On the passage of time: Temporal differences in video-mediated and face-to-face interaction

ARTICLE IN PRESS

Int. J. Human-Computer Studies 62 (2005) 521–542

1071-5819/$ -

doi:10.1016/j

�Correspo

E-mail ad

www.elsevier.com/locate/ijhcs

On the passage of time: Temporal differences invideo-mediated and face-to-face interaction

Rick van der Kleij�, Roos Paashuis, Jan Maarten Schraagen

TNO Human Factors, Department of Team Solutions, Kampweg 5, P.O. Box 23, 3769 ZG Soesterberg,

The Netherlands

Received 26 September 2003; received in revised form 13 September 2004; accepted 25 January 2005

Communicated by D. Damian

Abstract

This paper examines team work over time in video-mediated non-collocated and traditional

face-to-face same-room teams. In a longitudinal between-groups design, 22 three-person teams

were tested in 4 1-h test sessions at 2-week intervals. A paper-folding task was designed for the

experiment that had the potential to induce differences in team work under different

communication environments. Results showed that near the end of the experiment, initial

differences between groups on cooperative task performance in favour of the face-to-face

teams had disappeared. These findings are explained in terms of a technological adaptation

effect, which occurs when people learn how to use the technological tools available despite

technological limitations. No differences were found between groups on cohesion and

satisfaction. Likewise, cohesion did not increase over time as participants gained experience.

In comparison, satisfaction did increase over time for both groups as teams gained experience

with fellow team members and the task at hand. In contrast to what was expected, mental

effort was higher for face-to-face teams than for video-mediated teams. Furthermore, initial

differences between groups on mental effort did not disappear as participants gained

experience. This paper concludes with a discussion of the results in terms of their implications

see front matter r 2005 Elsevier Ltd. All rights reserved.

.ijhcs.2005.01.003

nding author. Tel.: +31346 356 211; fax: +31 346 353 977.

dresses: [email protected] (R. van der Kleij), [email protected] (J.M. Schraagen).

www.elsevier.com/locater/ijhcs

ARTICLE IN PRESS

R. van der Kleij et al. / Int. J. Human-Computer Studies 62 (2005) 521–542522

for natural work teams, the design of video-communication technologies and suggestions for

future research.

r 2005 Elsevier Ltd. All rights reserved.

Keywords: Technological adaptation; Video-mediated communication; Team work; Longitudinal design;

Telecommunication; Group development; Temporal differences

1. Introduction

The integration of computers and communication technologies has made newforms of team work possible, other than traditional face-to-face meetings. Electroniccommunication technologies enable collaboration without regard to temporal andspatial limitations. These technologies make collaboration in dispersed teams moreconvenient, less expensive and less time consuming than travelling to meet face-to-face (Hollingshead and McGrath, 1995).

While electronic communication technologies, such as electronic mail and internetchat, have proven their worth to organizations, video-communication technologiesstill appear to offer little more than traditional technologies such as the telephone(Whittaker, 2003). However, a number of recent technological advancements, suchas an increase in bandwidth, combined with reduced costs, have made video-communication technologies more attractive as means of supporting dispersed workteams. Nonetheless, to be able to develop and market video-communicationtechnologies, one must understand the effects that such products have on their users.

In this paper, we examine the effects of long-term use of a high-realism video-communication system on cooperative task performance, cohesion, membersatisfaction and mental effort. This research is primarily concerned with teams,although we will sometimes draw from the more extensive literature on small groupswhen relevant for the purpose of this research.

Most definitions of small groups include three attributes: (1) two or moreindividuals, (2) interaction among group members and (3) interdependence amongthem in some way (McGrath, 1984). To these three, McGrath adds that a smallgroup exists in some time frame; that is, it has a past, a present and a future. Both thepast and anticipation of the future will influence the present behaviours of small-group members.

When compared to small groups, teams share the aforementioned characteristicswith small groups, with some additional characteristics: teams have explicit goalsand, with a few exceptions, the members of a team have some level of awareness ofthe team goals (Ilgen et al., 1995). In addition, in teams, each member is assigned aspecific role or function to perform (Dyer, 1984). In other words, teams exist forsome task-oriented purpose and can be seen as a more resolute version of smallgroups.

The remainder of this introduction explores prior research on the effects ofcommunication media on team work, examines how temporal aspects shape

ARTICLE IN PRESS

R. van der Kleij et al. / Int. J. Human-Computer Studies 62 (2005) 521–542 523

interaction in teams and describes the research question addressed in the presentstudy.

1.1. Information richness and telepresence

Communication media differ in terms of the richness of the information that canbe transmitted (Daft and Lengel, 1984, 1986). Information richness refers to thedegree to which the information contains emotional, attitudinal, normative andother meanings beyond the literal cognitive denotations of the symbols to express it(McGrath and Hollingshead, 1993). Face-to-face communication is the richest formof information processing because it provides immediate feedback. With feedback,understanding can be checked and interpretations corrected. In addition, face-to-face conversations allow for the simultaneous observations of multiple cues,including body language, facial expression and tone of voice, which put acrossinformation beyond the spoken message. Audio conversations through telephone forexample, are less rich than face-to-face. Feedback capacity is fast, but visual cues areabsent; group members must rely on language content and audio cues to reach anunderstanding.

Daft and Lengel (1984, 1986) argue that task performance will be improved whentask needs, which differ in terms of the degree to which effective performance onthem requires the transmission of information that is more or less rich in its contents,are matched to the communication medium’s richness. For routine tasks, which arewell understood, media of lower richness, such as internet chat or e-mail, wouldprovide sufficient information. Emotional connotations about message and sourceare not required for routine tasks and are often considered to be a hindrance. Mediacapable of transmitting rich information, such as video-communication technolo-gies, are better suited to supporting equivocal tasks, where there are multipleinterpretations for available information.

In the same manner, the communication medium’s richness is used to refer to theapparent distance or ‘realness’ of the communicators at the remote site (Short et al.,1976). A similar but some-what broader construct is telepresence, which is used torefer to the impression of sharing space with the participants at the remote site(Bocker and Muhlbach, 1993; Muhlbach et al., 1995; Suwita et al., 1997; Draper etal., 1998).

Telepresence has received much attention up till now because most designers oftechnologies intended to mediate communication, consider a strong sense oftelepresence to be a desirable attribute (Welch, 1999). Herein lies the assumption thatthe more telepresence these technologies bring forth, the better the users’performance on the task for which the device is being used. Indeed, some evidencefor this assumption comes from Venkatesh and Johnson (2002); who found strongsupport for their hypothesis that higher telepresence leads to higher telecommutermotivation and higher sustained use of the system. In comparison, communicationmedia that are not capable of transmitting rich information, increase the group’s taskfocus at the expense of social interaction (Carletta et al., 2000). This could have animpact on satisfaction and the development of member relations, such as cohesion.

ARTICLE IN PRESS


For example, Straus (1997) reported a positive effect of social interaction on groupcohesion, as shown in the amount of non-task communication. Moreover, she foundthat computer-mediated groups developed lower cohesion than did face-to-facegroups.

As for satisfaction, Strauss found a similar effect such that computer-mediatedgroups were less satisfied than were face-to-face groups. This result is consistent withBaltes et al. (2002). In a meta-analysis Baltes et al. found that computer-mediatedgroups were less satisfied than their face-to-face counterparts. It appears that therelative lack of social context cues in computer-mediated settings leads to frustrationand dissatisfaction. Muhlbach et al. (1995) found similar results. In their experiment,a positive relationship was found between telepresence and satisfaction. The greaterthe feeling of being in direct contact with the participants at the remote site, the moresatisfaction with video-conferencing. According to their line of reasoning thecommunication setting creating the richest presence experience should lead to thehighest degree of satisfaction. This seems to underline the importance of mediatingthe sense of presence in the development of innovative communication technologiesoriented to create a perceptual immersion in remote or virtual environments.

1.2. Experience with team, task and communication medium

Empirical data describing media effects on temporal group processes are rare.Most studies done in this area can be described as single-session experiments withgroups that share no prior experience and expect no future interaction (i.e., ad hocgroups). There are some serious limitations to this approach, especially regarding thedegree to which it reflects the temporal processes of naturally occurring groups as weencounter them in our everyday life (see McGrath, 1991 for a comprehensivecritique). For instance, ad hoc teams bear no resemblance at all to many of thegroups we meet in everyday living which have numerous past experiences with eachother and anticipate a future with fellow group members (Alge et al., 2003).

Reasons for this failure to deal with temporal issues in past group research can beattributed to methodological and practical difficulties. Empirical research of groupsover time is difficult and requires considerable financial resources to carry out(McGrath and Hollingshead, 1993; McGrath et al., 1993). In addition, longitudinalstudies using experimentally created groups are contingent on having regularattendance by the same people for a long period of time. Problem are even moreacute for longitudinal studies using natural groups. Natural groups are prone toconsiderable fluctuation in membership and attendance over time, making it evenmore difficult to examine the effects of communication media on temporal groupprocesses.

The few studies that have examined the manifestations of temporal differences inface-to-face and mediated groups suggest that temporal aspects can have a largeimpact on group task performance, user reactions, and member relations (Hollings-head et al., 1993; McGrath and Hollingshead, 1993; McGrath et al., 1993; van derVelden, 1995; Alge et al., 2003; Wheelan, et al., 2003; van der Kleij et al., 2004).Hollingshead and co-workers (1993) found an adaptation effect in groups that were

ARTICLE IN PRESS


working together at remote sites.1 Hollingshead and her co-workers comparedgroups working face-to-face in the same room with groups working together usingan electronic message board on various tasks and under various sets of operatingconditions. The groups met in workshops once a week over a period of 14 weeks. Atthe beginning of the work group meetings, group task performance and satisfactionfor the computer-mediated groups was lower than for the traditional face-to-facegroups. The difference between groups on satisfaction with group performance didnot lesson over time. However, within a period of 3 weeks there were no significantdifferences in task performance between both communication environments. Thisresult suggests that the newness of the medium led to poorer task performance asopposed to the inherent information richness.

Chidambaram et al. (1991) found similar adaptation effects for computer-supported groups working on decision-making tasks with no right answer such asdescribed by McGrath’s (1984) task circumplex. Chidambaram and co-workersexamined the behaviour of face-to-face same-room groups using a group decisionsupport system across four sessions. The computer-supported groups used threeseparate group decision support tools to generate, review and evaluate decisionalternatives. Manual groups (i.e., groups that were not computer supported) usedflip charts for recording and reviewing alternatives, and for voting. Their resultsshowed significant differences over time between computer-supported groups andgroups that were not supported, in terms of conflict management and the degree ofcohesion. After adapting to the group decision support system, computer-supportedgroups displayed more productive conflict management (i.e., the ability to manageconflict as measured on a post-session questionnaire) and higher group cohesionthan groups that were not supported.

More recently, van der Kleij et al. (2004) studied the effects of long-term use ofvideo-mediated technologies on the conversational process. In a longitudinalbetween-groups experiment, three-person teams had to work on a rather simple‘intellective’ task (McGrath, 1984) in four test sessions at 2-week intervals. In eachsession, teams had to select and argue the correct answer out of three alternatives fora total of ten questions. Results showed that, compared to face-to-face teams, video-mediated teams took more time for turns, required fewer turns to complete the taskand interrupted each other less. At the end of the fourth test session, initialdifferences between groups on all these measures had disappeared, suggesting anadaptation effect. In addition, it was found that over time teams needed lessconversation to complete the task. For both groups, the overall number of turnsdecreased as teams gained experience, and the overall duration of pauses increased.Team members appeared to develop shared knowledge about the task, roles,responsibilities and informational needs of their team members. As a consequence,team members were able to anticipate each other’s information needs and providerelevant information before being asked to do so.

1Participants were in fact in the same room, but computer terminals either faced a wall or a partition

separating the room. Participants communicated to other group members through the computer network

and were not allowed to talk during the experiment.

ARTICLE IN PRESS


Wheelan et al. (2003) investigated the relationship between length of time thatnatural work groups had been meeting at their work sites and the verbal behaviourpatterns and perceptions of group members about their groups. Significantrelationships were found between the length of time that groups had been meetingand the perceptions of group members. More specific, members of older groupsperceived their groups to be more productive. In addition, the longer that a grouphad been meeting, the more verbal statements were made that represented goal-directed activities and task-oriented efforts.

In a lab experiment, Alge et al. (2003) found that media differences existed on bothbrainstorming and negotiation tasks for teams lacking past experiences with eachother; with face-to-face teams exhibiting higher perceived communication openness/trust, and explicit information sharing (i.e., the number of unique pieces of sharedtask information) than computer-mediated teams. However, computer-mediatedteams that did have numerous past experiences with each other (e.g., project teamsnearing the completion of a project) were able to eliminate these differences.

In line with the results described above, we predict that an increase in experiencewith video-communication technologies will have a positive effect on all aspects ofgroup work. We expect that in time groups learn how to use the technological toolsavailable despite technological limitations, such as restricted bandwidth or lowvideo-image resolution, a situation that is described as technological adaptation(Qureshi and Vogel, 2001; see also Dourish et al., 1996; Olson and Olson, 2000; andvan der Kleij et al., 2004). In sum, we hypothesize that, compared to video-mediatedcommunication, face-to-face communication will lead to better task performance,higher cohesion, more satisfaction and lower mental effort. We expect, however, thatat the end of the experiment these differences between face-to-face and video-mediated groups have disappeared due to technological adaptation.

2. Method

2.1. Participants

A total of 66 people participated in the experiment that was carried out in theHuman Factors lab of The Netherlands Organization for Applied ScientificResearch (TNO). Most participants were students and their age ranged from 18 to32 years. Of those participants 36 were female and 30 were male. We tested a total of22 teams four times, 11 teams in the face-to-face (control) group and 11 teams in thevideo-mediated (experimental) group. In the face-to-face group, 6 teams consisted of2 males and 1 female and 5 teams consisted of 2 females and 1 male. In the video-mediated group 4 teams of 2 males and 1 female and 5 teams of 2 females and 1 maleparticipated. In the video-mediated group there were 2 teams that consisted of 3females.

Participants in both groups were matched on two characteristics to equate groups.First we made sure that none of the team members knew each other prior to theexperiment and second we tried to maximize the number of mixed gender teams.

ARTICLE IN PRESS


Random allocation to experimental groups took place on all other participantcharacteristics to ensure the formation of equivalent groups.

For taking part in the experiment, each participant was paid EUR 40 per testsession. The total amount for participating in all test sessions was EUR 160, whichwas paid to the participants after the fourth test session.

2.2. Equipment

2.2.1. Face-to-face setting

Participants in the face-to-face teams were placed in the same room around around table. The table had a diameter of 180 cm. A video camera was set up in oneupper corner of the room for recording purposes. The video output was channelledthrough coaxial cable to a Panasonic AG 8700 VHS video recorder. During taskperformance participants wore Shure unidirectional microphone headsets; modelSM10A. The audio output from these headsets was also fed through coaxial cable tothe experimenter room. In the experimenter room, the audio output from separatetracks went to an audio mixer where all three voices were laid down on the audiotrack of the videocassette. The face-to-face setting is depicted in Fig. 1.

2.2.2. Video-mediated setting

Participants in the video-mediated teams were located in different rooms. Eachmember was seated in a room at a table which was shaped as one half of an oval. Thetable was 180 cm wide. In front of the participants was a large rear projection screenof 146 cm� 86 cm. This screen was contained in an aluminium frame, which added3 cm to the total size of the screen in each direction. The video-mediated setting isdepicted in Fig. 2.

Fig. 1. Participants in the face-to-face teams were placed in the same room around a round table. The

teams members each received unique task instructions. Collaborating with both other teams members was

required for each participant to complete the task.

ARTICLE IN PRESS

Fig. 2. Participants in the video-mediated teams were located in different rooms. All participants saw the

same split screen which allowed them to communicate with both other participants. Speech was amplified

and transmitted to the other two rooms so the participants could communicate with each other.


Behind each participant in the video-mediated teams, blue partitions were placedto create a uniform background. All participants in these teams wore Shureunidirectional microphone headsets; model SM10A, that recorded the audiocommunication. In the video-mediated setting, in contrast to the face-to-facesetting, the sound was amplified and transmitted to the other two rooms so theparticipants could communicate with each other. Two Panasonic WV-gp150Ecameras filmed the participants in the video-mediated teams. Two cameras weresituated on both sides of the screen and were used for the transmission of images tothe other rooms. The cameras were positioned on the sides of the screen to create anisotropic condition enabling selective gaze (i.e., gaze direction could be determinedby the position of the head of the speaker). Because the cameras were positioned atthe sides of the screen no eye contact, however, could be established betweenparticipants. The images recorded by the video cameras of both other participantswere projected on the rear projection screen by two Hitachi XGA cp x970 projectors.Fig. 3 depicts a schematic representation of the video-mediated setting.

A third camera was located on top of the screen and was used to monitor progresson the experimental task. In both conditions, a Panasonic VN 484p intercom systemwas used to inform the participants when to start and stop working on the task.

2.3. Task

For the experiment a team task was required in which a rich communicationenvironment was essential to performing the task well. To our knowledge, noexisting task that met this criterion was available at the time the study wasconducted. Consequently it was necessary to develop a new task specifically for thepresent study. The result is a complex object-centric team task in which the goal wasfor each team member to fold as quickly and precisely as possible two Japanese

ARTICLE IN PRESS

Fig. 3. Schematic representation of the experimental set-up (adapted from Okada et al., 1994).


Origami figures. On these properties our task bears a resemblance to Moreland’sradio-assembly task, in which group members work together to assembly a radiowith materials from a radio kit, that he used to investigate the role of sharedknowledge about the skills and knowledge of each group member on taskperformance (Moreland, 1999). Our task is different from Moreland’s task in thatteam members each received unique instructions about the same figures, which madethem interdependent for the completion of the task. One participant received thewritten instructions; one team member had the pictures that go with the writteninstructions; and one team member held the basic Origami folding symbols andtechniques. Collaborating with both other team members was required for eachparticipant to complete the two figures. Eight figures were selected and divided atrandom over the four versions of the task. The order of versions was randomizedover test sessions for each team. Furthermore, participants changed role, place at the(virtual) table and instructions in consecutive test sessions. The teams had 30 min toperform the task.

Additional instructions were given to the participants in the face-to-face teams.Those teams were instructed not to touch the work and instructions of the otherteam members. This instruction was given to make sure that the task was performedunder similar conditions as in the video-mediated group.

2.4. Experimental design

2.4.1. Independent variables

In a longitudinal between-groups design, 22 three-person teams were tested in 4 1-h test sessions at 2-week intervals. In the face-to-face group we started out with 13teams and in the video-mediated group there were 16 teams at test session one. At

ARTICLE IN PRESS


the end of test session four, 11 teams remained in both groups. The results describedin this paper are based on these 22 teams.

The experimental video-mediated group differed from the face-to-face group onfour aspects. First, although an isotropic condition was created enabling selectivegaze (i.e., gaze direction could be determined by the position of the head of thespeaker), no eye contact could be established between interlocutors. It is a knownfact that this could result in difficulties with turn-taking behaviour and securingother people’s attention (O’Conaill et al., 1993; Sellen, 1995; Whittaker andO’Conaill, 1997). Second, participants in the video-mediated teams did not share thesame physical space while teleconferencing. This means that, for example, speakershave no sense of how the other interlocutors perceive their voices (Sellen, 1995).Third, the image resolution of the participants perceived in the video-mediatedgroups is much lower compared to the face-to-face situation. The number of pixelswith which a interlocutor can be represented in the video-mediated setting is finite.This makes it perhaps more difficult to perceive details and to recognize facialexpressions. Fourth, participants in the video-mediated situation perceive otherparticipants and objects as two-dimensional images. This could have animpact on tasks that require manipulation of objects in space, such as ourOrigami task. To maintain as much control as possible, audio distortions,such as noise, reverberation and echoes, were kept to a minimum. Further, becausethere was no processing required for coding and decoding of the audio and videosignals, no unwanted round-trip delays and delays between audio and video wereintroduced.

2.4.2. Dependent variables

Two categories of dependent variables were measured to adequately address theresearch question, as described in Section 1.2: (1) objective measures of the outcomes(i.e., task performance); and, (2) individual perceptions of member relations and theprocess through which the team produced their outcomes (i.e., questionnaires oncohesion, member satisfaction and a rating scale on mental effort). Individualperceptions of the process are of importance because these measures offer additionalinsights to explain differences in outcomes, and means for assessing the effect ofcommunication medium and experience. For example, people are found to adapt toincreasing task demands by exerting additional effort to maintain an acceptable levelof performance. The level of task performance, therefore, only providesvaluable information when at the same time techniques are used to index theinvested effort. Both categories of dependent variables are discussed in the followingsubsections.

2.4.2.1. Task performance. Scores on the Origami task were made dependent ontwo dimensions, namely: completeness and precision. A maximum of 20 points couldbe obtained for each figure. Ten points for the completeness and 10 points for theprecision. The maximum score is thus 40 points for this task per person per sessionand 120 points for a team. The score for completeness was determined on a 10-pointscale by dividing 10 by the total number of steps needed to complete the figure, times

ARTICLE IN PRESS


the steps that were actually completed. The precision was determined on the basis ofthe neatness of the folded figures. For both scores the level of agreement betweentwo judges was calculated. It was found that judges reached a good level ofagreement on both scores (93 per cent for completeness and 95 per cent forprecision). Differences in task performance between both groups as well asdifferences in performance over time were tested.

2.4.2.2. Subjective measures. Questionnaires on cohesion, member satisfaction anda rating scale on mental effort were administered at the end of each test session aftercompletion of the task. All questionnaires and the rating scale were pre-tested andfound to be reliable, simple to administer and to take little time for participants tocomplete. The questionnaires and the rating scale are discussed in more detail in thefollowing paragraphs.

Cohesion: Cohesion can be described as the extent to which the group members areattracted to the group and each other (Chidambaram et al., 1991; Hogg, 1992).Groups characterized by greater interpersonal attraction communicate better andmore, and are less inhibited (Brown, 2000). Cohesion also appears to have a modestpositive effect on task performance. For instance, Mullen and Copper (1994)observed in a meta-analysis of 49 studies a positive correlation between cohesion andperformance (averaging around 0.25).

In the present study, cohesion was measured by means of a post-sessionquestionnaire (i.e., the questionnaire was given to the participants once at the end ofeach test session) on which ratings could be given on 7-point Likert scales, in which ascore of 1 corresponds to the most negative option and a score of 7 corresponds tothe most positive option (10 items, Cronbach’s alpha ¼ :96). The questionnaire wasan adapted version from the cohesion questionnaire from Dennis (1996). Thequestionnaire was translated into Dutch and extended to relate more to our settingand participants.

Member satisfaction: A satisfaction questionnaire was given to the participantsafter completion of the task. The questionnaire contains five questions onwhich ratings could be given on 7-point Likert scales in which a score of 1corresponds to the most negative option and a score of 7 corresponds to the mostpositive option (5 items, Cronbach’s alpha ¼ :95). This questionnaire was anadapted version of the satisfaction questionnaire from Dennis (1996) and includesscales that refer both to satisfaction with the group outcome and the discussionprocess. Process satisfaction refers to contentment with the interactions that occurwhile team members are devising decisions. Outcome satisfaction encompassesapproval of the final team decision (Thompson and Coovert, 2003). Thequestionnaire was also translated into Dutch and extended to relate more to oursetting and participants.

Mental effort: Mental effort can be defined as the ratio between the task demandsand the capacity of the operator working on the task (O’Donnell and Eggemeier,1986). Mental workload is high when the difference between task demands andcapacity is small. It is interesting to note that workload cannot be regarded as a staticconcept. In an attempt to cope with the task demands, people adapt by changing

ARTICLE IN PRESS


their strategies, delaying low priority tasks and by lowering performancecriteria (Veltman and Gaillard, 1996). Techniques for measuring mentalworkload can be divided into performance and physiological measures, andsubjective ratings.

To evaluate mental effort the Rating Scale Mental Effort (RSME) wasadministered once per test session directly after completion of the task. The RSME(Zijlstra, 1993) is a one-dimensional scale with ratings between 0 and 150. The scalehas nine descriptive indicators along its axis (e.g., 12 corresponds to not effortful, 58to rather effortful and 113 to extremely effortful). It is designed so as to minimizeindividual differences. The RSME was selected because it is simple to administer, isnot intrusive and at the same time it provides a good indication of the total mentalworkload (Veltman and Gaillard, 1996).

Comparative evaluations of subjective scales conducted in a wide range of tasksshowed little advantage for the multidimensional scales (see Wierwille andEggemeier, 1993, for a review). For example, it was shown that subjective workloadassessment techniques gain little from complex assessment and analysis methodssuch as those used in the multidimensional Subjective Workload AssessmentTechnique (SWAT) (Verwey and Veltman, 1996) and in the NASA Task Load Index(TLX) (Veltman and Gaillard, 1993). A simple one-dimensional rating scale appearssufficient for assessing subjective workload. It appears that people find it moredifficult to rate mental effort on multidimensional scales, than on one-dimensionalscales such as the RSME. Veltman and Gaillard argue that people make an overallestimation of mental workload and find it difficult to make a clear distinctionbetween the different sources of workload.

2.5. Procedure and instructions

At the start of each test session participants received a general instruction aboutthe experiment. Participants were told that the experiment examined interaction inteams. After all participants had read the general instructions, they were given task-specific instructions. After clarifying the instructions and answering possiblequestions about the task, the experimenter left the room and told the teams tostart working through the intercom. When time was up, the experimenter entered theroom again and collected the task materials. The team members then hadto complete the RSME and the questionnaires on both satisfaction andcohesion.

The video-mediated team members were seated in separate rooms. To maintain asmuch control as possible, participants were not allowed to leave their rooms duringthe experiment to prevent them from getting to know the other team membersoutside the test situation. For the same reason, upon arrival, we picked theparticipants up one at a time. At the end of the test session we escorted theparticipants one at a time at intervals to the main exit of our lab facilities.Furthermore, we asked our participants not to discuss the experiment duringunforeseen encounters with members of their own and other teams. One test sessiontook about one hour in total for both groups.

ARTICLE IN PRESS


3. Results

An analysis of variance (ANOVA) for repeated-measures designs was carried outon all variables. It should be noted that at the end our experiment 22 three-personteams remained, 11 in each communication setting. All analyses are at the team leveland are based on these remaining 22 teams. This means that the analyses presentedin this paper have quite low statistical power to detect effects, a limitation inherent inteam research. An alpha level of .10 was therefore adopted for all subsequentanalyses.

3.1. Task performance

The analysis revealed a significant main effect of groups on task performance,F ð1; 20Þ ¼ 6:34; p ¼ :020: The face-to-face teams performed significantly better(M ¼ 26:22; S.D. ¼ 5.34) than the video-mediated teams (M ¼ 22:25; S.D. ¼ 5.84).Post-hoc analysis revealed that a significant difference between groups in test session1 caused this effect, F ð1; 20Þ ¼ 9:32; pp:01: In test session 1, face-to-face teamsperformed significantly better (M ¼ 24:21; S.D. ¼ 5.38) than video-mediated teams(M ¼ 16:48; S.D. ¼ 5.02). No significant differences between groups were found ontest sessions 2, 3 and 4. Moreover, the analysis revealed a significant main effect oftest sessions on task performance, F ð3; 60Þ ¼ 5:84; pp:01: Overall task performanceincreased as teams gained experience. Additional analysis revealed that this effectwas due to the video-mediated teams, F ð3; 30Þ ¼ 5:98; pp:01: Significant linear,F ð1; 20Þ ¼ 12:17; pp:01 and significant quadratic, F ð1; 20Þ ¼ 3:49; p ¼ :077; trendscharacterized this increase. Furthermore, an interaction was found between groupsand test sessions, F ð3; 60Þ ¼ 2:48; p ¼ :069: Thus, while face-to-face teams remainedconstant in their performance over time, video-mediated teams got better at adecreasing rate as they gained experience. In short, these results provide support forour hypotheses that, compared with video-mediated communication, face-to-facecommunication leads to better task performance and initial differences in taskperformance reduce over time (Fig. 4).

3.2. Cohesion

On the cohesion questionnaire, no significant main effect of groups was found,F ð1; 20Þ ¼ 2:64; ns. In other words, no significant differences were found on cohesionbetween face-to-face (M ¼ 5:68; S.D. ¼ .70) and video-mediated teams (M ¼ 5:35;S.D. ¼ .78). Furthermore, no significant main effect of test sessions was found oncohesion, F ð3; 60Þ ¼ :47; ns. Likewise, no interaction effect was found betweengroups and test sessions, F ð3; 60Þ ¼ 1:16; ns (Fig. 5).

3.3. Member satisfaction

No significant differences in satisfaction were found between face-to-face(M ¼ 5:09; S.D. ¼ .73) and video-mediated teams (M ¼ 4:78; S.D. ¼ .81), F ð1; 20Þ ¼

ARTICLE IN PRESS

Test session

Deg

ree

of c

ohes

ion

4

5

6

7

1 2 3 4

Face-to-face

Video mediated

Fig. 5. Mean scores on the cohesion questionnaire as a function of test session and communication

environment.

Test session

Gro

up ta

sk p

erfo

rman

ce

14

16

18

20

22

24

26

28

30

1 2 3 4

Video mediated

Face-to-face

Fig. 4. Task performance as a function of test session and communication environment.


1:47; ns. The analysis did reveal a significant main effect of test sessions on perceivedsatisfaction, F ð3; 60Þ ¼ 6:21; pp:01: This implies that satisfaction on this taskincreased as teams gained experience. Significant linear, F ð1; 20Þ ¼ 9:47; pp:01 andsignificant quadratic, F ð1; 20Þ ¼ 5:92; p ¼ :024; trends characterized this increase.Thus, satisfaction increased at a decreasing rate as teams gained experience withfellow team members and task. Additional examination of our data revealedsignificant correlations between task performance and satisfaction in all test sessions(correlations ranging between .25 and .69). No significant interaction was found,however, between both groups and the test sessions, F ð3; 60Þ ¼ :43; ns. Overall, theseresults were not consistent with our expectations. Although satisfaction did increaseover time, no evidence was found for initial differences between both groups (Fig. 6).

ARTICLE IN PRESS

Test session

Deg

ree

of s

atis

fact

ion

4

5

6

7

1 2 3 4

Face-to-face

Video mediated

Fig. 6. Mean scores on the satisfaction questionnaire as a function of test session and communication

environment.

Test session

Deg

ree

of m

enta

l effo

rt

40

45

50

55

60

65

1 2 3 4

Face-to-face

Video mediated

Fig. 7. Mean scores on the Rating Scale Mental Effort (RSME) as a function of test session and

communication environment.


3.4. Mental effort

The analysis revealed a significant effect of groups on the RSME, F ð1; 20Þ ¼ 3:88;p ¼ :072: That is, the face-to-face teams (M ¼ 56:58; S.D. ¼ 17.89) reported higherlevels of mental effort than did the video-mediated teams (M ¼ 45:09; S.D. ¼ 21.02).No significant main effect was found for test sessions, F ð3; 60Þ ¼ 2:15; ns. Further,no significant interaction effect was found for groups and test sessions, F ð3; 60Þ ¼:32; ns. These results were not consistent or even opposite to our expectations. Incontrast to what was expected, it was found that the amount of mental effort wasgreater for face-to-face than for video-mediated teams. In addition, the initialdifferences between both groups did not disappear, as we expected. Perceived mentaleffort was also not found to decline over time (Fig. 7, Table 1).

ARTICLE IN PRESS

Table 1

Summary of cell means and standard deviations

Dependent variables Communication media Test session Row

1 2 3 4

Task performancea Video-mediated 16.48 25.58 21.79 25.15 22.25

(5.02) (4.12) (8.50) (5.92) (5.89)

Face-to-face 24.21 26.24 27.82 26.61 26.22

(5.38) (6.71) (5.37) (3.88) (5.34)

Column 20.35 25.91 24.80 25.88 24.24

(6.43) (5.44) (7.59) (4.94) (6.10)

Satisfactionb Video-mediated 4.29 4.94 4.88 5.01 4.78

(.89) (.79) (.78) (.94) (.85)

Face-to-face 4.65 5.02 5.36 5.34 5.09

(.68) (.87) (.51) (.86) (.73)

Column 4.47 4.98 5.12 5.18 4.94

(.79) (.81) (.69) (.89) (.80)

Cohesionb Video-mediated 5.46 5.38 5.24 5.30 5.35

(.73) (.69) (.82) (.88) (.78)

Face-to-face 5.62 5.70 5.66 5.74 5.68

(.69) (.66) (.78) (.69) (.70)

Column 5.54 5.54 5.45 5.52 5.52

(.71) (.69) (.82) (.81) (.74)

Mental effortc Video-mediated 48.86 45.54 42.24 42.71 45.09

(24.04) (20.48) (20.50) (22.26) (21.82)

Face-to-face 58.57 57.76 55.95 54.05 56.58

(21.60) (16.60) (16.10) (17.25) (17.89)

Column 54.21 51.65 49.10 48.38 50.84

(23.00) (19.42) (19.48) (20.49) (19.86)

Note. Values enclosed in parentheses represent standard deviations (S.D.).aThe values represent mean scores for the Origami task for each participant. Note. The maximum score

is 40.bThe values represent mean scores on 7-point Likert scales, in which a score of 1 corresponds to the

most negative option and a score of 7 corresponds to the most positive option.cThe values represent mean scores on a one-dimensional scale with ratings between 0 and 150. The scale

has nine descriptive indicators along its axis (e.g., 12 corresponds to not effortful, 58 to rather effortful and

113 to extremely effortful).


4. Discussion

In this paper, we examined if long-term usage of video-communicationtechnologies has an effect on cooperative task performance, cohesion, membersatisfaction and mental effort. In a longitudinal between-groups design, three-personvideo-mediated non-collocated teams were compared to traditional face-to-facesame-room teams in 4 1-h test sessions at 2-week intervals. In the following we willdiscuss the results and the limitations of our research, and draw implications forfuture research and for the design of video-mediated technologies.

ARTICLE IN PRESS


4.1. Effects of experience and communication environment

Consistent with our expectations, a technological adaptation effect was found.Results showed that after the first test session, significant differences between groupson task performance in favour of the face-to-face teams, had disappeared. In general,it seems that as dispersed work groups gain experience, team members compensatefor the technological limitations of the communication medium, such as restrictedbandwidth or low video-image resolutions.

This finding has an important implication for the design of video-communicationtechnologies. For instance, it means that designers of these technologies could runthe risk of devoting time, effort and financial resources to the development ofinnovative solutions to non-problems: problems that diminish in time as people learneffective practices to adopt to the technologies available to them (see also Dourish etal., 1996; van der Kleij et al., 2004).

There is an important lesson to be learned for practice as well. These findings canbe used for example to help project teams make informed decisions on when andwhether to engage in distributed collaboration. Despite obvious benefits associatedwith mediated communication, such as reductions in travel time and cost,communication media should be chosen with caution. For natural work groupsthat share no prior experience and expect no future interaction, so called ad hocteams, distributed collaboration may not be the most suitable communicationsolution. These teams should consider working together face-to-face. In comparison,for groups that meet on a regular basis the benefits associated with distributedcollaboration in all likelihood outweigh the initial limitations of the communicationmedium. These groups should consider, however, having at least one informalmediated meeting to familiarize themselves with the communication medium prior totheir work as a virtual team on the actual topic which makes up the reason for theirexistence.

A remarkable result was that for cohesion neither an effect was found betweengroups nor an effect over time. Differences in information richness between ourcommunication environments were predicted to result in increased task focus at theexpense of social interaction for video-mediated teams. As a result, we expectedparticipant perceptions of cohesion to be higher for face-to-face than for video-mediated teams. The limited time we allowed for our teams to develop could be seenas a possible explanation for the absence of cohesion effects in our experiment.Perhaps it takes natural work groups weeks or even months to develop cohesionbetween group members. Some evidence in favour of this explanation comes fromWalther and Burgoon (1992); who found that it took members of computer-mediated groups several weeks before subjective ratings on trust and socialorientation, both of which constitute an important part of cohesion, became morepositive.

Although we did find that member satisfaction increased over time, as wasexpected, no significant differences in satisfaction were found between groups. Thatno evidence was found for differences in satisfaction between group was somewhatunexpected, because a positive relationship between telepresence and satisfaction

ARTICLE IN PRESS


was reported by several other studies. For example, both Straus (1997) and Baltes etal. (2002) found that computer-mediated groups were less satisfied than were face-to-face groups. An explanation, that was confirmed to some degree by our participants,is that the novel high-realism communication environment in which our participantshad to interact and the perceptual immersion that was created led to a greaterenthusiasm, as compared to both the computer-mediated settings of studies whichdid report differences in satisfaction and our own face-to-face setting. Thisenthusiasm could have compensated for initial frustration and dissatisfactionassociated with the communication limitations in this setting (Baltes et al., 2002).

The fact that satisfaction increased over time for both groups means that this effectis, in all likelihood, due to increased experience of a team as a team, and with thetask rather than greater experience with the communication medium. Experience ofa team as a team tends to reduce the degree of interdependence and the level ofcoordination and information exchanges that are required in dispersed work groups.Moreover, increased experience with a given class of tasks will make collaboratingon those tasks easier in the sense that procedures for doing them can become more ofa routine. This again will tend to reduce the degree of coordination required and therichness of information that must be exchanged (McGrath et al., 1993).

Our hypothesis, that the amount of effort expenditure was greater for video-mediated teams than for face-to-face teams due to difficulties these teams experiencein for example the regulating of conversation, was not supported. The results showedan opposite effect. An explanation is the mere presence of others, which has beendemonstrated to increase attentional demands resulting in better group taskperformance on simple tasks and inhibiting performance on complex tasks (seeHoeksema-Van Orden, 2000 for a comprehensive review). The fact that theinterlocutors in the video-mediated teams were perceived as more distant and less‘real’ could have led to lower attentional demands as compared to the face-to-faceteams. Participants in the video-mediated group made some insightful comments onthis matter. Some participants stated that the fact that their team did not shared thesame physical location made it easier for them to distance themselves fromdiscussions and dodge their responsibilities, a negative condition that is described insocial psychological literature as social loafing (Harkins et al., 1980).

However, task performance suggest that something else might be going on here aswell. If social loafing did indeed occur, we would have expected lower taskperformance for video-mediated teams. This was in fact not the case. In sessions two,three and four, the video-mediated teams performed as well as the face-to-faceteams. As was aforementioned, this was accomplished with less mental effort. Analternative explanation for this findings could be that participants in the video-mediated teams were less distracted during task performance due to the factualabsence of others, making the task less mentally effortful to these participants.

Thus, it appears that the apparent distance or ‘realness’ of the interlocutors indeedhas an impact on users’ behaviour within mediated environments. Although themanner in which these differences are brought about is not entirely clear, this notionhas serious implications for actual work situations involving some sort of mediatedcommunication between team members. In situations in which team members are

ARTICLE IN PRESS


working on rather simple tasks, there is a real chance that they ‘free ride’ on theother team members’ efforts. In these situations, it is important to implementcountermeasures to reduce social loafing. For example, adding an individual elementto the task situation or providing public feedback on all members’ individualperformance on the task at hand helps to avoid social loafing (Hoeksema-VanOrden, 2000). On the other hand, as we have discussed, there is a good chance thatgeographical dispersion of team members is beneficial to the collaboration, becauseteam members are less distracted during task performance due to the factual absenceof others. It is clear that further research needs to be done, investigating therelationship between telepresence and mental effort in mediated teams.

In sum, the present results indicate that temporal aspects can have a large impacton team work. It seems that as dispersed teams gain experience, initial constraintsdue to technological limitations of the communication medium are compensated for.At the same time, however, this finding was not replicated for measures on cohesion,member satisfaction and mental effort. Either no differences between groups onthose measures were found, or initial differences did not disappear as teams gainedexperience.

4.2. Limitations and concluding remarks

Several limitations have to be taken into account in generalizing our findings toother settings and populations. For example, our tested teams consisted of no morethan three participants that had no significant stake in the outcome of theirinteraction. For larger groups (i.e., more than three members) the regulation ofconversational behaviour and the need for coordination becomes more important.For example, Suwita et al. (1997) found that an increase in the number ofparticipants in a video-conferencing setting had an effect on the regulation of non-verbal behaviour. Results of this experiment showed that with four interlocutors,there was more uncertainty as to whom was being addressed, more uncertainty interms of who was listening and more observed difficulties in turn taking than withtwo participants.

Another limitation of this research is inherent to all lab studies using longitudinaldesigns. Experimental control is more difficult to maintain, as participantinteractions cannot be monitored outside the lab setting (Chidambaram et al.,1991). To maintain as much control as possible, we did not allow participants toleave their rooms during the experiment to prevent them from getting to know theother team members outside the test situation. For the same reason we picked theparticipants up separately upon their arrival and escorted them one at a time to themain exit of our lab facilities at the end of each test session. Furthermore, we askedour participants not to discuss the experiment during possible encounters withmembers of their own and other teams.

In our research, we used a high-realism video-communication with large screensenabling selective gaze. This system provided high levels of immersion compared totraditional ‘talking head’ systems. Previous research has shown the advantage ofsuch standard systems (Werkhoven et al., 2001). It may therefore well be that the

ARTICLE IN PRESS


technological adaptation that we observed in the present study does not occur oroccurs less with traditional ‘talking head’ systems. Future research should explorethis issue into more detail.

To be able to investigate the effects of long-term use of video-communicationtechnologies on cooperative task performance, cohesion, satisfaction and mentaleffort, we tested teams at four test sessions of about 1 h each. However, naturalgroups take, in all likelihood, longer to develop. Our assumption that four sessionswould be sufficient to research the effects of long-term use was derived from theliterature. Both Chidambaram and co-workers (1991) and Hollingshead and co-workers (1993) found adaptation effects within a period of four test sessions. Still,the results of this research may not generalize to natural groups that exist over alonger period of time.

Natural groups show some variation in task, communication settings and exhibitconsiderable fluctuations in membership and attendance from time to time(McGrath et al., 1993). The studies that have focused on experience withcommunication medium, task and membership, including this research, have allbut a few dealt with groups of stable composition and structure, working on onetask, in the same communication setting over a given length of time. It would be aninteresting topic for future research, therefore, to examine the effects of variation intask, communication settings, membership and attendance over time. However, weshould keep in mind that more investigations in which a longitudinal approach isadopted will be needed to explore the overall effects that video-communicationtechnologies have on their users. In fact, to paraphrase Wheelan et al. (2003),considering what we know so far, it seems reasonable to adopt a longitudinalapproach in all team research.

Acknowledgements

Part of this research was funded by the European Commission, Directorate-General Information Society, project IST-1999-10044 VIRTUE. The authors wouldlike to thank Anja Langefeld for assistance with the statistical analysis. We wouldalso like to thank Richelle van Rijk and Wytze Hoekstra for their help and technicaladvice throughout this work. Also, the authors would like to thank Heather Young-Griffioen and the reviewers of this journal for their helpful suggestions forimprovement.

References

Alge, B.J., Wiethoff, C., Klein, H.J., 2003. When does the medium matter? Knowledge-building

experiences and opportunities in decision-making teams. Organizational Behavior and Human

Decision Processes 91, 26–37.

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 (1), 156–179.

ARTICLE IN PRESS


Bocker, M., Muhlbach, L., 1993. Communicative presence in videocommunications. Proceedings of the

Human Factors and Ergonomics Society 37th Annual Meeting, Human factors and ergonomics

society, Santa Monica, CA, pp. 249–253.

Brown, R., 2000. Group Processes: Dynamics Within and Between Groups. Blackwell Publishers Inc.,

Oxford, UK.

Carletta, J., Anderson, A.H., McEwan, R., 2000. The effects of multimedia communication technology on

non-collocated teams: a case study. Ergonomics 43 (8), 1237–1251.

Chidambaram, L., Bostrom, R.P., Wynne, B.E., 1991. A longitudinal study of the impact of group

decision support systems on group development. Journal of Management Information Systems 7,

7–25.

Daft, R.L., Lengel, R.H., 1984. Information richness: a new approach to managerial behavior and

organization design. Research in Organizational Behavior 6, 191–233.

Daft, R.L., Lengel, R.H., 1986. Organizational information requirements, media richness and structural

design. Management Science 32, 554–571.

Dennis, A.R., 1996. Information exchange and use in small group decision making. Small Group Research

27, 532–550.

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.

Draper, J.V., Kaber, D.B., Usher, J.M., 1998. Telepresence. Human Factors 40 (3), 354–375.

Dyer, J.L., 1984. Team research and team training: a state of the art review. In: Muckler, F.A. (Ed.),

Human Factors Review. The Human Factors Society, Santa Monica, CA, pp. 285–323.

Harkins, S.G., Latane, B., Williams, K., 1980. Social loafing: allocating effort or taking it easy? Journal of

Experimental Social Psychology 16, 457–465.

Hoeksema-Van Orden, C.Y.D., 2000. Fatigue and performance in groups. Unpublished Doctoral

Dissertation, Rijksuniversiteit Groningen, The Netherlands.

Hogg, M.A., 1992. The Social Psychology of Group Cohesiveness: From Attraction to Social Identity.

Harvester Wheatsheaf, New York.

Hollingshead, A.B., McGrath, J.E., 1995. Computer-assisted groups: a critical review of the empirical

research. In: Goldstein, I.L. (Ed.), Frontiers of Industrial and Organizational Psychology. Jossey-Bass,

San Francisco, CA, pp. 46–78.

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 (3), 307–333.

Ilgen, D.R., Major, D.A., Hollenbeck, J.R., Sego, D.J., 1995. Raising an individual decision-making

model to the team level: a new research model and paradigm. In: Guzzo, R., Salas, E. (Eds.), Team

Effectiveness and Decision-Making in Organizations. Academic Press Inc., San Diego, CA, pp.

99–136.

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 & Work 6, 57–59.

McGrath, J.E., 1984. Groups: Interaction and Performance. Prentice-Hall, Englewood Cliffs, NJ.

McGrath, J.E., 1991. Time, interaction, and performance (TIP). A theory of groups. Small Group

Research 22 (2), 147–174.

McGrath, J.E., Hollingshead, A.B., 1993. Putting the group back in group support systems: some

theoretical issues about dynamic processes in groups with technological enhancements. In: Jessup, L.,

Valacich, S. (Eds.), Group Support Systems: New Perspectives. Macmillan Publishing Company, New

York, pp. 78–96.

McGrath, J.E., Arrow, H., Gruenfeld, D.H., Hollingshead, A.B., O’Conner, K.M., 1993. Groups,

tasks and technology: the effects of experience and change. Small Group Research 24 (3),

406–420.

Moreland, R.L., 1999. Transactive memory: learning who knows what in work groups and organizations.

In: Thompson, L.L., Levine, J.M., Messick, D.M. (Eds.), Shared Cognition in Organizations: The

Management of Knowledge. Lawrence Erlbaum, Mahwah, NJ, pp. 3–32.

ARTICLE IN PRESS


Muhlbach, L., Bocker, M., Prussog, A., 1995. Telepresence in videocommunications: a study on

stereoscopy and individual eye contact. Human Factors 37 (2), 290–305.

Mullen, B., Copper, C., 1994. The relation between group cohesiveness and performance: an integration.

Psychological Bulletin 115 (2), 210–227.

O’Conaill, B., Whittaker, S., Wilbur, S., 1993. Conversations over video conferences: an evaluation

of the spoken aspects of video-mediated communication. Human–Computer Interaction 8, 389–428.

O’Donnell, C.R., Eggemeier, F.T., 1986. Workload assessment methodology. In: Boff, K.R., Kaufman,

L., Thomas, J.P. (Eds.), Handbook of Perception and Human Performance: vol II. Cognitive Processes

and Performance. Wiley, New York, pp. 42.1–42.29.

Okada, K., Maeda, F., Ichikawaa, Y., Matsushita, Y., 1994. Multiparty videoconferencing at virtual social

distance: MAJIC design. In: Furuta, R., Neuwirth, C. (Eds.), Proceedings of the ACM 1994 Conference on

Computer Supported Cooperative Work. Association for Computing Machinery, New York.

Olson, G.M., Olson, J.S., 2000. Distance matters. Human–Computer Interaction 15, 139–178.

Qureshi, S., Vogel, D., 2001. Adaptiveness in virtual teams: organisational challenges and research

directions. Group Decision and Negotiation 10, 27–46.

Sellen, A.J., 1995. Remote conversations: the effects of mediating talk with technology. Human–Computer

Interaction 10, 401–444.

Short, J., Williams, E., Christie, B., 1976. The Social Psychology of Telecommunications. Wiley, London.

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.

Suwita, A., Bocker, M., Muhlbach, L., Runde, D., 1997. Overcoming human factors deficiencies of

videocommunications systems by means of advanced image technologies. Displays 17, 75–88.

Thompson, L.F., Coovert, M.D., 2003. Teamwork online: the effects of computer conferencing on

perceived confusion, satisfaction, and postdiscussion accuracy. Group Dynamics: Theory, Research,

and Practice 7 (2), 135–151.

van der Velden, J.M., 1995. Samenwerken op afstand [Collaborating at a distance; in Dutch]. Unpublished

Doctoral Dissertation, Technical University Delft, The Netherlands.

Veltman, J.A., Gaillard, A.W.K., 1993. Measurement of pilot workload with subjective and physiological

techniques. Proceedings of the Workload Assessment and Aviation Safety Conference, Royal

Aeronautical Society, April 27, 28.

Veltman, J.A., Gaillard, A.W.K., 1996. Physiological indices of workload in a simulated flight task.

Biological Psychology 42, 323–342.

Venkatesh, V., Johnson, P., 2002. Telecommuting technology implementations: a within- and between-

subjects longitudinal field study. Personnel Psychology 55, 661–687.

Verwey, W.B., Veltman, H.A., 1996. Detecting short periods of elevated workload: a comparison of nine

workload assessment techniques. Journal of Experimental Psychology: Applied 2 (3), 270–285.

Walther, J.B., Burgoon, J.K., 1992. Relational communication in computer-mediated communication.

Human Communications Research 19, 50–83.

Welch, R.B., 1999. How can we determine if the sense of presence affects task performance? Presence:

Teleoperators and Virtual Environments 8, 225–240.

Werkhoven, P.J., Schraagen, J.M., Punte, P.A.J., 2001. Seeing is believing: communication performance

under isotropic teleconferencing conditions. Displays 22, 137–149.

Wheelan, S.A., Davidson, B., Tilin, F., 2003. Group development across time. Reality or illusion? Small

Group Research 34 (2), 223–245.

Whittaker, S., 2003. Things to talk about when talking about things. Human–Computer Interaction 18,

149–170.

Whittaker, S., O’Conaill, B., 1997. The role of vision in face-to-face and mediated communication. In:

Finn, K., Sellen, A., Wilbur, S. (Eds.), Video-Mediated Communication. Lawrence Erlbaum

Associates, Mahwah, NJ, pp. 23–50.

Wierwille, W.W., Eggemeier, F.T., 1993. Recommendations for mental workload measurement in a test

and evaluation environment. Human Factors 35, 263–281.

Zijlstra, F.R.H., 1993. Efficiency in work behaviour: a design approach for modern tools. Unpublished

Doctoral Dissertation, Delft University, The Netherlands.

Documents

On the passage of time: Temporal differences in video-mediated and face-to-face interaction