Individual and group performance of computerizededucational tasks

Joseph Klein

Published online: 1 February 2012# Springer Science+Business Media, LLC 2012

Abstract This study examines the conditions under which task performance isaccomplished more efficiently by a single individual or a group. 116 participants,ranging in age from high school level through holders of Master’s degrees, 58 womenand the same number of men, were presented with a computer game based oneducational software, arranged in five levels of difficulty. A comparison was madeof the speed in which objectives were attained when performed individually andwhen divided among groups of two to four partners, controlling for the type ofcoordination required. Cases of expedient and inexpedient division of labor wereidentified, as were factors affecting feasibility, among them the number of simulta-neous tasks to be performed, complexity level, number of participants and theintensity of coordination required among them. Basic principles, pertaining to thefunctional division of roles in learning activities and in educational administration,are discussed.

Keywords Computerized learning . Interactive simulations . Individual learning .

Cooperative learning

School administrators and teachers are frequently required to deal with multiple taskssimultaneously. Agents from within and outside demand immediate response topetitions, without consideration of other urgent issues with which principals areoccupied. The need to divide attention in order to cope with information flowing infrom a number of sources is observable in many work settings and in industrialsystems that require operators of sophisticated equipment to supervise a number offunctions at the same time (Ryu and Myung 2005). The present study examines theefficacy of dividing educational tasks among a number of partners and alternatively,concentrating them in the hands of one person.

Educ Inf Technol (2013) 18:443–458DOI 10.1007/s10639-012-9187-y

J. Klein (*)Bar-Ilan University, School of Education, Ramat-Gan 52900, Israele-mail: Joseph.Klein@biu.ac.il

1 Theoretical background

1.1 Divided attention

The term “divided attention” describes a cognitive condition that affects theability to perform a number of tasks simultaneously and at a suitable level(Barnes and Dougherty 2007). The difficulty of focusing attention on a number ofobjectives causes errors in performance (Girard 2007; Endsley and Smith 1996;Gopher 1993), a decline in the quality of treatment (Ruthruff et al. 2001) and longerreaction time (Levy et al. 2006). Two models explain the difficulty, the ‘limitedresource’ and the ‘bottleneck’ models. The limited resource model proposed byJames (1890), and later modified, describes a process for dealing with informationthat begins with absorption and retention in memory stores and processing at alater stage. Each memory depot has a limited capacity and reacts to a differenttype of stimulus, such as visual signals or aural input. The visual storehouse,for instance, can hold no more than a certain number of images, and the auralchannel can accommodate a small number of tonal composites. When differentactivities engage separate depots during processing, no attention overload iscreated, but such a problem does occur when two tasks compete for the limitedresources of the same repository or when the requirement exceeds the capacityof that store (Allport et al. 1972).

The bottleneck model, in contrast, assumes that simultaneous demands are pro-cessed sequentially. When a specific task engages a resource, other matters must waittheir turn for attention. Researchers are divided on the question of whether thebottleneck is created when visual and aural information is received or during a laterstage of neural activity, such as the apperception / encoding stage and responseprocessing stage (Pashler and Johnston 1998; Elidan and Friedman 2006). Pashler(1994) and Watter and Logan (2006) conducted empirical studies to validate thevarious models. Their findings do not negate the limited resources approach but docontribute to verification of the bottleneck model.

Other researchers have suggested details of mechanisms for memory storage andretrieval. Lashley (1951) discerned indications that information might be stored in ahierarchical fashion, rather than serially. Such a structure would increase storagepotential and prevent the loss of information that could occur if a linear arrangementwould be disrupted. Terrace and Chen (1991) raised the possibility that new bits ofrelated information from the environment might be introduced serially to the samememory storage site, forming chunks and increasing overall capacity.

The Multi-store Model (Atkinson and Shiffrin 1968) describes the memoryprocess as a flow of information from the environment via the sense organs to asensory memory store. If the individual takes note of the input, it goes on to the nextstore, the short-term memory. If the content of the short-term store is rehearsed, it isincorporated into the long-term store from which it can be retrieved by recall.

The Multi-store Model has been largely replaced by Working Memory(Baddeley and Hitch 1974), which presents the sensory memory as a more complexWorking Memory. It was originally thought to be composed of a supervisory systemand two slave systems, specializing the verbal and visuo-spatial domains respectively.Later, an additional slave system, the episodic buffer, was found; its function is the

444 Educ Inf Technol (2013) 18:443–458

integration of the information in the verbal and visuo-spatial systems, and perhapsfrom other sources, into a single episodic presentation. The working memory is theshort-term site, with input to the long-term memory.

In their proposal of the Levels of Processing Model of memory, Craik andLockhart (1972) did not consider the structural aspect, but rather the manner in whichmemory is established. Instead of discussing short- and long-term memory, they dealwith the depth of processing, which is defined by the authors as “the meaningfulnessextracted from the stimulus rather than in terms of the number of analyses performedupon it”.

Sharon (1997) identified the ability of most members of the population to performseveral simple activities simultaneously. However, they have difficulty in makingsimultaneous decisions (Pashler & Johnston, op. cit.). Few people can divide theirattention among a large number of tasks. Individual differences are explained by theemployment of different cognitive strategies in dealing with large masses of infor-mation. The cognitive devices that enable some people to perform tasks simulta-neously employ a method of imposing organization or order, thus assisting inhandling information loads and economizing on processing (Pashler & Johnston,op. cit.). Maturana and Varela (1980, p. 137) define this ability as “the relations thatdefine a system as a unity, and determine the dynamics of interaction and trans-formations which it may undergo as such a unity…”

Multiple demands can be managed in a number of ways:

1. Work overload may be relieved by flexible and rapid transition between tasksuntil all have been resolved, with preference given to pressing issues (Yeung andMonsell 2003; Gopher et al. 2000).

2. Adaptive executive control strategies and procedural rules previously learnedmay be used for dealing with new situations, in order to minimize uncertainty inaddressing them (Polk and Seifert 2002; Meyer and Kieras 1999). Learningadvanced procedural rules helps to formulate semi-automatic procedures thatrequire low levels of attention, like a driver who can perform certain actionswith almost no awareness of having performed them. To this end work schedulesare reordered so that, for example, several activities requiring superficial atten-tion are performed together rather than sequentially. Even in this situationcognitive limitations do not prevent the simultaneous processing of complexthinking tasks (Rosen 2008).

3. Responsibility for implementation of complex assignments may be placed onindividuals whose cognitive style enables them to deal effectively with suchchallenges (Sharon 1997).

4. A complex problem may be divided into a series of hierarchical actions, as in adecision tree, with focus of attention on one problem at each stage (Frederiksenand White 1989; Taatgen 2007).

5. Work may be divided among a group of people working together in coordination.This procedure is common in many sectors, among them educational systems(Garbis and Artman 2004). The decision to organize teamwork presents adilemma as to the optimal division of subtasks among partners. The additionof more participants beyond that optimum will not improve performance(Michaelson et al. 1989).

Educ Inf Technol (2013) 18:443–458 445

1.2 Factors affecting the efficacy of division of tasks

The effectiveness of the division of tasks is not something to be taken for granted.The literature poses the question whether two heads are better than one, and theanswers are not consistent. This is explained by a lack of methodological uniformityamong the various studies and by the many factors that affect individual and groupefficacy, not all of which were controlled for in the studies (Blinder and Morgan2005; Kerr et al. 1996).

The advantages ascribed to individual work in comparison to group efforts includespeed of decision making in simple situations and quality of decisions in situationsrequiring centralized organization and coordination of information. Individuals arenot subject to the problems characteristic of group functions, such as fear that voicingdissenting ideas might engender criticism and conflict with the leader’s position(Davis and Toseland 1987). Other studies indicate little difference between individualand group judgment (Kocher and Sutter 2005; Nijstad et al. 2006).

The advantages attributed to group decision making include the possibility ofdealing with larger amounts of information, brainstorming to elicit new ideas,reciprocal exchanges of ideas, and minimizing faulty conclusions by identifyingerrors at an early stage of decision formulation. Cooperative projects also benefitfrom collective wisdom, the opportunity for analysis of situations from differentviewpoints, and the ability to choose better options than those that would be selectedby individuals (Pinto et al. 1993; Kerr and Tindale 2004; Cox and Hayne 1998;Blinder and Morgan 2005). Porush (1984) suggests a higher statistical probability ofattaining quality decisions rapidly and accurately. Joint activities, by their nature,involve gradual decision processes rather than a single, possibly hasty, action (Shaw1976). Groups are also characterized as being more systematic in their decisions,because all members are committed to explaining their proposals convincingly to theother members (Davis and Toseland 1987; Kameda et al. 2003; Komiya et al. 2007).Group decisions also represent collective democratic thought, which is not reflectedin individual decisions.

The issue of sharing takes on greater importance in this age of information andcommunication technologies (ICTs) and the development of Web 2.0 and eLearning2.0. These tools provide a universal technological platform for transmitting messages,thinking and engaging in shared multi-age learning. The new environment is rich inits sources of information, accessibility to experts and flexibility in terms of sched-uling activity. Group thinking leads to insights that at times exceed the knowledgeavailable from a single expert. An experiment conducted by Tetlock (2005) found thatpredictions made by people who were knowledgeable in a given field of knowledgebut not considered experts, were better than predictions proposed by the best expert.The study of the influence of collective cognition on promoting personal cognition,and the abilities shared by learning and problem solving was begun in the precedingcentury, and one of its major proponents was Vygotsky (1978). Web 2.0 was ground-breaking because of the tools it included, which enabled individuals from around theworld to engage together in community action while making the cognitive processesinvolved accessible to all. The need for joint cognition will continue to grow as theenvironment in which people work becomes more complex and dynamic and newand relevant joint solutions will be needed for this environment (Passig 2007).

446 Educ Inf Technol (2013) 18:443–458

Learning in this new social age has been transformed from passive to constructiveand dynamic. As such, it requires the use of appropriate cognitive strategies, socialdata processing, regulation of attention and emotions, and flexibility in cognition.It also requires the use of higher order thinking skills such as methods forseeking information, the ability to evaluate the quality of information sources anddealing with several sources of information simultaneously (Lombardozzi 2001;Passig, op. cit.).

The efficacy of group work is affected by the quality of inter-member coordinationand the degree to which aspirations to attain common goals are shared. Varioustypologies of coordination have been offered. One such distinction differentiatesbetween collaboration and cooperation (Malone and Kevin 1994; Kogut and Zadner1996). Teasley and Roschelle’s (1993) description of collaboration is of “a co-ordinated, synchronous activity that is the result of a continued attempt to constructand maintain a shared conception of a problem. This is distinct from co-operation,which they define as ‘the division of labor among participants … where each personis responsible for a portion of the problem-solving’ ”. In addition to simultaneouscooperative efforts, some projects require chain processing, in which each individualdoes his/her assigned part in the effort and passes the results to the next in line.

When the probability of communication failure is very high, it is preferable to havethe work done when practical by one operator rather than a number of agents(Michaelson et al. 1989). An individual performing a complete job independentlyavoids the pitfalls of teamwork. However, the effort might fail as a result of the needto divide attention among multiple sub-tasks.

Work in schools entails many kinds of joint undertakings, such as staff meetingsand day-to-day cooperation with work colleagues and outside agents such as localauthorities and the community. The ability to organize a group activity and to gainsupport of participants must be learned. Some challenges involve intensive coordi-nation of disparate elements, such as planning a school’s teaching schedule with itsmyriad demands (Sharan 2002), with regard to teacher and facility availability. Skillis also required in conducting the “orchestra” of learning in class (Sharan 2002).

Empirical research has supported computer-supported collaborative learning(CSCI). It enriches pedagogical activity, and promotes coordinated activation oflearning processes that mediate between the teacher and daily life (Hamalainen2008; Hämäläinen et al. 2008). It has been reported to help in meeting learningchallenges at different levels of reciprocal relations in modern society and to assist inpresentation individual and social abilities (Dillenbourg et al. 2009; Arvaja et al.2008). Difficulties have been reported in computer-assisted learning. These wereattributed to personal style, work habits, lack of training for group projects, andcomputer hardware that is suited more for individual use than for team work (Scott etal. 2000; Laurillard et al. 2000).

Kerawalla et al. (2008) developed software based on cooperative efforts, whichdoes not allow individuals to perform tasks by themselves. Even in such an extremesituation some pupils simply copied tasks from classmates, without contributing theirown independent thinking to a supposedly joint effort. The implementation of anefficient course of action in group projects requires educational and psychologicalforethought, as well as an empirical examination of productive and unproductivepatterns of teamwork.

Educ Inf Technol (2013) 18:443–458 447

The decision of whether to assign a task to one person or to divide it among severalindividuals should take into account the complexity of the task and the amount ofcoordination required among participants (Cox and Hayne 1998). When very intensecooperation is required, it is often preferable to leave the task in the hands of oneperson. Conversely, when the assignment is very complicated, involving muchdivision of attention, there is an advantage in dividing it among several individuals,each of whom performs a simple sub-task (Davis and Toseland 1987). The interactionbetween these conditions creates diverse variants, for which the relative input of eachof the elements must be weighed in order to determine whether a project is besthandled by a single individual or by a group. The optimal cut-off points in makingthis decision are the subject of our discussion. In the absence of sufficient empiricalknowledge, limited and intuitive personal knowledge serves as the basis for decisionsand procedures pertaining to the centralization–or division–of performance of edu-cational tasks. This is evident in the manner in which an individual or group-learningactivity is assigned, without conducting a systematic study of the most favorablelearning environment for the specific task.

In order to shed light on the optimal conditions for sharing tasks in an educationalproject, this paper examines the performance of participants in a highly controlledcomputerized assignment.

2 Methods and materials

2.1 Tools

A computer simulation was used in this study. The participants’ ability to transitionflexibly between tasks was tested, as was their skill in dealing rapidly with essentialissues, which is necessary when several tasks must be coped with simultaneously andeffectively (Yeung and Monsell 2003; Gopher et al. 2000). Speed and effectiveness ofperformance were compared between a single player and from two to four players, insimple and complex tasks. This is not the first simulator designed to test the efficacyof team work (Holzinger et al. 2009). Its uniqueness lies in its programming whichenables it to monitor the efficacy of cooperative activity for varying numbers ofparticipants and for five different levels of activity complexity. This set-up simulatesconditions in the education field with its variety of cooperative learning programs atdiffering levels of complexity.

2.2 Essence of the game

The simulator screen displayed two objects, the “tracker” and the “target”. Partic-ipants, aided by the keyboard, controlled the movements of the tracker and attemptedto close in on the target, which moved randomly on the screen. There was nocommunication, verbal or otherwise, between players of a team. The only indicationthat one received of the activity of others was derived from observation of themovements of the tracker on the screen.

The exercise was presented in five versions of increasing complexity toindividuals and to groups of two, three or four participants. The computer

448 Educ Inf Technol (2013) 18:443–458

recorded the exact amount of time that elapsed between the beginning and endof each game.

Level A - one vs. two players.

The target moves randomly over the length (Y) and width (X) of the screen. Inorder to reach the target, the tracker must be moved by means of the keyboard. Asingle player controls the keys for both coordinates simultaneously. When twoplayers are involved, one operates on each coordinate.

Level B - one, two, or three players.

In addition to the movement of the target as described above, it exhibits randomchanges of color to one of four hues: red, green, blue, or yellow. The player(s) mustnot only reach the target, but also adjust the color of the tracker to match that of thetarget. A single player is required to cope simultaneously with directional and colorvariations. When two players cooperate, one controls the direction and the other thecolor. In the case of three players, one manages the color alterations, and each of theothers is responsible for regulating movement on one of the coordinates.

Level C - one, two, or three participants in a more complex setting.

The color remains constant, as in level A. The target moves on three coordinates, X,Y, and Z (depth). The addition of the Z coordinate, involving a series of highly diversechanges, presents a challenge that is more difficult than that of level B, in which thecolor switches were limited to four possibilities. A single player must direct movementof the tracker on all three coordinates. When two participants cooperate, one is in chargeof the X and Y directional movements, and the second controls the depth factor. With theaddition of a third partner, each of the players works with one coordinate.

Level D - one to four players.

The color modifications of Level B are added to the program of Level C. Whenmore than one participant plays, the tasks are divided in the manner described above.

Level E - one to four players.

The game design is the same as that of Level D. However, the target is programmedto move faster on all three coordinates, and the colors change more rapidly.

The research design is detailed in Table 1.Legend: x-horizontal axis; y-vertical axis; z-depth axis; h-changing color of object

A dash separating the letters, such as x-y-z indicates division of tasks. One participantis in charge of x, the second of y and the third of z. When the same player performsmore than one function, no dash appears between the symbols.

Three comparisons were conducted:

Group A: Comparison of performances of one versus two participants at all fivelevels of complexity.Group C: Comparison of performances of one to three participants at the fourhighest levels of complexity, in which three players can be included.Group C: Comparison of performances of one to four participants at the twohighest levels of complexity, in which four players can be included.

Educ Inf Technol (2013) 18:443–458 449

2.3 The educational aspect

The game helps participants to adapt to team work on the computer. It is arranged insuch a way that when cooperation is needed, all participants are dependent on theirpartners’ activity and none of them can progress by ignoring the others. The gamewas so attractive that participants of all ages asked to continue with the activity whenthe research measurement was finished.

2.4 Controlled variables

Factors that are thought by some researchers to influence decision-making processeswere chosen as controlled variables. These included seniority at work (Hartman et al.1990), age, educational level (Lehman et al. 1988), and cognitive style (Eugene 2004;Hough and Ogilvie 2005; Hayes et al. 2003).

Learning style affects the activation of a large variety of individual strategies andsolutions for stressful situations. Therefore, a questionnaire (Felder and Brent 2005)was used as a control for the cognitive style of the participant. Similarly, type ofcoordination among participants was controlled. Participants observed the actions oftheir colleagues on screen, but no oral communication was permitted betweenparticipants during the activity.

2.5 Participants

The random sample of 116 participants included an equal number of men andwomen, 58, who ranged in age from 16 to 36 (mean 0 26.12) and varied in theireducational level from high school level to master’s degree. The high school studentsvolunteered to participate in the experiment, which was conducted in their schools’computer laboratories. The decision to use high school students was intended toexamine the issue under study in the context of school studies as well. The olderpopulation was also composed of volunteers. The age limitation of 35 was predicatedon the need to afford appropriate representation to each age group. Expanding therange of ages would have required an increase in the number of participants, which

Table 1 Research design

No. of players Level A Level B Level C Level D Level E

Game complexity

One Group: a Group: a, b Group: a,b Group: a,b,c Group: a,b,c

Tasks: x y Tasks: xyc Tasks: xyz Tasks: xycz Tasks: xycz

Two Group: a Group: a, b Group: a,b Group: a,b Group: a,b,c

Tasks: x-y Tasks: xy-c Tasks: xy-z Tasks: xy-cz Tasks: xy-cz

Three Group: b Group: b Group: b,c Group: b,c

Tasks: x-y-c Tasks: x-y-z Tasks: xy-c-z Tasks:xy-c-z

Four Group: c Group: c

Tasks: x-y-c-z Tasks: x-y-c-z

450 Educ Inf Technol (2013) 18:443–458

would have been difficult because of the complex nature of the experiment. It shouldbe noted that many teachers begin to work in schools at age 25 or earlier, andtherefore the sampling is also relevant for educators of these ages. Because of thelimited age range, follow-up studies are needed to examine additional age groups. Allthe participants were experienced in operating a variety of computer programs, amongthem components of Office and the internet. The only skills that were required in thisstudy were basic familiarity with operating a keyboard and mouse. The participantsreceived detailed instruction about the nature of the study in advance and were giventime to practice working with all elements of the program before it began. All partic-ipants completed all parts of the experiment. Their motivation to participate wastriggered by the presentation of the study aim, which was to clarify a heretoforeunresolved problem of when two heads are better than one, and when they are not.The challenging nature of the experiment, which was exemplified by the computer-based program that was used, contributed greatly to their willingness to invest their time.

3 Results

In order to detect interactions, if any, between performance, as measured by elapsedtime, different levels of complexity and varying numbers of participants, a two-directional analysis of variance (ANOVA) was conducted with repeated measures.The dependent variable was elapsed time and the independent variables were: taskcomplexity level (five levels) and number of participants performing the task(one to four participants). The controlled variables were: educational backgroundof participants, gender.

Means and standard deviations are presented in Table 2.A comparison of performance by a single participant and two participants

shows a significant difference in elapsed time at the various levels of complexity,F(4,504) 0 113.30, P<.001, η²00.473. There is a significant interaction betweentask complexity level and the number of participants performing it, F(4,504) 0 3.67,P<.01, η²00.028. A Bonferroni analysis yielded a significant difference between eachof the five levels of complexity in elapsed time. A single player attained the result fasterthan two participants in the first and third tasks, while two participants were better in theother tasks. Educational background, age, and gender of the participants were notsignificant in the various analyses and were therefore omitted from the remaining tests.

In comparing the performance of a single versus two and three participants, adifference in elapsed time was found at the various levels of complexity, F(3,567) 0108.34, P<.001, η²00.364. A Bonferroni analysis revealed a significant difference inelapsed time to result between each of the five levels of complexity. A significantinteraction exists between task complexity level and the number of participantsperforming it, F(2,189) 0 10.08, P<.001, η²00.096. A Scheffe test found a significantdifference in elapsed time between one, two and three players (M045.4 vs. M033.02vs. M024.56). Two participants had the lowest elapsed time to result in the secondlevel of task complexity. In the other levels of complexity three players reached thetarget faster.

In comparing the performances of one versus two, three and four participants, adifference was found in elapsed time to result at the various levels of complexity,

Educ Inf Technol (2013) 18:443–458 451

F(1,252) 0 82.47, P<.001, η²00.247. A significant interaction is found between tasklevel of complexity and number of participants performing it, F(3,252) 0 6.78, P<.001,η²00.075. A Bonferroni analysis revealed a significant difference in elapsed time totarget between two levels of complexity. In the Scheffe test, three players attained theresult faster than four players. The slowest result was recorded by a single player.

4 Discussion

Decisions, with respect to assignment of educational and organizational roles andcentralization or decentralization, are always influenced by personal values, experi-ence, and intuition, rather than research on the relevant issues. The results show that

Table 2 Player function at different levels of complexity

452 Educ Inf Technol (2013) 18:443–458

the direct connection, expected intuitively between the complexity of a task and thenumber of participants involved, is not necessarily confirmed by empirical assess-ment. Obviously, this issue takes on special importance in the age of Web 2.0 andeLearning 2.0, which intensify the practice of shared and multi-age learning world-wide. This activity includes simultaneous interactive discussions with a number ofpartners.

Two heads are not always better than one. The interaction between variousaspects of an assignment may at times require a different assessment of therequired staff size. It should be emphasized that the conclusions drawn fromthis work were based on an experimental design in which the tasks and thegoal were well defined at the time of assignment. A continuation of the studywould be the investigation of cooperative exercises in which participants mustdecide on the steps required to meet the goal.

In the experimental setup discussed here, the optimal number of partners dependedon the type of action prescribed. When the demand for division of attention was notonerous, evidence indicated a preference for a single participant who did not have todeal with the demands of interpersonal coordination. With a slight increase incomplexity, two or three partners were more successful than an individual performer.As the game plan became even more intricate, three partners had the advantage. In alltrials, a team of four was consistently less successful. The results suggest that in acomplex format, the problem of divided attention was a more dominant factor thandifficulty in coordination between two or three players. Cooperation became a majorissue again when groups of four were activated—a clear case of too many cooksspoiling the broth.

The interaction between the elements of cooperation and divided attention is welldemonstrated at the third of the five levels of complexity. It could have been expectedthat three partners would perform more efficiently than two, and that two would bebetter than one. In the tests, however, the single participant reached the target fasterthan the pair. Two explanations for this outcome are suggested. The advantage of theindividual over the pair is ascribable to the fact that in both cases the problem ofdividing attention was present, but the individual had the advantage of not having todeal with interpersonal coordination.

An alternative, or additional possibility is that the manipulation of the Z coordinatewas, as described in the Methods section, more complicated than the alternationbetween four colors in the second level, but division of attention proved to be lesscritical when all activity was based on directional manipulation rather than on twoconcepts, direction and color. A single player, coping with the added Z coordinate andthe consequent increased demand for division of attention, encountered less difficultythan. Did two players who were required to coordinate their movements. In athreesome, the need for internal coordination increased, but task division amongthe three simplified performance and precluded the difficulties of dividing attention.

When the level of task complexity rose to that of levels 4 and 5, individuals foundit harder to divide their attention between the various tasks. This overshadowed theiradvantage in terms of coordination, and therefore in these cases the twosome loggedbetter results than the individual. The findings emphasize the need to base decisionswith respect to centralization or division of tasks on an understanding of the stressesinvolved, as well as a thorough awareness of the demands of the specific assignment.

Educ Inf Technol (2013) 18:443–458 453

Because of the interactivity of the game, participants of all ages asked to continuethe activity even after the research measurements had been concluded. Although thisobservation does not relate directly to the main theme, it demonstrates severaladvantages to utilization of such devices in educational work. An exercise such asthis one introduces participants to computer learning in an entertaining way, andencourages them to engage in a cooperative enterprise.

The rich findings indicate that decisions about centralizing or dividing tasksshould be based on knowledge and not simply on logic and intuition. This alsoapplies to the prevailing assumptions about differences in decision making betweenmen and women and between different age groups. In certain tasks background dataseem to make a difference in their decisions while in other tasks no differences arefound. This was also true in the specific tasks tested in this study. Each task should beexamined individually.

In the preceding paragraphs, emphasis was placed on the nature of the task. Whatabout the individuals involved? During the last century, product planning underwenta revolution. Instead of a “consumer be damned” philosophy that obliged buyers tomake do with the products available, the new approach of human engineering placedresponsibility on the planners to adapt products to suit the needs of the end users. Asimilar transformation is essential in the educational field and in other areas as well.Roles should not be divided or assigned only on the basis of intuitive decisions takenby those in authority, but only after the human factor, including individual abilitiesand limitation, has also been considered.

The results show that cooperative action may, under certain conditions, requiremore time than a single individual might need to do the same job. However in many,probably most cases, speed is not of the essence; quality of the finished product iswhat counts. The decision with regard to cooperative or individual activity must takeexpected achievement into consideration. Under conditions in which individuals areas efficient as a team and can accomplish the purpose within a reasonable period oftime, there is little justification for compelling them to work in partnership. Even ifthe exercise has been planned with the purpose of training for cooperation, theoutcome will be positive only if the participants are able to see the advantage ofteam work.

The findings obtained are of significance when planning effective discussionsamong members through the Web and eLearning. Aspects to be taken into consider-ation include the intensity of the expected discussion, the number of simultaneousactivities they will be expected to perform and the reaction speed required. At times itis preferable to divide a large group of participants into smaller subgroups, which willallow participants to coordinate their activities effectively with the others. Thedecision to perform individual or group activity depends on many factors. Our studyfocused on clarifying some of them, but they are not the only ones. Passig (2007)reports on the need for high order cognitive skills for engaging in interactive groupactivity. In the cognitive and affective context, there are those whose cognitive andaffective learning styles define them as ‘loners’ who thrive in individual work and arelimited in group work. There are those whose cognitive and affective learning stylesdefine them as ‘collaborators’ who function better in group work, and there are thosewhose cognitive and affective learning styles define them as ‘leaders’ a trait mostnecessary when group learning is undertaken (Vermunt 1996; Pirola-Merlo et al. 2002).

454 Educ Inf Technol (2013) 18:443–458

Another consideration to be taken into account is what the task is expected toachieve. The task that was tested in our study was defined as attaining the goal inminimal time. Where an individual can attain this objective faster than a group, thereis no reasonable cause for requiring collaborative work in such a task. Managers ofeducational institutions often broadcast mixed signals regarding their expectationswith respect to cooperation among teachers as well as pupils. Administrators maylaud the importance and centrality of group work because of its social and academicbenefits (Littleton and Light 1999; Sharan 2002), but teachers are aware thatindividual quality is often the main criterion in evaluation of their own work. Inprofessional ties with their colleagues, they may avoid sharing special pet projects oftheir own, fearing that any distinctive contribution would be buried in the collectiveresponse (Klein 2005).

In an era of expectations of personal excellence, it is no wonder that those withhigh abilities ignore instructions to work cooperatively, preferring instead to adhere totheir own individual work style to prove their merit. If teamwork is truly important toa company or educational institution, it should be included in achievement evaluationsheets and be given its due worth. Any attempt to channel pupils and teachers toparticipate in individual or group activity without considering these factors willsucceed briefly, if at all.

In this report, considerable information with respect to the interactions betweenwork load and cooperative effort was provided by an exercise that required nothingmore than following a set of rules. This offers an enticing prospect of the knowledgeto be derived from an investigation of individual and cooperative work when the taskinvolves elaboration of new ideas as well as planning and implementation. Results ofsuch studies could provide insight into organizational issues such as centralization vs.decentralization, within the educational framework as well as in other spheres. A fewexamples are presented here.

Uncertainties about work load and division of attention arise when a limitednumber of staff members are assigned to key posts in school. When, for instance,assistant principals and social coordinators also function as homeroom teachers andsubject instructors, their effectiveness declines. In many cases these agents areunaware of the deleterious effect on the quality of their work, and fail to recognizethe advantage of dividing tasks among a larger number of staff members. Managersof educational institutions often prefer to centralize undertakings and avoid assigningduties to other colleagues. The question of concentrating or decentralizing educa-tional responsibility can be demonstrated by the manner in which dealing with theneeds of pupils is handled in various schools. In some systems, a single educator isappointed to coordinate all efforts involved in matters of the values, social activityand behavior of pupils. Elsewhere, this duty is assigned to grade coordinators, each ofwhom is responsible for all the classes of a specific age level, or to homeroomteachers, working with their own classes in cooperation with relevant personnelwhen necessary.

Each of these work patterns represents a different educational approach. Thoseadvocating the appointment of homeroom teachers for social purposes see them assymbolic parents who serve as the children’s main source of support and as individ-uals who can take responsibility for all of their school affairs. Those preferring a morecentralized approach of the task view the educational institution as a tool for

Educ Inf Technol (2013) 18:443–458 455

preparing children for life, in which adults are required to function independently,without the luxury of an all encompassing, hothouse-type nurturing habitat. To date,no empirical study has been conducted to compare the efficacy of the various models,controlling for factors of complexity, division of attention and coordination.

The tug-of-war between the difficulties represented by multiple work demands andcooperative efforts has been demonstrated here in its simplest form, affording a mereglimpse into an issue that requires more rigorous study. An investigation of higherlevel cooperative activity should reveal a richer and more complex set of underlyingcognitive and behavioral principles. Intensive coordination requires well timedcommunication, as can be seen in the reciprocal ties between conductor andorchestra, or among members of a large medical team engaged in complex surgery.A rise in the level of coordinative intensity increases the probability that difficultiesin synchronization will develop between members, as a result of cultural, social,psychological and technical obstacles.

Understanding of these factors should provide a firmer basis for determination ofthe relative merits of independent effort and cooperative action as an answer tospecific needs. Educators of all ranks should be trained in utilizing new insights onthe subject to design an effective learning environment.

References

Allport, D. A., Antonis, B., & Reynolds, P. (1972). On the division of attention: A disproof of the singlechannel hypothesis. Quarterly Journal of Experimental Psychology, 24, 225–235.

Arvaja, M., Häkkinen, P., & Kankaanranta, M. (2008). Collaborative learning and computer-supported collaborative learning environments. In J. Voogt & G. Knezek (Eds.), Internationalhandbook of information technology in primary and secondary education (pp. 267–279). New York:Springer.

Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. InK. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation (vol 2). London:Academic.

Baddeley, A. D., & Hitch, G. (1974). Working memory model. In G. Bower (Ed.), Recent advances onlearning and motivation, vol. 8 (pp. 47–89). New York: Academic.

Barnes, K. A., & Dougherty, M. R. (2007). The effect of divided attention on global judgment of learningaccuracy. The American Journal of Psychology, 120(3), 347–359.

Blinder, A. S., & Morgan, J. (2005). Are two heads better than one? monetary policy by committee. Journalof Money, Credit, and Banking, 37(5), 789–811.

Cox, J. C., & Hayne, S. C. (1998). Group versus individual decision making in strategic market games.Mimeo: University of Arizona.

Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journalof Verbal Learning and Verbal Behavior, 11, 671–684.

Davis, L. V., & Toseland, R. W. (1987). Group versus individual decision making: An experimentalanalysis. In R. W. Toseland & P. H. Ephross (Eds.), Working effectively with administrative groups(pp. 110–195). Binghamton: Howorth.

Dillenbourg, P., Järvelä, S., Fischer, F. (2009). The evolution of research on computer-supportedcollaborative learning: From design to orchestration. In N. Balacheff, S. Ludvigsen, T. de Jong,A. Lazonder, S. Barnes (Eds.), Technology-enhanced learning. Principles and products (p 3–19).

Elidan, G., & Friedman, N. (2006). Learning hidden variable networks: The information bottleneckapproach. Journal of Machine Learning Research, 6, 81–127.

Endsley, M. R., & Smith, R. P. (1996). Attention distribution and decision making in tactical air combat.Human Factors, 38(2), 232–249.

Eugene, S. S. (2004). Cognitive style and the management of small and medium-sized enterprises.Organization Studies, 25(2), 155–181.

456 Educ Inf Technol (2013) 18:443–458

Felder, R. M., & Brent, R. (2005). An exploration of differences in student learning styles, approaches tolearning (deep, surface, and strategic), and levels of intellectual development. Understanding StudentDifferences. Journal of Engineering Education, 94(1), 57–72.

Frederiksen, J. R., & White, B. Y. (1989). An approach to training based upon principal task decomposi-tion. Acta Psychologica, 71(1–3), 14–89.

Garbis, C., & Artman, H. (2004). Team situation awareness as communicative practice. In S. Banbury & S.Tremblay (Eds.), A cognitive approach to situation awareness: Theory and application. Aldershot:Ashgate & Town.

Girard, N. (2007). Multitasking: How much is too much? AORN, 85(3), 505–506.Gopher, D. (1993). The skill of attention control. In D. E. Meyer & S. Kornblum (Eds.), Attention and

performance, XVI (pp. 299–322). Hillsdale: Erlbaum.Gopher, D., Amony, L., & Greenshpan, Y. (2000). Switching tasks and attention policies. Journal of

Experimental Psychology. General, 129, 308–339.Hamalainen, R. (2008). Designing and evaluating collaboration in a virtual game environment for voca-

tional learning. Computers in Education, 50(1), 98–109.Hämäläinen, R., Manninen, T., Oksanen, K., & Mannila, B. (2008). Designing and investigating collab-

orative 3D learning spaces for future learning. Computers in Human Behavior, 24(6), 2496–2506.Hartman, S. J., Lundberg, O., & White, M. (1990). Effect of background and organizational position on

executive planning. Journal of Social Psychology, 130, 801–811.Hayes, J., Allinson, C. W., Hudson, R. S., & Keasey, K. (2003). Further reflections on the nature of

intuition-analysis and the construct validity of the Cognitive Style Index. Journal of Occupational andOrganizational Psychology, 76(2), 269–278.

Holzinger, A., Kickmeier-Rust, M. D., Wassertheurer, S., & Hessinger, M. (2009). Learning performancewith interactive simulations in medical education: Lessons learned from results of learningcomplex physiological models with the HAEMOdynamics SIMulator. Computers in Education,52(2), 292–301.

Hough, J. R., & Ogilvie, D. T. (2005). An empirical test of cognitive style and strategic decision outcomes.Journal of Management Studies, 42(2), 417–448.

James, W. (1890). Principles of psychology. New York: Holt.Kameda, T., Tindale, R. S., & Davis, J. H. (2003). Cognitions, preferences, and social sharedness: Past,

present, and future directions in group decision making. In S. L. Schneider & J. Shanteau (Eds.),Emerging perspectives on judgment and decision research (pp. 215–240). Cambridge: CambridgeUniversity Press.

Kerawalla, L., Pearce, D., Yuill, N., Luckin, R., & Harris, A. (2008). “I’m keeping those there, are you?”the role of a new user interface paradigm–Separate Control of Shared Space (SCOSS)–in thecollaborative decision-making process. Computers in Education, 50(1), 193–206.

Kerr, N. L., & Tindale, R. S. (2004). Group performance and decision making. Annual Review ofPsychology, 55, 623–655.

Kerr, N. L., MacCoun, R. J., & Kramer, G. P. (1996). Biases in judgment comparing individuals andgroups. Psychological Review, 103, 687–719.

Klein, J. (2005). Effectiveness of school staff meetings: Implications for teacher-training and conduct ofmeetings. International Journal of Research & Methods in Education, 28(1), 67–81.

Kocher, M. G., & Sutter, S. (2005). The decision maker matters: Individual versus group behavior inexperimental beauty-context games. The Economic Journal, 115, 200–223.

Kogut, B., & Zadner, U. (1996). What firms do? coordination, identity and learning. Organization Science,7(5), 502–518.

Komiya, A., Kusumi, T., & Watabe, M. (2007). Regret in individual and group decision making. JapaneseJournal of Psychology, 78(2), 165–172.

Lashley, K. S. (1951). The problem of serial order in behavior. In L. A. Jeffries (Ed.), Cerebral mechanismsin behavior (pp. 112–136). New York: Wiley.

Laurillard, D., Stratfold, M., Luckin, R., Plowman, L., Taylor, J. (2000). Affordances for learning in a non-linear narrative medium. Journal of Interactive Media in Education, 2 [www-jime.open.ac.uk/00/2].

Lehman, R. L., Lempert, R. O., & Nisbett, R. E. (1988). The effects of graduate training onreasoning: Formal discipline and thinking about everyday-life events. American Psychologist, 43,431–442.

Levy, J., Pashler, H., & Boer, E. (2006). Central interference in driving: Is there any stopping thepsychological refractory period? Psychological Science, 17(3), 228–235.

Littleton, K., & Light, P. (1999). Learning with computers, analysing productive interaction. London &New York: Routledge.

Educ Inf Technol (2013) 18:443–458 457

Lombardozzi, C. (2001). Learning journal: Thoughts on learning in organizations http://learningjournal.wordpress.com

Malone, T. W., & Kevin, C. (1994). The interdisciplinary study of coordination. ACM Computing Surveys,26(1), 87–119.

Maturana, H. R., & Varela, F. J. (1980). Autopoiesis and cognition. Dordrecht: Reidel.Meyer, D. E., & Kieras, D. E. (1999). Précis to a practical unified theory of cognition and action:

Some lessons from EPIC computational models of human multiple-task performance. In D. Gopher &A. Koriat (Eds.), Attention and performance XVII (pp. 17–88). Cambridge: MIT.

Michaelson, L. K., Watson, W. E., & Black, R. H. (1989). A realistic test individual versus group consensusdecision making. Journal of Applied Psychology, 74(5), 834–839.

Nijstad, B. A., Stroebe, W., & Lodewijkx, H. F. M. (2006). The illusion of group productivity: A reductionof failures explanation. European Journal of Social Psychology, 36, 31–48.

Pashler, H. (1994). Dual-task interference in simple tasks: Data and theory. Psychological Bulletin, 116,220–244.

Pashler, H., & Johnston, J. (1998). Attentional limitation in dual-task performance. In H. Pashler (Ed.),Attention (pp. 155–189). Hove: Taylor and Francis.

Passig, D. (2007) Melioration as a higher thinking skill to enhance intelligence. Teachers college record.Columbia University. 109(1), 24–50.

Pinto, M., Pinto, J., & Prescott, J. (1993). Antecedents and consequences of project team cross functionalcooperation. Management Science, 39, 1281–1297.

Pirola-Merlo, A., Härtel, C., Mann, L., & Hirst, G. (2002). How leaders influence the impact of affectiveevents on team climate and performance in R&D teams. The Leadership Quarterly, 13(5), 561–581.

Polk, T. A., & Seifert, C. M. (Eds.). (2002). Cognitive modeling. Cambridge: MIT.Porush, J. (1984). On consultants’ negligence or: The best of doctors should go to hell. Israel Law,

12, 119–124.Rosen, C. (2008). The myth of multitasking. The New Atlantis: A Journal of Technology & Society, 20,

105–110.Ruthruff, E., Johnston, J. C., & Van Selst, M. V. (2001). Why practice reduces dual-task interference.

Journal of Experimental Psychology. Human Perception and Performance, 27, 3–21.Ryu, K., & Myung, R. (2005). Evaluation of mental workload with a combined measure based on

physiological indices during a dual task of tracking and mental arithmetic. International Journal ofIndustrial Ergonomics, 35(11), 991–1009.

Scott, S. D., Shoemaker, G. B. D., Inkpen, K. M. (2000). Towards seamless support of natural collaborativeinteractions. In Proceedings of Graphics Interface (GI) 2000. May 2000, Montréal, PQ, pp. 103–110.

Sharan, S. (2002). Differentiating method of cooperative learning in research and practice. Asia PacificJournal of Education, 22, 106–116.

Sharon, M. (1997). Divided attention: A study on high organizing abilities. Tel-Aviv: Gomeh ScientificPublications, Tcherikover Publishers.

Shaw, M. (1976). Group decision dynamics: The psychology of small group behavior. New York:McGraw-Hill Book Company.

Taatgen, N. (2007). The minimal control principle. In W. D. Gray (Ed.), Integrated models of cognitivesystems (pp. 368–379). New York: Oxford University Press.

Teasley, S. D., & Roschelle, J. (1993). Constructing a joint problem space: The computer as a tool forsharing knowledge. In S. P. Lajoie & S. J. Derry (Eds.), Computers as cognitive tools (pp. 229–260).Mahwah: Erlbaum.

Terrace, H. S., & Chen, S. (1991). Chunking during serial learning by a pigeon: III. What are the necessaryconditions for establishing a chunk? Journal of Experimental Psychology. Animal Behavior Processes,17(1), 107–118.

Tetlock, P. (2005). Expert political judgment: How good is it? how can we know?. Princeton UniversityPress.

Vermunt, J. D. (1996). Metacognitive, cognitive and affective aspects of learning styles and strategies: Aphenomenographic analysis. Higher Education, 31, 25–50.

Vygotsky, L. S. (1978). Mind in society. Cambridge, Massachusetts, London, England: Harvard UniversityPress.

Watter, S., & Logan, G. D. (2006). Parallel response selection in dual-task situations. Perception &Psychophysics, 68(2), 254–277.

Yeung, N., & Monsell, S. (2003). Switching between tasks of unequal familiarity: The role of stimulus-attribute and response-set selection. Journal of Experimental Psychology. Human Perception andPerformance, 29(2), 455–469.

458 Educ Inf Technol (2013) 18:443–458