of problem-solving tactic, information
session had been less cooperative than when exchange and heuristic trial and error, were
talking across a barrier. also correlated with total profit. However,
The results described so far seem coun- failure to find similar orientation effects for
terintuitive if one considers that face-to- the data set as a whole is puzzling.
face contact is more intimate than contact
across a barrier. Why then should there be References
more efforts to dominate and more compe-
tition in the face-to-face condition? To an- Bartunek, J. M., Benton, A. A., & Keys, C. B. Third
swer this question, we must recall that bar- party intervention and the bargaining behavior of
group representatives. Journal of Conflict Resolution,
gaining tends to evoke a competitive 1975, 19, 532-557.
definition of the situation, especially under Benton, A. A. Accountability and negotiations between
high accountability. In such a context, we group representatives. Proceedings of the 80th An-
postulate that various elements of nonverbal nual Convention of the American Psychological As-
sociation, 1972, 7, 227-228.
behavior, such as staring, are employed in Benton, A. A., & Druckman, D. Salient solutions and
an effort to dominate and are interpreted as the bargaining behavior of representatives and non-
domineering behavior. When there is no vi- representatives. International Journal of Group Ten-
sual access, such nonverbal tactics are not sions, 1973, 3, 28-39.
available. Hence, there is less temptation to Benton, A. A., & Druckman, D. Constituent's bargain-
ing orientation and intergroup negotiations. Journal
try to dominate and less reason to view the of Applied Social Psychology, 1974,4, 141-150.
other party as trying to dominate. The result Carnevale, P. J. D., Pruitt, D. G., & Britton, S. D.
is fewer pressure tactics and more coopera- Looking tough: The negotiator under constituent sur-
tion. veillance. Personality and Social Psychology Bulle-
tin, 1979,5, 118-121.
One alternative interpretation would hold Deutsch, M. The resolution of conflict. New Haven,
that the visual image of the adversary in a Conn.: Yale University Press, 1973.
competitive context acts as a releasing cue Druckman, D. Dogmatism, prenegotiation experience,
for domineering and hostile behavior toward and simulated group representation as determinants
that adversary. Other alternative explana- of dyadic behavior in a bargaining situation. Journal
of Personality and Social Psychology, 1967, 6, 279-
tions are plausible. For example, negotiators 290.
within the visual presence of one another Druckman, D., Solomon, D., & Zechmeister, K. Effects
may meet with heightened objective self- of representational role obligations on the process of
awareness that leads them to act in a manner children's distribution of resources. Sociometry, 1972,
35, 387-410.
more appropriate for the given situation. The Frey, R. L., Jr., & Adams, J. S. The negotiator's di-
appropriate behavior, for example, employ- lemma: Simultaneous in-group and out-group con-
ing pressure tactics, may be more clearly flict. Journal of Experimental Social Psychology,
defined in the high accountability conditions. 1972,5, 331-346.
The significant main effects and the cor- Gahagan, J. P. Effects of promise credibility, outside
options and social contact on interpersonal conflict.
relations under high accountability suggest Dissertation Abstracts International, 1970, 30,
the following four-element causal explana- 3546-A.
tion for the accountability effect in the face- Cruder, C. L. Relationships with opponent and partner
to-face condition: Accountability produces in mixed-motive bargaining. Journal of Conflict Res-
olution, 1971, IS, 403-416.
dominance behavior that precludes negoti- Cruder, C. L., & Rosen, N. A. Effects of intragroup
ator understanding of the other party's prior- relations or intergroup bargaining. International
ities, resulting in the development of agree- Journal of Group Tensions, 1971, /, 301-317.
ments involving low mutual gain. Kelley, H. H., & Schenitzki, D. P. Bargaining. In
The results also provide further support C. G. McClintock (Ed.), Experimental social psy-
chology. New York: Holt, Rinehart & Winston, 1972.
for the flexible-rigidity hypothesis, which Kimmel, M., Pruitt, D. G., Magenau, J., Konar-Gold-
holds that high joint benefit results from a band, E., & Carnevale, P. J. D. Effects of trust, as-
combination of high, rigid aspirations and piration, and gender on negotiation tactics. Journal
the adoption of problem-solving tactics. As of Personality and Social Psychology, 1980, 38, 9-
22.
implied by this hypothesis, aspiration level Klimoski, R. J. The effects of intragroup forces on in-
was correlated with total profit. Further- \ergroup conflict resolution. Organizational Behavior
more, under high accountability the two and Human Performance, 1972, 8, 363-383.
120 P. CARNEVALE, D. PRUITT, AND S. SEILHEIMER

Klimoski, R. J., & Ash, R. A. Accountability and ne- P. J. D., Magenau, J. M., Peragallo, J., and Engram,
gotiator behavior. Organizational Behavior and Hu- P. The effect of accountability and surveillance on
man Performance, 1974, / / , 409-425. integrative bargaining. In H. Sauermann (Ed.), Con-
Kogan, N., Lamm, H., & Trommsdorff, G. Negotiation tributions to experimental economics. Tubingen,
constraints in the risk-taking domain: Effects of being Germany: Mohr, 1978.
observed by partners of higher or lower status. Jour- Pruitt, D. G., & Lewis, S. A. Development of integrative
nal of Personality and Social Psychology, 1972, 23, solutions in bilateral negotiation. Journal of Person-
143-156. ality and Social Psychology, 1975, 31, 621-633.
Lewis, S. A., & Fry, W. R. Effects of visual access and Pruitt, D. G., & Lewis, S. A. The psychology of inte-
orientation on the discovery of integrative bargaining grative bargaining. In D. Druckman (Ed.), Negoti-
alternatives. Organizational Behavior and Human ations: A social-psychological perspective. Beverly
Performance, 1977, 20, 75-92. Hills, Calif.: Sage-Halsted, 1977.
Milgram, S. Obediance to authority. New York: Harper Schulz, J. W., & Pruitt, D. G. The effects of mutual
& Row, 1974. concern on joint welfare. Journal of Experimental
Morley, I., & Stephenson, G. The social psychology of Social Psychology, 1978, 14, 480-491.
bargaining. London: George Allen & Unwin, 1977. Sen, A. K. Collective choice and individual values. New
Pruitt, D. G., & Carnevale, P. J. D. The development York: Holden-Day, 1970.
of integrative agreements in social conflict. In V. J. Walton, R. E., & McKersie, R. B. A behavior theory
Derlega & J. Grzelak (Eds.), Living with other peo- of labor negotiation. New York: McGraw-Hill, 1965.
ple: Theories and research on cooperation and help-
ing behavior. New York: Academic Press, 1980.
Pruitt, D. G., Kimmel, M. J., Britton, S., Carnevale, Received March 17, 1980

Manuscripts Accepted for Publication in the Section Interpersonal Relations and

Group Processes
Property Rights, Deservingness, Reciprocity, and Friendship: The Transactional Character of Children's
Sharing Behavior. Ervin Staub (Department of Psychology, University of Massachusetts, Amherst,
Massachusetts 01003) and Henry Noerenberg.
Is Empathic Emotion a Source of Altruistic Motivation? C. Daniel Batson (Department of Psychology,
University of Kansas, Lawrence, Kansas 66045), Bruce D. Duncan, Paula Ackerman, Terese Buckley,
and Kimberly Birch.
x ldentifiability as a Deterrent to Social Loafing: Two Cheering Experiments. Kipling Williams, Stephen
Harkins, and Bibb Latane (Behavioral Sciences Laboratory, Department of Psychology, Ohio State
University, 404B West 17th Avenue, Columbus, Ohio 43210).
Verbal, Vocal, and Visible Factors in Judgments of Another's Affect. Robert M. Krauss (Department
of Psychology, Schermerhorn Hall, Columbia University, New York, New York 10027), William
Apple, Nancy Morency, Charlotte Wenzel, and Ward Winton.
Heiderian Balance as a Group Phenomenon. Theodore M. Newcomb (1045 Cedar Bend Drive, Ann
Arbor, Michigan 48105).
The Effects of Prior Experience on Coalition Bargaining. S. S. Komorita (Department of Psychology,
University of Illinois, Champaign, Illinois 61820) and David A. Kravitz.
Sex Differences in Reward Allocation: Subjects, Partners, and Tasks. Harry T. Reis (Department of
Psychology, University of Rochester, Rochester, New York 14627) and Linda A. Jackson.
Audience-Induced Inhibition of Overt Practice During Learning. Seymour M. Berger (Department of
Psychology, University of Massachusetts, Amherst, Massachusetts 01003), Katherine L. Hampton,
Linda L. Carli, Paul S. Grandmaison, Janice S. Sadow, Clifford H. Donath, and Laura R. Herschlag.
Low-Ball Compliance Technique: Task or Person Commitment? Jerry M. Burger and Richard E. Petty
(Department of Psychology, University of Missouri, Columbia, Missouri 65211).
Effects of Communication Timing on Duopoly Bargaining Outcomes. Frank Stech and Charles G.
McClintock (Department of Psychology, University of California, Santa Barbara, California 93106).

View publication stats