How Conversations Change Over Time in Face-to-Face and Video-Mediated Communication

Abstract

An experiment was conducted to examine how communication patterns and task performance differ as a function of the group's communication environment and how these processes change over time. In a longitudinal design, three-person groups had to select and argue the correct answer out of a set of three alternatives for ten questions. Compared with face-to-face groups, video-teleconferencing groups took fewer turns, required more time for turns, and interrupted each other less. Listeners appeared to be more polite, waiting for a speaker to finish before making their conversational contribution. Although groups were able to maintain comparable performance scores across communication conditions, initial differences between conditions in communication patterns disappeared over time, indicating that the video-teleconferencing groups adapted to the newness and limitations of their communication environment. Moreover, because of increased experience with the task and the group, groups in both conditions needed less conversation to complete the task at later rounds. Implications are discussed for practice, training, and possibilities for future research.

Full text

http://sgr.sagepub.com
Small Group Research
DOI: 10.1177/1046496409333724
2009;
2009; 40; 355 originally published online Apr 15,Small Group Research
W. De Dreu
Rick van der Kleij, Jan Maarten Schraagen, Peter Werkhoven and Carsten K.
Video-Mediated Communication
How Conversations Change Over Time in Face-to-Face and
http://sgr.sagepub.com/cgi/content/abstract/40/4/355
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:Small Group Research Additional services and information for
http://sgr.sagepub.com/cgi/alerts Email Alerts:
http://sgr.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://sgr.sagepub.com/cgi/content/refs/40/4/355 Citations
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

355
Small Group Research
Volume 40 Number 4
August 2009 355-381
© 2009 SAGE Publications
10.1177/1046496409333724
http://sgr.sagepub.com
hosted at
http://online.sagepub.com
Authors’ Note: Part of this research was funded by the European Commission, Directorate-
General Information Society, project IST-1999-10044 VIRTUE. Please address correspondence to
Rick van der Kleij, Business Unit Human Factors, TNO Defence, Security and Safety, Kampweg 5,
P.O. Box 23, 3769 ZG Soesterberg, The Netherlands; e-mail: rick.vanderkleij@tno.nl.
How Conversations Change
Over Time in Face-to-Face
and Video-Mediated
Communication
Rick van der Kleij
Jan Maarten Schraagen
Peter Werkhoven
TNO Defence, Security and Safety, Soesterberg, The Netherlands
Carsten K. W. De Dreu
University of Amsterdam, The Netherlands
An experiment was conducted to examine how communication patterns and
task performance differ as a function of the group’s communication environ-
ment and how these processes change over time. In a longitudinal design,
three-person groups had to select and argue the correct answer out of a set
of three alternatives for ten questions. Compared with face-to-face groups,
video-teleconferencing groups took fewer turns, required more time for
turns, and interrupted each other less. Listeners appeared to be more polite,
waiting for a speaker to finish before making their conversational contribu-
tion. Although groups were able to maintain comparable performance scores
across communication conditions, initial differences between conditions in
communication patterns disappeared over time, indicating that the video-
teleconferencing groups adapted to the newness and limitations of their com-
munication environment. Moreover, because of increased experience with
the task and the group, groups in both conditions needed less conversation to
complete the task at later rounds. Implications are discussed for practice,
training, and possibilities for future research.
Keywords: geographical distribution; virtual teams; communication patterns;
team adaptation; video teleconferencing; longitudinal experiment
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

356 Small Group Research
C
lear and concise communication is essential to good teamwork (McIntyre
& Dickinson, 1997; McIntyre & Salas, 1995; Rochlin, LaPorte, &
Roberts, 1987; Salas, Burke, & Samman, 2001; Seifert & Hutchins, 1992;
Smith-Jentsch, Johnston, & Payne, 1998). Communication is a tool for
teams to accomplish their tasks and to coordinate their actions (Kanki,
Folk, & Irwin, 1991). It enables team members to state their intentions, and
to send and receive information among one another. Furthermore, it allows
team members to keep up-to-date with the changes in the situation, and to
evaluate and improve task performance (Rasker, 2002).
Because organizations increasingly rely on virtual teams, scholars and
practitioners alike have vented a considerable interest in understanding the
impact of distributed teamwork on communication and the nature of the
constraints imposed when one has to communicate over a communications
network. In virtual teams, the channel that teams use to communicate often
imposes restrictions on the communication process. A broad range of stud-
ies shows that when people have the ability to conduct face-to-face discus-
sions, group commitment, socialization, and control are improved. People
can use the full range of linguistic, paralinguistic, and nonverbal behaviors
to communicate and coordinate when physically collocated (Connaughton &
Daly, 2004; Kiesler & Cummings, 2002; Kraut, Fussell, Brennan, & Siegel,
2002). It is possible to use deictic gestures (i.e., pointing) to refer quickly
and easily to people, locations, and objects. Furthermore, because face- to-
face communication is produced in real time, speakers have direct feedback
on how their message is being understood as it is being delivered.
In virtual teams, current bandwidth limitations in telecommunication
networks and cost considerations make it difficult to transmit all the
sources of information available in face-to-face conversations at the same
rate and fidelity at which they occur naturally (Fussell & Benimoff,
1995). Consequently, correspondence is often unclear and infrequent
(Seetharaman, Samarah, & Mykytyn, 2004; see also Friedman & Currall,
2003). Miscommunication is often acerbated by the trouble that virtual
teams have in recognizing when disruptions or communication mix-ups
have occurred. Virtual teams are also slower to correct these misunder-
standings (Thompson & Coovert, 2006). Shin (2005) argues that in distrib-
uted settings it is difficult for members to receive guidance or instruction
on their tasks and roles from supervisors. At the same time, dispersed mem-
bers have fewer opportunities to elucidate their tasks and roles than collo-
cated members do. Therefore, virtual team members are more likely to
experience task, role, and responsibility ambiguity. This ambiguity can be
a source of conflict in virtual teams.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 357
To make things worse, use of the linguistic (verbal) channel of informa-
tion (cf. Fussell & Benimoff, 1995) that consists of conventional signals,
such as the words of a language, is also limited in virtual teams. Team
members in global virtual teams often do not speak the same language and
need to communicate in a nonnative tongue. Nonnative language use can
result in less depth and candidness of exchange, which has been found to
increase psychological distance and fears of misuse, miscommunication,
and misinterpretation of knowledge (Gibson & Cohen, 2004).
Finally, virtual teams have fewer opportunities for spontaneous, infor-
mal communication, which have been found to have a positive impact on
group functioning (Kiesler & Cummings, 2002). People who work in proxi-
mate settings run into one another at the water cooler, coffee machine, or
copier, and meet in the cafeteria. Therefore, it takes relatively little effort
for them to start interacting. These spontaneous, informal communication
opportunities increase the convenience and pleasure of conversations, the
frequency of communication, and allow for unplanned and multipurpose
interactions (Kraut et al., 2002). Ongoing work progresses better when
people have spontaneous casual communication and strong social and work
ties are supported. In short, one of the biggest challenges of virtual teams
is to overcome ineffective and inefficient interpersonal communication (Van
der Kleij, 2007).
The current research focuses on the role of experience in overcoming
communication deficiencies in distributed settings. Our main goal is to inves-
tigate how communication, satisfaction, and task performance differ as a
function of the group’s communication environment, whether distributed
or collocated, and how these variables change over time. Moreover, we
will explore the interactions between communication environment and
experience.
Interpersonal Communication
Conversations can be characterized in terms of the length of pauses, the
duration and number of turns, and number of interruptions (Sellen, 1995).
This aspect of communication refers more to how things are communi-
cated as opposed to the speech content (Kanki et al., 1991). Sellen (1995)
argues that patterns of communication are of interest because they may
reveal the manner in which a conversation is conducted and offer an alter-
native to task performance as a means for assessing the effect of commu-
nication media. For example, difficulties in the regulation of conversation
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

358 Small Group Research
may reveal itself in long turns, frequent interruptions, or unusually long
pauses.
Several studies have looked at how communication patterns change with
communication mode (for reviews, see Doherty-Sneddon et al., 1997;
Sellen, 1995). Most of these studies have involved comparisons of face-to-
face versus audio or text-based conditions. Although findings have been
inconsistent across studies, clearly differences exist between language use
in distributed and collocated face-to-face settings (Boyle, Anderson, &
Newlands, 1994). For example, it has been found that telephone conversa-
tions tend to be more formal, with fewer interruptions and longer utter-
ances than face-to-face conversations (Doherty-Sneddon et al., 1997). In
addition, text-based communication has been found to disrupt the regula-
tion of turn taking and to undermine higher-level language processes (i.e.,
the construal of meaning; Hancock & Dunham, 2001). Similarly, Ferrara,
Brunner, and Whittemore (1991) observed that interlocutors in text-based
settings tend to omit copulas, subject pronouns, and articles as a response
to the increased costs of formulating and producing utterances in these set-
tings as opposed to speaking.
Few studies have focused on communication patterns in video-mediated
communication (O’Conaill & Whittaker, 1997; Sellen, 1995). These studies
to some extent suggest that video-mediated communication has similar effects
on the conversational process as communicating through, for example, the
telephone. Furthermore, there is some indication that video-teleconferencing
technologies are attenuating the social context cues available in face-to-face
meetings. Social context cues are nonverbal signals (e.g., facial expres-
sions), paraverbal cues (e.g., voice volume), status and interpersonal cues
(e.g., age, physical appearance), and features of the physical surroundings
(e.g., office size; Dietz-Uhler & Bishop-Clark, 2001; Sproull & Kiesler,
1986; Straus, 1997). These cues help teams regulate interaction, express
information, and monitor feedback from others (Straus, 1997). During
computer-mediated communication (e.g., Internet chat) participants are
usually unaware of the facial expressions, tone of voice, posture, and so
forth of the other participants (Dietz-Uhler & Bishop-Clark, 2001). This is
often problematic. Attenuation of the social context cues available in face-
to-face meetings causes disruptions in conversations and difficulties in fol-
lowing and understanding discussions (Straus & McGrath, 1994). Indeed,
a recent survey of 214 managers revealed that 61% and 43% of these man-
agers have misinterpreted coworkers via e-mail and phone, respectively
(TANDBERG, 2007). Moreover, Wilson, Straus, and McEvily (2006)
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 359
found that the reduced availability of social context cues in distributed
communication environments changed the fundamental communication proc-
ess and constrained relational development in virtual teams. Wilson et al.
(2006) observed a lower incidence of behaviors associated with the devel-
opment of interpersonal relationships in virtual teams using Internet chat to
communicate, such as sharing and attending to personal information.
These studies notwithstanding, naturally occurring groups show some
variation in task, communication settings, and exhibit considerable fluctua-
tions in membership and attendance from time to time (McGrath, Arrow,
Gruenfeld, Hollingshead, & O’Connor, 1993). The few experimental stud-
ies that investigated the effects of long-term use of communication tech-
nologies suggest that temporal aspects can have a large impact on group
processes and outcomes (e.g., Hollingshead, McGrath, & O’Connor, 1993;
McGrath, 1993; McGrath et al., 1993; Van der Kleij, Paashuis, Langefeld,
& Schraagen, 2004; Van der Kleij, Paashuis, & Schraagen, 2005; Wilson et al.,
2006). The question remains, however, how experience with the task, group,
and communication mode affects the conversational process.
Experience and Change in Distributed Groups
Several studies acknowledge that one of the most important aspects
of teamwork is the ability to adapt to environmental opportunities and
constraints (Blickensderfer, Cannon-Bowers, & Salas, 1998; Entin &
Serfaty, 1999; Kozlowski, 1998; LePine, 2005; Serfaty, Entin & Johnston,
1998; Serfaty, Entin, & Volpe, 1993). Team members in effective teams
are able to use information from the task environment to adjust team
strategies, back each other up, make mutual adjustments, and reallocate
team resources (Blickensderfer et al., 1998). Serfaty et al. (1993) found
that teams with records of superior performance are extremely adaptable
to varying task demands. These teams switched between several coordi-
nation strategies and organizational structures, with different lines of
authority, communication patterns, and task responsibilities as they
move from normal to high-tempo or emergency situations. Serfaty et al.
(1993) asserted that reliable teams have not one, but several organiza-
tional structures and shift between them when needed. These teams
maintain open and flexible communication lines that promote the free
flow of information from the lowest to the highest levels as well as the
other way around. In adaptive teams, team members were also found to
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

360 Small Group Research
be sensitive to other members’ workload and performance in high-tempo
situations.
Adaptive response models stress response to environmental opportuni-
ties and constraints as a guiding force in group change (Arrow, 1997;
Arrow, Poole, Henry, Wheelan, & Moreland, 2004; McGrath, 1991). One
important aspect of these models is that the notion of equifinality is incor-
porated, a condition in which different initial conditions lead to similar
effects. This means that these models assume that different temporal pat-
terns or paths can lead to the same result: good performance. The path cho-
sen depends on task features or environmental constraints, such as time
pressure or limitations in the communication setting in which virtual teams
have to collaborate. These models assume that there exists no single best
path to match task and contextual demands, but assume that adaptation of
temporal patterns to these changing demands is essential to effective group
performance (Poole & DeSanctis, 1990).
McGrath et al. (1993) argue that adaptation is a function of the effects
of increased experience, as a group, with its tasks, with its technology, and
in its context. Adaptation is seen as a reaction to a mismatch between the
group’s structure and composition, its task structure, its set of tools and
resources for carrying out those tasks, and in the physical, organizational,
and temporal context within which the work is being performed. Increased
experience of a group reduces doubt with regard to distribution of labor and
responsibilities. This is expected to reduce the degree of interdependence
and the level of coordination and information exchange required in dis-
persed work groups. Moreover, increased experience with a given tool will
make collaborating easier in the sense that group members can become
more accustomed to the procedures involved in efficient collaboration.
This, too, reduces the degree of coordination required. As a group gains
experience using groupware, it will change the ways in which it uses those
tools in such a manner that initial constraints due to the newness and tech-
nological limitations of the groupware are compensated for (McGrath et al.,
1993). For example, Van der Kleij et al. (2005) found that distributed
teams, as compared to face-to-face teams, were able to compensate for
initial constraints due to their dispersed nature because of increased experi-
ence. This finding was explained in terms of a technological adaptation
effect. Technological adaptation occurs when people learn how to use the
technological tools available despite technological limitations, such as
restrictions in bandwidth (Dourish, Adler, Bellotti, & Henderson, 1996;
Olson & Olson, 2000; Qureshi & Vogel, 2001).
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 361
Based on the aforementioned literature review and discussion on the
effects of experience with group, task, and communication tools we formulate
the following hypotheses:
Hypothesis 1: Face-to-face groups will experience (a) fewer difficulties in
the regulation of conversation and (b) more satisfaction with group proc-
esses and outcomes. (c) Face-to-face groups will outperform video-
teleconferencing groups.
Hypothesis 2: Differences between face-to-face groups and video-teleconferencing
groups will disappear over time, as video-teleconferencing groups adjust to the
newness and limitations of the communication medium available to them.
Hypothesis 3: Increased experience with the group and the task will result in
fewer difficulties in the regulation of conversation, more satisfied group
members, and better performance in both communication conditions.
Method
Participants
A total of 66 persons (30 men) participated in a longitudinal experiment.
Participants were students and their age ranged from 18 to 32 years. In the
face-to-face condition, six groups consisting of 2 men and 1 woman and
five groups consisting of 2 women and 1 man participated. In the video-
teleconferencing condition, four groups of 2 men and 1 woman and five
groups of 2 women and 1 man participated. In the video-teleconferencing
condition, two groups consisted of 3 women. None of the group members
knew each other prior to the experiment. For participating in the experi-
ment, we awarded each participant with €40 per test session, which is about
US$55. The total amount for participating in all test sessions was €160,
which we paid to each participant after the fourth test session. The best per-
forming group over all four test sessions in each experimental condition also
had the prospect of an extra bonus of €40 for the group as a whole, to enhance
motivation and stimulate groups to remain in the experiment until the end.
Apparatus
Face-to-face environment. Groups in the face-to-face condition were
placed in the same room with a round table (diameter of 180 cm). A video
camera was set up in one upper corner of the room; participants wore
Shure unidirectional close-talking microphone headsets (model SM10A).
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

362 Small Group Research
The audio output from headsets was fed through coaxial cables to a
speech-storage unit and stored as separate tracks on removable hard disks
for archiving and further analysis.
Video-mediated environment. Participants of the video-teleconferencing
groups were located in different rooms. Each group was seated in a room
at an oval half round table (diameter of 180 cm). In front of them was a
large rear projection screen of 146 × 86 cm. Figure 1 shows a schematic of
the experimental video-teleconferencing condition.
Behind each participant in the video-teleconferencing condition, blue parti-
tions were placed to create a uniform background. Participants in the video-
teleconferencing condition wore the same unidirectional microphone
headsets as participants in the face-to-face condition to record the audio
communication. In this condition, the sound was also amplified and trans-
mitted to the other two rooms so group members could communicate with
NETWORK
B
C
A
C
AB
B
C
A
STEREO SPEAKERS
VIRTUAL PA RTICIPA NTS
ON SCREEN
VIRTUAL TA BLE
NETWORK
B
C
A
C
AB
B
C
A
REAL
TABLE
BINOCULAR (2)
CAMERA SYSTEM
LOCAL PA RTCIPANT
EQUIPPED WITH
MICROPHONE
HEADSET
Figure 1
Experimental Setup
Note: Two cameras were situated on both sides of the screen and were used for the transmis-
sion of images to the other locations. The cameras were positioned on the sides of the screen
to create an isotropic condition enabling selective gaze. Speech was amplified and transmitted
to the other two rooms so the participants could communicate with each other.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 363
each other. Two Panasonic WV-gp150E cameras filmed the participants
and transmitted images to the other rooms in the video-teleconferencing
condition. The cameras were situated on both sides of the screen. Although
the camera positions did not allow participants the establish eye contact,
gaze direction could be determined by the position of the head of the
speaker (i.e., an isotropic condition enabling selective gaze was created).
The images recorded by the video cameras were projected by two Hitachi
XGA cp x970 beamers.
A third camera was located on top of the screen and was used to monitor
progress on the experimental task. A Panasonic VN 484p intercom system
was used to inform the participants when to start and stop working. Factors
affecting the quality and intelligibility of the speech, and those interfering
with the process of communication were kept to a minimum. For example,
we were able to reduce noise levels, echoes, reverberation, and cross-talk
between the sound source and the receiver using directional microphones.
Further, because there was no processing required for coding and decoding
of the audio and video signals, no unwanted round-trip delays and delays
between audio and video were present.
The Science Quiz Task
For this experiment, we designed a group task that had the potential to
induce differences in communication patterns between communication
conditions. Groups had to select and argue the correct answer out a set
of three alternatives, a task that can be classified as an intellective task
(McGrath, 1984). There were 4 sets of 10 questions each, one set for each
consecutive test session. The questions (including the answers and writ-
ten argumentations) were selected from the archives of an annual Dutch
television show called The Science Quiz. Questions that were selected
had to be at least 4 years old, so the probability of our participants knowing
the questions from previous television shows was reduced. We divided
the questions at random over the four versions of the task. Topics of the
questions ranged on a variety of fields, such as mathematics and biology.
Sample questions are “Why are bananas curved? (a) The curve is geneti-
cally defined; (b) Because of gravity; (c) Because of exposure to (sun)
light”; and “Imagine that you have a glass of water with one large ice
cube. What happens to the level of the water when the ice cube melts?
(a) Nothing, the level does not change; (b) The level rises; (c) The level
sinks.”
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

364 Small Group Research
Design
We tested 22 groups in two communication environments (face-to-face
same-room and video-teleconferencing noncollocated communication) at
four consecutive test sessions for 1 hour at 2-week intervals making a 2 × 4
mixed repeated measures design. In the face-to-face condition, we started
out with 13 groups and in the video-teleconferencing condition there were
16 groups at test session 1. At the end of Test Session 4, 11 groups remained
in both conditions. The other 7 groups dropped out due to several reasons,
such as illnesses of group members or problems with making follow-up
appointments with the experimenter or fellow group members. The findings
in this article are based on the remaining 22 groups.
Dependent Variables
The following three dependent variables were assessed: (a) communica-
tion patterns, (b) satisfaction, and (c) task performance. We discuss these
categories in the following subsections.
Communication Patterns
Specialized speech-tracking equipment was used to converse the speech
waves that were stored on removable hard disk into on-off patterns for each
participant. When a participant spoke louder than a preset threshold, a talk
spurt was recorded and stamped with a time code. For each participant
these talk spurts were laid down on a speech-time line showing when the
on–off patterns of speech took place with respect to each other. This ena-
bled us to analyze the lower levels of language use (i.e., the length of
pauses, the duration and number of turns, number of interruptions, etc.).
Based on the speech time-lines that were generated, the software, threshold,
and specialized equipment were checked for correctness against videotape
data, and parameters were adjusted, if appropriate, to compensate for dif-
ferences between the speech time-lines and the videotape data (for a more
elaborate discussion, see Sellen, 1995). The categorization definitions we used
to analyze the data are described below.
Duration of pauses. Pauses are moments of silence between two successive
speakers. The average duration of pauses was measured in milliseconds.
Turn duration
. Turn duration was defined as the average time in milli-
seconds for each turn. Speech was identified as a turn when it lasted for at
least 1.5 seconds and was not interrupted during that time.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 365
Number of turns. The numbers of turns was defined as the average
number of turns per group. The average number of turns was divided by the
total amount of seconds that the group was working on the task. In this man-
ner, the score was made independent of task duration. As with turn duration,
speech was identified as a turn when it lasted for at least 1.5 seconds and
was not interrupted during that time.
Number of simultaneous speech events. Simultaneous speech was defined
as two or more participants that begin speaking at the same time (i.e., within
1 second after each other). We calculated the average number of simultaneous
speech events.
Number of interruptions with a turn switch. This variable was defined as
an interruption of the current speaker resulting in a turn switch. The inter-
ruption must occur more than 1 second after the floor holder has started
his turn.
Number of interruptions without a turn switch. This variable was defined as
a short interruption of the current speaker. After the interruption the same
speaker still holds the floor. If two persons interrupt the speaker at the same
time this is categorized both as simultaneous speech and as an interruption. The
interruption must occur more than one second after the floor holder has started
his turn to be scored as an interruption. Scores were computed per group.
Satisfaction
A satisfaction questionnaire was administered at the end of each test
session after completion of the task. The questionnaire was adapted from
Dennis (1996), translated into Dutch, and tailored to our setting and par-
ticipants. It contained the following five items pertaining to satisfaction
about the task, other members, and the communication processes of inter-
acting with other members: “I am satisfied about the course of discussions
in our team,” “I am satisfied about the quality of the interactions in our
team,” “I am satisfied with the quality of the outcome of our team,” “To
what extent do you feel that you have contributed to the team’s final out-
come?,” and “To what extent do you feel that all team members had equal
input to the team’s final outcome?” (Cronbach’s α = .95). All items were
measured on a 7-point Likert-type scale, in which a score of 1 corresponds
to the most negative response to a statement and a score of 7 corresponds
to the most positive response.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

366 Small Group Research
Task Performance
Scores on the science quiz task were determined by the number of ques-
tions that were answered correctly and by the quality of the group’s argu-
mentation. For each correct answer, we awarded 1 point to the team to a
maximum score of 10 points for each test session. If the answer was correct,
an additional point was awarded for the correctness of the written argu-
ment. Thus, the maximum score was 20 points per team per test session.
Two raters simultaneously judged the arguments of the groups. To exclude
effects of raters’ unintentional preferences, group products were numbered,
presented in random order to the raters, and we removed all information
linking them to the experimental conditions.
Procedure and Instructions
As soon as participants arrived at the lab, we escorted them to their seats
where they received a general instruction about the experiment. Participants
were told that the experiment was designed to study group processes. After
all participants had read the general instructions, we gave them task-specific
instructions. After clarifying the instructions and answering questions
about the task, 10 questions were divided at random among the three group
members, which meant that one person received 4 questions and both oth-
ers received 3 questions each. One person was made responsible for writing
down the answers and accompanying arguments on which the group had to
reach consensus. The participants were told that the questions had to be
answered in fixed order within 30 minutes, which was more than enough
time to answer all questions. Then the experimenter left the room and through
the intercom told the groups to start working. When time was up or the
group had finished before time, the experimenter entered the room again
and collected the task materials. The group members then completed ques-
tionnaires on satisfaction and telepresence. The latter questionnaire we
used to check for the correctness of our dispersion manipulation.
The video-teleconferencing groups were seated in separate rooms and
communication was mediated through a videoconferencing system. These
participants received additional instructions pertaining to the use of the
video-teleconferencing system. To maintain as much control as possible,
participants were not allowed to leave their rooms during the experiment to
prevent them from getting to know the other group members outside the
test situation. For the same reason, we collected the participants one at a
time at their arrival and escorted them in the same manner to the main exit
of our lab facilities at the end of the test session. Furthermore, we asked our
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

367
Table 1
Cell Means and Standard Deviations as Function of Communication Condition
(Face-to-Face Versus Video Teleconferencing) and Test Session (N = 22)
Test Session
Dependent Variable Condition 1 2 3 4 Row Average
Duration of pauses
a
FTF 0.86 (0.12) 0.97 (0.31) 0.95 (0.14) 1.06 (0.37) 0.96 (0.26)
VTC 0.91 (0.22) 1.02 (0.28) 1.08 (0.27) 1.06 (0.29) 1.02 (0.26)
Turn duration
a
FTF 1.32 (0.14) 1.35 (0.13) 1.33 (0.10) 1.33 (0.13) 1.33 (0.12)
VTC 1.48 (0.12) 1.45 (0.09) 1.47 (0.13) 1.44 (0.10) 1.46 (0.11)
Number of turns
b
FTF 0.20 (0.02) 0.19 (0.03) 0.19 (0.02) 0.18 (0.03) 0.19 (0.02)
VTC 0.18 (0.02) 0.17 (0.02) 0.16 (0.01) 0.17 (0.02) 0.17 (0.02)
Number of simultaneous FTF 28.67 (4.94) 25.98 (6.12) 25.67 (4.52) 23.41 (5.95) 25.88 (5.58)
speech events
b
VTC 21.44 (5.38) 19.47 (3.16) 17.91 (2.63) 19.52 (4.11) 19.58 (4.03)
Number of interruptions FTF 15.63 (6.66) 14.17 (5.72) 14.08 (4.94) 12.50 (4.98) 14.09 (5.53)
with turn switch
b
VTC 13.67 (4.89) 12.61 (5.30) 10.96 (5.71) 11.84 (3.72) 12.27 (4.89)
Number of interruptions FTF 14.96 (7.93) 15.13 (7.59) 13.08 (5.39) 13.79 (6.54) 14.24 (6.74)
without turn switch
b
VTC 14.75 (10.37) 9.57 (4.00) 8.64 (4.83) 10.21 (3.11) 10.79 (6.49)
Satisfaction
c
FTF 5.66 (0.36) 5.95 (0.59) 5.88 (0.46) 5.68 (0.63) 5.79 (0.52)
VTC 5.53 (0.49) 5.41 (0.59) 5.42 (0.49) 5.48 (0.56) 5.46 (0.52)
Task performance
d
FTF 7.36 (3.28) 6.82 (2.40) 6.82 (3.94) 7.09 (3.01) 7.02 (2.51)
VTC 7.91 (3.96) 8.82 (4.17) 7.18 (2.70) 8.55 (3.95) 8.11 (2.31)
Note: Values enclosed in parentheses represent standard deviations (SD). FTF = face-to-face; VTC = video teleconferencing.
a. The values on this scale represent the average time in milliseconds.
b. The values on this scale represent the total number of speech occurrences.
c. The values represent mean scores on 7-point Likert-type scales, in which a score of 1 corresponds to the most negative option and a score of 7 corresponds
to the most positive option.
d. The values represent mean task scores for each condition. The maximum score is 20.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

368 Small Group Research
participants to refrain from discussing the experiment during unforeseen
encounters with members of their own and other groups. After the final test
session, we debriefed and interviewed participants, thanked them for their
cooperation, and excused them.
Results
An analysis of variance (ANOVA) for repeated measures designs was
carried out on all variables at the team level to account for statistical inter-
dependence. Data were analyzed in a 2 (communication group) × 4 (test
session) mixed-model ANOVA with the last factor within groups. The prob-
ability of a Type I error was maintained at .05 for all analyses. Table 1 sum-
marizes the means and standard deviations for the dependent variables
across conditions.
Manipulation Check
The perceived physical distance between group members, also called tel-
epresence, was assessed at the end of each test session on a questionnaire that
included the following item: “During the experiment I had the feeling we
were all in the same room.” We used a 7-point Likert-type scale, in which
a score of 1 corresponds to the most negative response to a statement and a
score of 7 corresponds to the most positive response. The analysis revealed
a significant main effect for communication condition, F(1, 20) = 42.98, p <
.001. Face-to-face groups scored higher on this question (M = 6.63; SD =
0.39) than video-teleconferencing groups (M = 4.52; SD = 1.06), meaning
that the experimental manipulation produced the desired psychological
effect. The average score for the video-teleconferencing groups was actually
quite high, indicating that the perceived physical distance between group
members was low and, therefore, that our teleconferencing system was capa-
ble of inducing the impression of sharing space with the participants at the
remote site. No significant main effect for test sessions was found, F(3, 60) =
1.54, ns. Likewise, no interaction effect was found between communication
condition and test sessions, F(3, 60) = .96, ns.
Communication Patterns
Duration of pauses. No significant differences in duration of pauses
were found between the face-to-face groups (M = 0.96; SD = 0.26) and the
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 369
video-teleconferencing groups (M = 1.02; SD = 0.26), F(1, 20) = 0.48, ns.
The analysis did reveal a significant main effect of test sessions for duration
of pauses, F(3, 60) = 3.19, p < .05. As can be seen in Table 1, overall dura-
tion of pauses increased as groups gained experience. No interaction
effect was found between communication condition and test sessions,
F(3, 60) = 0.40, ns.
Turn duration. The analysis revealed a significant main effect for
communication condition on turn duration, F(1, 20) = 9.66, p < .01. The
face-to-face groups took less time for turns (M = 1.33; SD = 0.12) than the
video-teleconferencing groups (M = 1.46; SD = 0.11). No significant main
effect for test sessions was found, F(3, 60) = 0.20, ns, suggesting that over
time no increase or decrease was present in turn duration. Although the
interaction between communication condition and test sessions was not
significant, F (3, 60) = 0.59, ns, post hoc analysis revealed significant dif-
ferences between groups on Test Sessions 1, 2, and 3 but not on Test Session
4 (see Table 1). This is indicative of an adaptation effect.
Number of turns. A significant main effect for communication condition
was found for number of turns, F(1, 20) = 14.33, p < .001. The video-
teleconferencing groups required fewer turns to complete the task (M =
0.17; SD = 0.02) than face-to-face groups (M = 0.19; SD = 0.02). Post hoc
analysis showed that significant differences between conditions on Test
Sessions 1, 2, and 3 were responsible for this effect (also see Table 1). We
found no significant differences between conditions on Test Session 4, sug-
gesting an adaptation effect. Furthermore, a significant main effect for test
sessions was revealed, F(3, 60) = 3.30, p < .05. Examination of the data
revealed that the overall number of turns decreased as groups gained expe-
rience. No interaction effect was found between communication condition
and test sessions, F(3, 60) = 0.92, ns.
Number of simultaneous speech events. The analysis revealed a signifi-
cant main effect for communication condition for number of simultaneous
speech events, F(1, 20) = 20.40, p < .001. Simultaneous speech events
occurred less often for video-teleconferencing groups (M = 19.58; SD = 4.03)
than for face-to-face groups (M = 25.88; SD = 5.58). Post hoc analysis
revealed that significant differences between conditions on Test Sessions
1, 2, and 3 caused this effect. No significant differences between conditions
were found on Test Session 4, again suggesting an adaptation effect.
Furthermore, the analysis revealed a significant main effect of test sessions
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

370 Small Group Research
for number of simultaneous speech events, F(3, 60) = 3.78, p < .05. The
overall number of simultaneous speech events decreased as groups gained
experience. Additional analysis revealed that this effect was due to face-to-
face groups, F(3, 30) = 8.68, p < .01. A significant linear trend character-
ized this decrease, F(1, 20) = 8.37, p < .01. Thus, the number of
simultaneous speech events for face-to-face groups decreased at a constant
rate as those groups gained experience (also see Table 1). No interaction
effect was found between communication condition and test sessions for
the number of simultaneous speech events, F(3, 60) = 1.14, ns.
Number of interruptions with turn switch. For the number of interrup-
tions with turn switch, no significant main effect for communication condi-
tion was found, F(1, 20) = 1.02, ns. In other words, no significant differences
were present on this variable between face-to-face groups (M = 14.09;
SD = 5.53) and video-teleconferencing groups (M = 12.27; SD = 4.89).
Further, no significant main effect for test sessions was found, F(3, 60) =
1.97, ns. Likewise, no interaction effect was found between communication
condition and test sessions, F(3, 60) = 0.42, ns.
Number of interruptions without turn switch. No significant differences
were found for the number of interruptions without turn switch between
face-to-face groups (M = 14.24; SD = 6.74) and video-teleconferencing
groups (M = 10.79; SD = 6.49), F(1, 20) = 2.67, ns. The analysis revealed a
significant main effect for test sessions, F(3, 60) = 2.45, p < .05. Additional
analysis revealed a significant effect for the video-teleconferencing condi-
tion on test sessions, F(3, 30) = 2.92, p < .05. Significant linear, F(1, 20) =
4.02, p < .05, and significant quadratic, F(1, 20) = 3.50, p < .05, trends
characterized this decrease. Thus, the numbers of interruptions without a
turn switch for video-teleconferencing condition decreased at a decreasing
rate as those groups gained experience. No interaction effect was found
between communication condition and test sessions, F(3, 60) = 1.15, ns.
Satisfaction
The analysis revealed a significant effect between communication condi-
tions on the satisfaction questionnaire, F(1, 20) = 3.72, p < .05. Face-to-face
groups (M = 5.79; SD = 0.52) were more satisfied than video-teleconferenc-
ing groups (M = 5.46; SD = 0.52). No significant main effect was found for
test sessions, F(3, 60) = 0.32, ns, and no significant interaction effect was
found for conditions and test sessions, F(3, 60) = 1.34, ns.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 371
Task Performance
No significant main effect for communication condition was found for
task performance, F(1, 20) = 1.05, ns, indicating that groups did not differ
in task performance across communication conditions. In addition, no sig-
nificant main effect for test sessions was found on task performance, imply-
ing that there was no occurrence of adaptation on task performance over
time F(3, 60) = 0.37, ns. Likewise, no interaction was found between com-
munication condition and test sessions, F(3, 60) = 0.36, ns.
Discussion
Our goal was to examine how communication patterns, satisfaction, and
task performance differ as a function of the group’s communication envi-
ronment, and how these variables change over time. To investigate the
effects of different communication environments, we compared video-
teleconferencing sessions with face-to-face same-room meetings. In the
following subsections, we discuss the results and the limitations of our
research and draw implications for future research, training, practice, and
for the design of teleconferencing technologies.
Effects of Communication Environment
We expected that, due to the challenges imposed on distributed team-
work, face-to-face groups would experience fewer difficulties in the regula-
tion of conversation, experience more satisfaction, and would outperform
video-teleconferencing groups. The results partially confirmed our first
hypothesis. The communication patterns of video-teleconferencing groups
indicate that these groups experienced more difficulties in the regulation of
conversation than face-to-face groups (Hypothesis 1a). Moreover, members
of face-to-face groups were more satisfied than video-teleconferencing
groups (Hypothesis 1b). However, video-teleconferencing groups suffered
no decrement in performance as compared with face-to-face groups
(Hypothesis 1c).
First, consistent with previous research comparing face-to-face and tel-
ephone conversations (O’Conaill & Whittaker, 1997; Sellen, 1995), our
results showed that, compared with face-to-face communication, video-
teleconferencing groups took fewer turns, required more time for turns, and
interrupted each other significantly less. In other words, the interaction
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

372 Small Group Research
process in video-teleconferencing groups was more formal, lecture-like
with long turns. Listeners appeared to be more polite, waiting for a
speaker to finish before making their conversational contribution. In gen-
eral, these results suggest that video-teleconferencing technologies, although
capable of inducing the impression of sharing space with the participants at
the remote site, are to some extent attenuating the social context cues that
are considered important for coordinating turn taking during face-to-face
communication (cf. Boyle et al., 1994).
Second, we found lower levels of perceived satisfaction with group
processes and outcomes in video-teleconferencing groups as compared to
face-to-face groups. Baltes, Dickson, Sherman, Bauer, and LaGanke (2002)
found similar results. With meta-analysis, Baltes et al. (2002) compared
decision making in face-to-face with distributed computer-mediated com-
munication groups (e.g., chat groups). Their results suggest that, in general,
and compared with face-to-face groups, computer-mediated communica-
tion leads to decreases in satisfaction. Presumably, the relative lack of
social context cues in the video-teleconferencing setting led to the frustra-
tion and dissatisfaction of the video-teleconferencing group members.
Likewise, Mühlbach, Böcker, and Prussog (1995) found a positive relation
between telepresence and satisfaction: The greater feeling of being in direct
contact with the participants at the remote site led to a higher degree of
satisfaction with video teleconferencing. According to their line of reason-
ing, the communication setting creating the richest presence experience
(i.e., face-to-face communication) should lead to the highest degree of sat-
isfaction. This seems to underline the importance of mediating a sense of
sharing space with the participants at the remote site in the development of
innovative communication technologies.
Third, although video-teleconferencing groups took fewer turns, required
more time for turns, and interrupted each other less, we found no perform-
ance differences between face-to-face and video-teleconferencing groups.
Groups were able to maintain comparable performance scores across com-
munication conditions in spite of differences in communication patterns. It
appears that group members adapted their communication patterns to their
distributed working environment in an effort to maintain high and stable
performance. This finding is consistent with previous studies (Boyle et al.,
1994, Williams, 1977). In general, group members adapt to limitations in
their working environment by accelerating their cognitive processing,
increasing the amount of their implicit coordination by sending more infor-
mation without being asked for it, and, to a lesser extent, omitting certain
activities (Adelman, Miller, Henderson, & Schoelles, 2003). It is likely that
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 373
to maintain a high performance on the task, communication strategies were
altered in such a manner that task performance required less communica-
tion and resources were freed for the more demanding cognitive processing
required in distributed group work.
These finding have some interesting implications for practice. In accord-
ance with previous research, our results confirm the beneficial effects of
proximity on communication. As it seems, proximity helps to conduct a
conversation and has beneficial effects on satisfaction (cf. Kiesler &
Cummings, 2002; Kraut et al., 2002). In collocated settings, people can use
the full range of linguistic, paralinguistic, and nonverbal behaviors to com-
municate (Kraut et al., 2002), and use deictic gestures (i.e., pointing) to refer
quickly and easily to people, locations, and objects. This general notion
implies that collocation makes it less difficult for group members to com-
municate information to others and to interpret the communications of oth-
ers (cf. Driskell, Radtke, & Salas, 2003). Thus, given the importance of
collocation, a potential challenge for virtual teams becomes how to incorpo-
rate face-to-face meetings into their routines (Connaughton & Daly, 2004).
Alternatively, it is possible that in time and with careful management,
technological substitutes will provide the same critical routine as face-to-
face meetings. Thompson and Coovert (2006) argue that chat rooms can be
a viable alternative for teams lacking the resources to hold face-to-face
meetings. Chat allows people to have a real-time conversation, either between
two people or in chat rooms between more people at the same time. Nardi
and Whittaker (2002) suggested that chat could aid in developing a sense
of social connection and awareness, and stimulate impromptu, or water-
cooler conversations that nurture understanding and trust.
Indeed, Zheng, Bos, Olson, and Olson (2001) and Zheng, Veinott, Bos,
Olson, and Olson (2002) found that team members who were unable to
meet face-to-face but rather engaged in chat-based conversations to be
acquainted with each other before collaborating demonstrated more trust
than teams who did not hold such meetings. In addition, Zheng et al. (2002)
demonstrated that, when it is not possible to hold a kick-off meeting (i.e.,
the first meeting with the team members to discuss the work that needs to
be completed), it is valuable to present a photograph of each distributed
team member in the chat room to promote interpersonal trust. In contrast to
the static presentation of biographical information of fellow team members,
Zheng et al. (2002) argue that a photo improves the building of trust
because the visual identification emphasizes interlocutor’s humanity and
implies accountability (i.e., one can be visually recognized by their fellow
team members in the future). Future research, however, is required in this
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

374 Small Group Research
area before clear mechanisms emerge on how to incorporate face-to-face
meetings, or alternatives like chat meetings, into the routines of virtual
teams.
Effects of Experience
Relating to adaptation, we hypothesized that initial differences between
face-to-face groups and video-teleconferencing groups would disappear
over time, as video-teleconferencing groups would adjust to the newness
and limitations of the communication medium available to them (Hypothesis 2).
This hypothesis was partly confirmed. At the end of the experiment, initial
differences between communication conditions on turn duration, number of
turns, and number of simultaneous speech events had disappeared, suggest-
ing an adaptation effect on these communication patterns. This implies that
video-teleconferencing groups are able to adapt to their given communica-
tion environment and that possible problems in the regulation of conversa-
tion are, to some extent, compensated for.
This finding has an important implication for the design of groupware. It
means that groupware engineers and designers could run the risk of devoting
time, effort, and financial resources to the development of innovative solutions
to nonproblems: problems that diminish in time as people learn effective
practices to adapt to the technologies available to them (see also Dourish et al.,
1996). In practice, the implication of an adaptation effect is that video-
teleconferencing groups that are new to the communication medium and
must learn to operate it, must be allowed some time to adapt the medium’s
structures into their work processes. This finding is also relevant for the
training of virtual teams. Some argue that the nature and characteristics of
virtual teams require initial training of communication skills, and skills and
knowledge that enable team members to operate available groupware
(Cascio, 2000; Hertel, Geister, & Konradt, 2005; Horvath & Tobin, 2001;
Knoll & Jarvenpaa, 1998). However, rather than developing and delivering
expensive training sessions, this research shows that virtual teams that have
recurring tasks and anticipate continued future interaction should simply be
allowed to learn and adapt while performing their task.
Also interesting was the finding that over time, groups in both commu-
nication conditions needed less conversation to complete the task, provid-
ing evidence in favor of our third hypothesis (Hypothesis 3). The overall
numbers of turns decreased as groups gained experience and the overall
duration of pauses increased. Furthermore, the number of simultaneous
speech events for face-to-face groups, and, for the video-teleconferencing
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 375
groups, the numbers of interruptions without a turn switch decreased. In
other words, the volume of communication diminished over time (cf. Briggs
& Johnston, 1967).
A plausible explanation for this finding is the development of shared and
thus more efficient language. The development of such a common vocabu-
lary is called grounding (Clark & Brennan, 1991; Clark & Carlson, 1982).
According to Clark and Brennan (1991), grounding is essential to commu-
nication. Once a message is formulated, people must do more than just send
it off. They need to assure themselves that it has been understood as it was
intended to be. Otherwise, there is little assurance that the discourse they
are taking part in will proceed in an orderly way. Thus, grounding involves
reaching a state of mutual understanding about what was said and meant.
By the principle of least collaborative effort, people try to ground with as
little combined effort as needed. Although groups need to update their com-
mon ground moment by moment, especially when starting up a new dis-
course the associated costs of reaching a state of mutual understanding and
common vocabulary can be compounded. This explains why at the begin-
ning of the collaboration there was more communication.
An alternative explanation is that groups developed a shared knowledge
structure, or transactive memory system, which allowed them make better
use of each individual’s expertise (Hollingshead, 2001; Jackson & Klobas,
2008). Transactive memory is the shared division of cognitive labor with
respect to the encoding, storage, retrieval, and communication of informa-
tion from different domains that often develops in close relationships
(Hollingshead, 1998; Wegner, 1987). Through self-disclosure and shared
experiences, group members learn who the expert is across knowledge
domains, and members rely on one another to access information in appro-
priate domains. When group members know who is good at what, coordina-
tion is likely to improve. Familiarity with one another’s strengths and
weaknesses helps members to anticipate behavior rather than merely react
to it (Moreland, Argote, & Krishnan, 1996). Even when specific tasks have
not been assigned to particular members, group members can coordinate
more smoothly, efficiently, and implicitly. This means that experienced
groups need less discussion to complete the same amount of work as groups
that share no task experiences.
Future research should examine more fully the impact of communica-
tion environment on the development of shared knowledge structures, a
special case of which is a transactive memory system. To develop a shared
knowledge structure it is important that collaborators are enabled to moni-
tor their partners’ activities and to understand the impact of their partners’
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

376 Small Group Research
progress on their own work (Kraut et al., 2002). Monitoring can help peo-
ple determine when and which collaborative actions are required. In other
words, monitoring helps determine whether it is time to harass someone to
complete his or her section of the task. The major challenge, according to
Kraut et al., is to develop tools or procedures for virtual teams that allow
people to monitor their partners’ activities and aid the development of
awareness. It is well imaginable that a communication environment capable
of conveying cues containing emotional, attitudinal, normative, and other
meanings, beyond the literal cognitive denotations of the symbols used to
express it (e.g., face-to-face communication), fosters more swiftly the
development of shared knowledge of the task, roles, responsibilities, and
informational needs of fellow group members.
There are several examples of groupware that claim to help virtual teams
in the development of shared awareness. Most of them focus on maintain-
ing so called passive awareness. This means that these technologies aim to
provide collaborators with knowledge of the other team members. For
example, Bolstad and Endsley (1999) investigated the use of shared dis-
plays as a means of enhancing shared awareness. A comparison was made
between teams in which the members were able to view each other’s screen
while working and teams in which team members could not view each
other’s screen. Bolstad and Endsley concluded that the shared-display set-
ting supported the development of shared awareness and improved per-
formance. Nevertheless, team members who could view each other’s screen
were initially slower. The reason for this was twofold. First, team members
used extra time looking at the teammate’s screen. Second, team members
used more time to develop knowledge of the teammate’s task. The major
design challenge is therefore to share the information needed to maintain
awareness, detect relevant changes in the environment, and then present
these changes to interested parties without overwhelming them so much
that they are unable to do their own work (Bolstad & Endsley, 1999).
Contrary to our expectations, neither satisfaction nor task performance
increased as groups gained experience with the task, the group, and the
communication setting. An explanation, which is compatible with the other
accounts we have presented, involves the task we used in the experiment.
Because we used a rather simple multiple-choice task that did not require
skilled behavior, it is well possible that our distributed environment did not
pose a real hindrance to our participants to maintain high and stable
performance. This could also have led to the relatively high levels of satis-
faction. Because there was ample time to perform the task, there was no
challenge involved and groups in both communication conditions per-
formed at their maximum at the beginning of the experiment: Participants
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 377
either knew or did not know the answer to a specific question. It is not
likely that the knowledge our participants possessed required answering the
questions increased over time due to increased experience with the task.
Clearly, more research is needed to more fully understand the effects of
experience. For example, it would be interesting to replicate this longitudi-
nal approach using a task that requires rich communication channels and
has the potential to lead to substantial variances in group performance to
investigate the effects of experience on task performance more thoroughly
(cf. Van der Kleij et al., 2005). Furthermore, it would also be interesting to
impose a more stringent time limit on teamwork. This can have severe
detrimental effects on task performance (Kelly, Jackson, & Hutson-
Comeaux, 1997). This is especially interesting in distributed settings,
because, according to Walther (2002), the associated challenges of distrib-
uted teamwork are accentuated when time is, or becomes, short.
References
Adelman, L., Miller, S. L., Henderson, D., & Schoelles, M. (2003). Using Brunswikian theory
and a longitudinal design to study how hierarchical teams adapt to increasing levels of
time pressure. Acta Psychologica, 112, 181-206.
Arrow, H. (1997). Stability, bistability, and instability in small group influence patterns.
Journal of Personality and Social Psychology, 72, 75-85.
Arrow, H., Poole, M. S., Henry, K. B., Wheelan, S., & Moreland, R. (2004). Time,
change, and development: The temporal perspective on groups. Small Group Research,
35, 73-105.
Baltes, B. B., Dickson, M. W., Sherman, M. P., Bauer, C. C., & LaGanke, J. S. (2002).
Computer-mediated communication and group decision making: A meta-analysis.
Organizational Behavior and Human Decision Processes, 87, 156-179.
Blickensderfer, E. L., Cannon-Bowers, J. A., & Salas, E. (1998). Cross-training and team
performance. In J. A. Cannon-Bowers & E. Salas (Eds.), Making decisions under stress:
Implications for individual and team training (pp. 299-311). Washington, DC: American
Psychological Association.
Bolstad, C. A., & Endsley, M. R. (1999). Shared mental models and shared displays: An
empirical evaluation of team performance. In Proceedings of the Human Factors and
Ergonomics Society 43rd Annual Meeting (pp. 213-217). Santa Monica, CA: Human
Factors and Ergonomics Society.
Boyle, E. A., Anderson, A. H., & Newlands, A. (1994). The effects of visibility on dialogue
and performance in a cooperative problem solving task. Language and Speech, 37, 1-20.
Briggs, G. E., & Johnston, W. A. (1967). Team training (Tech. Rep. No. 1327-4). Orlando,
FL: Naval Training Device Center.
Cascio, W. F. (2000). Managing a virtual workplace. The Academy of Management Executive,
14, 81-90.
Clark, H. H., & Brennan, S. E. (1991). Grounding in communication. In L. Resnick, J. M. Levine,
& S. D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 127-149). Washington,
DC: American Psychological Association.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

378 Small Group Research
Clark. H. H., & Carlson, T. (1982). Hearers and speech acts. Language, 58, 332-373.
Connaughton, S. L., & Daly, J. A. (2004). Leading from afar: Strategies for effectively leading
virtual teams. In S. H. Godar & S. P. Ferris (Eds.), Virtual and collaborative teams:
Process, technologies and practice (pp. 49-75). Hershey, PA: Idea Group.
Dennis, A. R. (1996). Information exchange and use in small group decision making. Small
Group Research, 27, 532-550.
Dietz-Uhler, B., & Bishop-Clark, C. (2001). The use of computer-mediated communication to
enhance subsequent face-to-face discussions. Computers in Human Behavior, 17, 269-283.
Doherty-Sneddon, G. D., O’Malley, C., Garrod, S., Anderson, A., Langton, S., & Bruce, V.
(1997). Face-to-face and video-mediated communication: A comparison of dialogue struc-
ture and task performance. Journal of Experimental Psychology: Applied, 3, 105-125.
Dourish, P., Adler, A., Bellotti, V., & Henderson, A. (1996). Your place or mine? Learning
from long-term use of audio-video communication. Computer Supported Cooperative
Work, 5, 33-62.
Driskell, J. E., Radtke, P. H., & Salas, E. (2003). Virtual teams: Effects of technological media-
tion on team performance. Group Dynamics: Theory, Research, and Practice, 7, 297-323.
Entin, E. E., & Serfaty, D. (1999). Adaptive team coordination. Human Factors, 41, 312-325.
Ferrara, K., Brunner, H., & Whittemore, J. (1991). Interactive written discourse as an emergent
register. Written Communication, 8, 8-34.
Friedman, R. A., & Currall, S. C. (2003). Conflict escalation: Dispute exacerbating elements
of e-mail communication. Human Relations, 56, 1325-1348.
Fussell, S. R., & Benimoff, I. (1995). Social and cognitive processes in interpersonal com-
munication: Implications for advanced telecommunications technologies. Human Factors,
37, 228-250.
Gibson, C. B., & Cohen, S. G. (2004). Virtual team effectiveness. General Management Review.
Retrieved February 20, 2006, from http://www.etgmr.com/GMRjan-mar04/art1.html
Hancock, J. T., & Dunham, P. J. (2001). Language use in computer-mediated communication:
The role of coordination devices. Discourse Processes, 31, 91-110.
Hertel, G., Geister, S., & Konradt, U. (2005). Managing virtual teams: A review of current
empirical research. Human Resource Management Review, 15, 69-95.
Hollingshead, A. B. (1998). Retrieval processes in transactive memory systems. Journal of
Personality and Social Psychology, 74, 659-671.
Hollingshead, A. B. (2001). Cognitive interdependence and convergent expectations in trans-
active memory. Journal of Personality and Social Psychology, 81, 1080-1089.
Hollingshead, A. B., McGrath, J. E., & O’Connor, K. M. (1993). Group task performance and
communication technology: A longitudinal study of computer-mediated versus face-to-
face work groups. Small Group Research, 24, 307-333.
Horvath, L., & Tobin, T. J. (2001). Twenty-first century teamwork: Defining competencies for
virtual teams. In M. M. Beyerlein, D. A. Johnson, & S. T. Beyerlein (Eds.), Virtual teams
(Vol. 8, pp. 239-258). Amsterdam: JAI Press.
Jackson, P., & Klobas, J. (2008). Transactive memory systems in organizations: Implications
for knowledge directories. Decision Support Systems, 44, 409-424.
Kanki, B. G., Folk, V. G., & Irwin, C. M. (1991). Communication variations and aircrew
performance. International Journal of Aviation Psychology, 1, 149-162.
Kelly, J. R., Jackson, J. W., & Hutson-Comeaux, S. L. (1997). The effects of time pressure and
task differences on influences modes and accuracy in decision-making groups. Personality
and Social Psychology Bulletin, 23, 10-22.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 379
Kiesler, S., & Cummings, J. N. (2002). What do we know about proximity and distance in
work groups? A legacy of research. In P. J. Hinds & S. Kiesler (Eds.), Distributed work
(pp. 57-80). Cambridge: MIT Press.
Knoll, K., & Jarvenpaa, S. L. (1998). Working together in global virtual teams. In M. Igbaria
& M. Tan (Eds.), The virtual workplace (pp. 2-23). Hershey, PA: Idea Group.
Kozlowski, S. W. J. (1998). Developing adaptive teams. In J. A. Cannon-Bowers & E. Salas
(Eds.), Making decisions under stress: Implications for individual and team training
(pp. 115-153). Washington, DC: American Psychological Association.
Kraut, R. E., Fussell, S. R., Brennan, S. E., & Siegel, J. (2002). Understanding effects of prox-
imity on collaboration: Implications for technologies to support remote collaborative work.
In P. J. Hinds & S. Kiesler (Eds.), Distributed work (pp.137-162). Cambridge: MIT Press.
LePine, J. A. (2005). Adaptation of teams in response to unforeseen change: Effects of goal
difficulty and team composition in terms of cognitive ability and goal orientation. Journal
of Applied Psychology, 90, 1153-1167.
McGrath, J. E. (1984). Groups: Interaction and performance. Englewood Cliffs, NJ: Prentice Hall.
McGrath, J. E. (1991). Time, interaction, and performance (TIP): A theory of groups. Small Group
Research, 22, 147-174.
McGrath, J. E. (1993). Introduction. The JEMCO workshop: Description of a longitudinal
study. Small Group Research, 24, 285-306.
McGrath, J. E., Arrow, H., Gruenfeld, D. H., Hollingshead, A. B., & O’Connor, K. M. (1993).
Groups, tasks and technology: The effects of experience and change [Special issue]. Small
Group Research, 24, 406-420.
McIntyre, R. M., & Dickinson, T. L. (1997). A conceptual framework for teamwork measurement.
In M. T. Brannick, E. Salas., & C. Prince (Eds.), Team performance assessment and measure-
ment: Theory, methods, and applications (pp. 19-45). Mahwah, NJ: Lawrence Erlbaum.
McIntyre, R. M., & Salas, E. (1995). Measuring and managing for team performance:
Emerging principles for complex environments. In R. A. Guzzo & E. Salas (Eds.), Team effec-
tiveness and decision-making in organizations (pp. 9-45). San Francisco: Jossey-Bass.
Moreland, R. L., Argote, L., & Krishnan, R. (1996). Socially shared cognition at work:
Transactive memory and group performance. In J. L. Nye & A. M. Brower (Eds.), What’s
social about social cognition? (pp. 57-84). Thousand Oaks, CA: Sage.
Mühlbach, L., Böcker, M., & Prussog, A. (1995). Telepresence in video communications:
A study on stereoscopy and individual eye contact. Human Factors, 37, 290-305.
Nardi, B. A., & Whittaker, S. (2002). The place of face-to-face communication in distributed work.
In P. J. Hinds & S. Kiesler (Eds.), Distributed work (pp. 83-112). Cambridge: MIT Press.
O’Conaill, B., & Whittaker, S. (1997). Characterizing, predicting, and measuring video-
mediated communication: A conversational approach. In K. Finn, A. Sellen, & S. Wilbur
(Eds.), Video-mediated communication (pp. 107-131). Mahwah, NJ: Lawrence Erlbaum.
Olson, G. M., & Olson, J. S. (2000). Distance matters. Human-Computer Interaction, 15,
139-178.
Poole, M. S., & DeSanctis, G. (1990). Understanding the use of decision support systems: The
theory of adaptive structuration. In J. Fulk & C. Steinfield (Eds.), Organizations and
communication technology (pp. 175-195). Newbury Park, CA: Sage.
Qureshi, S., & Vogel, D. (2001). Adaptiveness in virtual teams: Organizational challenges and
research directions. Group Decision and Negotiation, 10, 27-46.
Rasker, P. C. (2002). Communication and performance in teams. Unpublished doctoral
dissertation, University of Amsterdam.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

380 Small Group Research
Rochlin, G. I., LaPorte, T. R., & Roberts, K. H. (1987). The self-designing high reliability
organization: Aircraft carrier flight operators at sea. Naval War College Review, Autumn,
76-90. Retrieved January 12, 2006, from http://www.nwc.navy.mil/press/review/1998/
summer/art7su98.htm
Salas, E., Burke, C. S., & Samman, S. N. (2001). Understanding command and control teams oper-
ating in complex environments. Information Knowledge Systems Management, 2, 311-323.
Seetharaman, P. S., Samarah, I., & Mykytyn, P. P. (2004). Impact of heterogeneity and
collaborative conflict management style on the performance of synchronous global virtual
teams. Information and Management, 41, 303-321.
Seifert, C. M., & Hutchins, E. L. (1992). Error as opportunity: Learning in a cooperative task.
Human-Computer Interaction, 7, 409-435.
Sellen, A. J. (1995). Remote conversations: The effects of mediating talk with technology.
Human-Computer Interaction, 10, 401-444.
Serfaty, D, Entin E. E., & Johnston, J. H. (1998). Team coordination training. In J. A. Cannon-
Bowers & E. Salas (Eds.), Making decisions under stress: Implications for individual and
team training (pp. 221-270). Washington, DC: American Psychological Association.
Serfaty, D., Entin, E. E., & Volpe, C. (1993). Adaptation to stress in team decision-making and
coordination. In Proceedings of the Human Factors and Ergonomics Society 37th Annual
Meeting (pp. 1228-1232). Santa Monica, CA: Human Factors and Ergonomics Society.
Shin, Y. (2005). Conflict resolution in virtual teams. Organizational Dynamics, 34, 331-345.
Smith-Jentsch, K. A., Johnston, J. A., & Payne, S. C. (1998). Measuring team-related expertise
in complex environments. In J. A. Cannon-Bowers & E. Salas (Eds.), Making decisions
under stress: Implications for individual and team training (pp. 61-88). Washington,
DC: American Psychological Association.
Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organiza-
tional communication. Management Science, 32, 1492-1512.
Straus, S. G. (1997). Technology, group process, and group outcomes: Testing the connections
in computer-mediated and face-to-face groups. Human-Computer Interaction, 12, 227-266.
Straus, S. G., & McGrath, J. E. (1994). Does the medium matter: The interaction of task and
technology on group performance and member reactions. Journal of Applied Psychology,
79, 87-97.
TANDBERG. (2007, January 22). Face to face with management reality: A telework research
report. Alexandria, VA. Retrieved January 25, 2007, from http://www.teleworkexchange.com/
managementstudy/Face-to-Face-with-Management-Reality-Study-012207.pdf
Thompson, L. F., & Coovert, M. D. (2006). Understanding and developing virtual computer-
supported cooperative work teams. In C. Bowers, E. Salas, & F. Jentsch (Eds.), Creating
high-tech teams: Practical guidance on work performance and technology (pp. 213-242).
Washington, DC: American Psychological Association.
Van der Kleij, R. (2007). Overcoming distance in virtual teams: Effects of communication
media, experience, and time pressure on distributed teamwork. Doctoral dissertation, University
of Amsterdam. Retrieved March 3, 2009, from http://dare.uva.nl/record/234997
Van der Kleij, R., Paashuis, R. M., Langefeld, J. J., & Schraagen, J. M. C. (2004). Effects of
long-term use of video-communication technologies on the conversational process.
International Journal of Cognition, Technology and Work, 6, 57-59.
Van der Kleij, R., Paashuis, R. M., & Schraagen, J. M. C. (2005). On the passage of time:
Temporal differences in video-mediated and face-to-face interaction. International Journal
of Human-Computer Studies, 62, 521-542.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from

Van der Kleij et al. / How Conversations Change 381
Walther, J. B. (2002). Time effects in computer-mediated communication: Past, present, and
future. In P. J. Hinds & S. Kiesler (Eds.), Distributed work (pp. 235-258). Cambridge: MIT
Press.
Wegner, D. M. (1987). Transactive memory: A contemporary analysis of the group mind. In
B. Mullen & G. R. Goethals (Eds.), Theories of group behavior (pp. 185–208). New York:
Springer-Verlag.
Williams, E. (1977). Experimental comparisons of face-to-face and mediated communication:
A review. Psychological Bulletin, 84, 963-976.
Wilson, J. M., Straus, S. G., & McEvily, B. (2006). All in due time: The development of trust
in computer-mediated and face-to-face teams. Organizational Behavior and Human
Decision Processes, 99, 16-33.
Zheng, J., Bos, N., Olson, J. S., & Olson, G. M. (2001). Trust without touch: Jumpstart trust
with social chat. Proceedings of the ACM Conference on Human Factors and Computer
Systems (pp. 293-294). New York: ACM Press.
Zheng, J., Veinott, E., Bos, N., Olson, J. S., & Olson, G. M. (2002). Trust without touch:
Jumpstart long-distance trust with initial social activities. In Proceedings of the ACM
Conference on Human Factors and Computer Systems (pp. 141-146). New York: ACM
Press.
Rick van der Kleij is a research psychologist at the Human Factors lab of TNO Defence,
Security and Safety in Soesterberg, the Netherlands. His main area of interest is on making
effective teams with people and collaborative technologies in collocated and distributed set-
tings. He holds a PhD in psychology from the University of Amsterdam, the Netherlands.
Jan Maarten Schraagen is a senior research scientist at the Human Factors lab of TNO
Defence, Security and Safety and a professor of applied cognitive psychology at Twente
University, the Netherlands. His research interests include task analysis, team decision mak-
ing, trust in collaborative repositories, and adaptive human–computer collaboration.
Peter Werkhoven is a professor at the Department of Information and Computing Sciences
at Utrecht University, the Netherlands. His work is focused on multimodal perception and
interaction in virtual environments. He is also Director Research, of TNO Defence, Security
and Safety, managing a research portfolio in the field of observation and information systems,
operational analysis and human factors.
Carsten K. W. De Dreu is Professor and Chair of the Organizational Psychology Program at
the University of Amsterdam, the Netherlands. He was president of the International
Association for Conflict Management and the European Association of Social Psychology,
and served as scientific director of the Kurt Lewin Graduate School for social and organiza-
tional psychology. His research interests center on group processes and performance including
conflict and negotiation, and creativity and innovation.
by Books Editorial on December 31, 2009 http://sgr.sagepub.comDownloaded from