The effects of team environment on attentional resource allocation and cognitive workload

The effects of team environment on attentional resource allocation and cognitive workload
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  The Effects of Team Environment on Attentional ResourceAllocation and Cognitive Workload Matthew W. Miller Auburn University Lawrence J. Groman University of Maryland Jeremy C. Rietschel Maryland Exercise and Robotics Center of Excellence, Veteran’s Health Administration,Baltimore, Maryland Craig G. McDonald George Mason University Seppo E. Iso-Ahola and Bradley D. Hatfield University of MarylandDespite the frequency with which individuals perform in team environments as well as therobust relationship between attentional resource allocation/cognitive workload and perfor-mance, the impact of team environment on attentional resource allocation and cognitiveworkload has only recently begun to be investigated. To address this shortcoming, we useda dual-task paradigm (concurrent auditory oddball discrimination and cognitive-motortasks) and recorded phenomenological reports (National Aeronautics and Space Adminis-tration [NASA]-Task Load Index) to assess attentional resource allocation and cognitiveworkload, respectively, while participants engaged in neutral, adaptive, and maladaptiveteam environments. We observed that individuals performing the cognitive-motor task inmaladaptive team environments exhibited poorer oddball performance relative to neutral(  p  .004,  d   0.836) and adaptive (  p  .01,  d   0.754) team environments. Furthermore,participants reported higher Task Load Index scores in maladaptive team environmentsrelative to neutral (  p    .007,  d     1.229) and adaptive (  p    .002,  d     1.590) teamenvironments. Thus, individuals engaging in adaptive and neutral team environmentsallocated their attentional resources much more efficiently and experienced considerablyless cognitive workload relative to maladaptive team environments. Additionally, individ-uals performing in adaptive team environments exhibited substantially better cognitive-motor task performance relative to neutral (  p  .009,  d   0.931) and maladaptive (  p  .017,  d   1.005) team environments. These results illustrate the importance of (1) avoidingmaladaptiveteamenvironmentssoastopreventteammembersfrominefficientlyallocatingtheir attentional resources and experiencing excessive levels of cognitive workload and (2)generating adaptive team environments to enhance task performance. Keywords:  team environment, attention, cognitive workload Attentional resource allocation and cognitiveworkload impact human performance such thatthe efficient allocation of attentional resourcesand the maintenance of manageable levels of  This article was published Online First November 26, 2012.Matthew W. Miller, Department of Kinesiology, Au-burn University; Lawrence J. Groman, Department of Kinesiology, University of Maryland; Jeremy C. Riet-schel, Maryland Exercise and Robotics Center of Excel-lence, Veteran’s Health Administration, Baltimore,Maryland; Craig G. McDonald, Department of Psychol-ogy, George Mason University; Seppo E. Iso-Ahola,Department of Kinesiology, University of Maryland;Bradley D. Hatfield, Department of Kinesiology andNeuroscience and Cognitive Science Program, Univer-sity of Maryland.The authors thank Tanner Nelson for his assistance inthe piloting of this study.Correspondence concerning this article should be ad-dressed to Bradley D. Hatfield, University of Maryland,2351 School of Public Health Building #255, College Park,MD 20742. E-mail:      T     h     i   s     d   o   c   u   m   e   n    t     i   s   c   o   p   y   r     i   g     h    t   e     d     b   y    t     h   e     A   m   e   r     i   c   a   n     P   s   y   c     h   o     l   o   g     i   c   a     l     A   s   s   o   c     i   a    t     i   o   n   o   r   o   n   e   o     f     i    t   s   a     l     l     i   e     d   p   u     b     l     i   s     h   e   r   s .     T     h     i   s   a   r    t     i   c     l   e     i   s     i   n    t   e   n     d   e     d   s   o     l   e     l   y     f   o   r    t     h   e   p   e   r   s   o   n   a     l   u   s   e   o     f    t     h   e     i   n     d     i   v     i     d   u   a     l   u   s   e   r   a   n     d     i   s   n   o    t    t   o     b   e     d     i   s   s   e   m     i   n   a    t   e     d     b   r   o   a     d     l   y . Sport, Exercise, and Performance Psychology © 2012 American Psychological Association2013, Vol. 2, No. 2, 77–89 2157-3905/13/$12.00 DOI: 10.1037/a0030586 77  cognitive workload are associated with superiorperformance, whereas inefficient attentional re-source allocation and excessive levels of cogni-tive workload are associated with poor perfor-mance (see Hillyard & Kutas, 1983; Navon & Gopher, 1979). Individuals’ attentional resourceallocation and cognitive workload have beenexamined in a number of settings wherein indi-viduals perform tasks in nonteam environments.For example, distracted-driving research has in-vestigated how individuals’ performance of asecondary task while driving reallocates atten-tion away from driving and increases cognitiveworkload (Strayer, Watson, & Drews, 2011). However, humans frequently perform tasksin team environments. Examples of such teamenvironments are apparent in military, law en-forcement, medical, sport, and industrial set-tings. Team environments vary with regard totheir quality such that adaptive team environ-ments, which can be characterized by high lev-els of perceived competence of and trust inone’s teammates, as well as task cohesivenesswith one’s teammates, are associated with su-perior individual performance, whereas mal-adaptive team environments, which can be char-acterized by low levels of these aforementionedcomponents, are associated with poor perfor-mance (Carron, Colman, & Wheeler, 2002;Dirks, 1999; Marcos, Miguel, Oliva, & Calvo,2010).It is surprising that the impact of team envi-ronment on attentional resource allocation andcognitive workload has only recently begun tobe investigated (Stevens, Galloway, Berka, &Sprang, 2009). We believe this area of investi-gation is important given (1) the frequency withwhich humans perform in team environmentsand (2) the strength of the relationship betweenattentional resource allocation/cognitive work-load and performance.The purpose of this study was to determinewhether the quality of team environment wouldinfluence participants’ attentional resource allo-cation and cognitive workload while perform-ing a cognitive-motor task in an adaptive, amaladaptive, and a neutral team environment.To assess the impact of team environment onindividuals’ attentional resource allocation, weobserved each participant’s performance on anoddball auditory discrimination task (Squires,Squires, & Hillyard, 1975) while s/he engagedin a cognitive-motor task in each team environ-ment. To investigate the effects of team envi-ronment on cognitive workload, we evaluatedparticipants’ subjective workload using the Na-tional Aeronautics and Space Administration(NASA)-Task Load Index (TLX) after they en-gaged in the team environments.As maladaptive team environments and inef-ficient attentional resource allocation/cognitiveworkload are generally associated with poorperformance, it was predicted that participantswould exhibit less efficient attentional resourceallocation, higher levels of cognitive workload,and poorer cognitive-motor task performancewhile engaging in the maladaptive team envi-ronment relative to the adaptive team environ-ment. Additionally, it was predicted that whileperforming in a neutral team environment, theefficiency of participants’ attentional resourceallocation, the level of their cognitive workload,and their cognitive-motor task performancewould fall in between those observed in theadaptive and maladaptive team environments. MethodsParticipants Nine right-handed young adults (five women,with an age range of 18–26 years,  M     21.8, SD    2.2 years) recruited from a large Mid-Atlantic university gave informed consent andcompleted the study, which was approved by aninstitutional review board. Cognitive-Motor Task For the cognitive-motor task, participantsused their right hand to play the videogameTetris while the song “Korobeiniki” (“Music A”in the standard Tetris game) was played (72–76dB SPL) from a speaker next to the computermonitor on which Tetris was being played.Tetris asks individuals to manipulate different-shaped game pieces presented on a video screen(in the present case, a computer monitorscreen) to place them in their optimal locationon the game board (monitor screen). Tetriscan be played at various difficulty levels,which are modulated by the velocity at whichthe game pieces move down the game board(e.g., at level 1, the pieces fall at a velocity of 1.67 cm/s, whereas at level 8, the pieces fall ata velocity of 3.56 cm/s). Each participant estab- 78 MILLER ET AL.      T     h     i   s     d   o   c   u   m   e   n    t     i   s   c   o   p   y   r     i   g     h    t   e     d     b   y    t     h   e     A   m   e   r     i   c   a   n     P   s   y   c     h   o     l   o   g     i   c   a     l     A   s   s   o   c     i   a    t     i   o   n   o   r   o   n   e   o     f     i    t   s   a     l     l     i   e     d   p   u     b     l     i   s     h   e   r   s .     T     h     i   s   a   r    t     i   c     l   e     i   s     i   n    t   e   n     d   e     d   s   o     l   e     l   y     f   o   r    t     h   e   p   e   r   s   o   n   a     l   u   s   e   o     f    t     h   e     i   n     d     i   v     i     d   u   a     l   u   s   e   r   a   n     d     i   s   n   o    t    t   o     b   e     d     i   s   s   e   m     i   n   a    t   e     d     b   r   o   a     d     l   y .  lished a Tetris difficulty level and played at thatlevel throughout the experiment. The establish-ment of Tetris level was determined by havingparticipants begin play at level one, five, orseven, dependent on each participant’s re-sponses to a Tetris expertise questionnaire(beginner, advanced beginner, or fairly good,respectively).After beginning play at the appropriate level,participants played until they failed at the task (i.e., the Tetris pieces accumulated to the top of the monitor screen) twice. If a participant com-pleted 10 horizontal lines that contained no gapsbetween the pieces, the current level was com-pleted and the participant advanced to the nextlevel. If a participant advanced to a new leveland then failed at that level, s/he began play atthe level at which s/he failed the first time andcontinued until s/he failed a second time. Thelevel at which the participant failed a secondtime became the level used throughout the re-mainder of the experiment, unless s/he had notcompleted any lines at this level, in which case,s/he played at the previous level. Participants’mean Tetris level was 6.5 with a standard devi-ation of 1.5, indicating that they were playing atmedium levels of difficulty (6 being about half-way between the lowest level, 1, and the highestlevel, 10). Tetris score was determined by giv-ing each participant one point for each horizon-tal line of game pieces completed and subtract-ing five points each time the participant failed atthe task. Oddball Task (Attentional ResourceAllocation) Assessing individuals’ speed and accuracy onthe oddball task while they concurrently per-form another task is an established method of inferring attentional resource allocation (Maclinet al., 2011). Specifically, faster and more ac-curate responses on the oddball task indicatethat attentional resources formerly dedicated tothe other task (in the present case, Tetris) havebeen freed up and reallocated to the oddballtask, thus signifying efficient attentional re-source allocation (see Kahneman, 1973; Maclin et al., 2011).For the oddball task, participants engaged ina difficult version of the oddball paradigm (Tro-che, Houlihan, Stelmack, & Rammsayer, 2009).This version of oddball asked participants to usetheir left hand to press the spacebar on a key-board (different from the one they were using toplay Tetris) every time a target tone (1,000 Hz,275 ms) was played through speakers posi-tioned 75 cm behind the participants and towithhold a response when nontarget tones(1,000 Hz, 200 ms) were played. Three-hundredtones (60 targets and 240 nontargets, 2,000-msinterstimulus interval) were played (92 dB SPL)in each experimental condition. So as to avoidthe possibility of speed–accuracy trade-offs dueto different strategies among the different con-ditions (Fitts, 1954), each participant’s oddball score was determined by summing his or herstandardized (  z -scores calculated across all task conditions) median reaction times (RTs) andstandardized accuracy scores (error rates: falsealarms    missed targets). Thus, oddball scorewas calculated as follows: standardized medianRT    standardized accuracy score, with loweroddball scores indicating better performance(shorter RTs and fewer errors). For a visualdescription of the experimental setup, pleaserefer to Figure 1. NASA-TLX (Cognitive Workload) While the oddball task offers a measure of attentional resource allocation, the NASA-TLXprovides a metric of subjective cognitive work-load, thus offering a convergent measure of cognitive efficiency, but from a phenomenolog-ical perspective instead of a behavioral one(Hart & Staveland, 1988). Additionally, the TLX provides a description of the sources of cognitive workload impacting cognitive effi-ciency (Hart and Staveland). The TLX is aquestionnaire that asks participants to rate theirperceptions of the cognitive workload imposedby a task (or task condition) (Hart and Stave-land). To fill out the questionnaire, participantsrate the magnitudes of six sources of cognitiveworkload (mental demand, physical demand,temporal demand, performance, effort, and frus-tration) on scales from 0 to 100 (100 indicatingthe greatest magnitude), thus providing a scorefor each source. On completing the question-naire, participants make pairwise comparisonsbetween each of the sources of cognitive work-load to indicate which source contributed moreto the overall workload imposed by the task.Each source of cognitive workload is thenweighted by multiplying each source’s score by 79TEAM ENVIRONMENT, ATTENTION, AND COGNITIVE WORKLOAD      T     h     i   s     d   o   c   u   m   e   n    t     i   s   c   o   p   y   r     i   g     h    t   e     d     b   y    t     h   e     A   m   e   r     i   c   a   n     P   s   y   c     h   o     l   o   g     i   c   a     l     A   s   s   o   c     i   a    t     i   o   n   o   r   o   n   e   o     f     i    t   s   a     l     l     i   e     d   p   u     b     l     i   s     h   e   r   s .     T     h     i   s   a   r    t     i   c     l   e     i   s     i   n    t   e   n     d   e     d   s   o     l   e     l   y     f   o   r    t     h   e   p   e   r   s   o   n   a     l   u   s   e   o     f    t     h   e     i   n     d     i   v     i     d   u   a     l   u   s   e   r   a   n     d     i   s   n   o    t    t   o     b   e     d     i   s   s   e   m     i   n   a    t   e     d     b   r   o   a     d     l   y .  the number of pairwise comparisons eachsource “won” (had a greater relative contribu-tion to workload than the source to which it wascompared). The weighted source scores are thensummed together and divided by 15 (the totalnumber of pairwise comparisons made), thusyielding a TLX score between 0 and 100. ThisTLX score indicates the level of subjective cog-nitive workload such that a high score indicatesa high level of workload, whereas a lower scoresignifies a lower level of workload (Hart andStaveland). The TLX is considered to be one of the most effective measures of cognitive work-load and is highly correlated with other subjec-tive cognitive workload metrics (Rubio, Diaz, Martin, & Puente, 2004; Warm, Dember, &Hancock, 1996). Participants completed aTLX immediately after performing in eachteam environment. Experimental Conditions Each participant engaged in four experimen-tal task conditions: Oddball, Neutral Team En-vironment, Adaptive Team Environment, andMaladaptive Team Environment. During theOddball condition, participants engaged in theoddball task while watching Tetris beingplayed. During the Neutral Team Environmentcondition, participants engaged in the Tetris andoddball tasks concurrently in the presence of thetwo teammates (both of whom were male) withwhom they would play/had played in the Adap-tive and Maladaptive Team Environment con-ditions. The task in the Adaptive and Maladap-tive Team Environment conditions was thesame as in the Neutral Team Environment con-dition, but one of the teammates offered theparticipants recommendations on how to ma-neuver the Tetris game pieces, whereas theother teammate was present but did not offerany advice. Henceforth, the teammate who of-fered advice in the Adaptive Team Environmentcondition will be referred to as the “goodteammate,” whereas the teammate who of-fered advice in the Maladaptive Team Envi-ronment condition will be referred to as the“bad teammate.”Previous research has revealed that one’s per-ception of his or her teammates’ competency atperforming a task is strongly and positivelycorrelated with both one’s trust in his or herteammates’ abilities to help him/her success-fully perform a task and one’s reported task cohesiveness with his or her teammates (Marcoset al., 2010; Miller et al., 2011). Accordingly, we sought to alter participants’ perceptions of their teammates’ Tetris aptitudes before partic-ipants’ engagement in the Adaptive and Mal-adaptive Team Environment conditions. Bymodulating participants’ perceptions of their Figure 1.  The experimental setup demonstrating a participant engaging in the cognitive-motor task (Tetris) with his right hand and responding to the oddball task with his left hand.The participant is receiving recommendations from his teammate, who is seated to his right.80 MILLER ET AL.      T     h     i   s     d   o   c   u   m   e   n    t     i   s   c   o   p   y   r     i   g     h    t   e     d     b   y    t     h   e     A   m   e   r     i   c   a   n     P   s   y   c     h   o     l   o   g     i   c   a     l     A   s   s   o   c     i   a    t     i   o   n   o   r   o   n   e   o     f     i    t   s   a     l     l     i   e     d   p   u     b     l     i   s     h   e   r   s .     T     h     i   s   a   r    t     i   c     l   e     i   s     i   n    t   e   n     d   e     d   s   o     l   e     l   y     f   o   r    t     h   e   p   e   r   s   o   n   a     l   u   s   e   o     f    t     h   e     i   n     d     i   v     i     d   u   a     l   u   s   e   r   a   n     d     i   s   n   o    t    t   o     b   e     d     i   s   s   e   m     i   n   a    t   e     d     b   r   o   a     d     l   y .  teammates’ abilities, we were confident theother team environment components of inter-est would likewise be altered, effectively gen-erating adaptive and maladaptive team envi-ronments.We manipulated perceived competence in athree-step process. First, the good teammateinformed participants that he was a much moreexperienced and, therefore, better Tetris playerthan the bad teammate, who acknowledged thisinformation. Second, in the Oddball condition,participants watched the good teammate playTetris for 5.5 min and the bad teammate play for5.5 min. With the participants watching, thegood teammate played Tetris to the best of hisability, attempting to optimize the placement of every game piece. Conversely, the bad team-mate did not play to the best of his ability, as heattempted to optimize the placement of only25% of the game pieces, intentionally misplac-ing the other 75%. Finally, in the AdaptiveTeam Environment condition, the good team-mate offered advice to the best of his ability forevery game piece presented, whereas, in theMaladaptive Team Environment condition, thebad teammate offered advice to the best of hisability for only 25% of the pieces, intentionallygiving nonoptimal advice for the other 75%.As determined before the experiment, theteammates were comparable with regard toTetris ability, so that they could switch whoplayed the role of the good teammate and whoplayed the role of the bad teammate, thus con-trolling for differences beyond the accuracy of the advice they were offering. An additionalattempt was made to control for communicationstyle differences between the two teammates, inthat each teammate offered advice through thesame three hand signals: drawing a circle in theair to indicate that game pieces should be ro-tated, pointing directly on the screen to wheregame pieces should be placed, and giving a“thumbs-up” when game pieces were beingmoved toward the recommended location. Wetold participants that their teammates receivedthe same incentive-based monetary rewardas the participants. We told participants that,given this reward system, they should decidehow much to follow their teammates’ recom-mendations because, if participants found betterplaces for game pieces than their teammates hadrecommended, their teammates would benefitfrom this more optimal placement.To determine whether the experimental ma-nipulations were effective, we asked partici-pants to fill out a questionnaire about eachteammate immediately after playing with thatteammate. The Teammate Questionnaire askedparticipants to use a 5-point Likert scale (high-est scores equal to 5) to respond to one questionregarding how competent they believed theirteammate to be at Tetris, a second questionconcerning how much they trusted their team-mate’s abilities to help them successfully playTetris, and a third question inquiring about thelevel of task cohesiveness they felt with theirteammate (see Appendix for specific questions).This novel and brief questionnaire was usedbecause it asked questions about perceivedcompetence, trust, and task cohesiveness spe-cifically related to Tetris and was minimallyintrusive to the experiment. We were confidentin the construct validity of the questions (Cron-bach & Meehl, 1955) contained in the question-naire, given that pilot data revealed significant(all  p  values    .001) group differences in theexpected directions for responses to the ques-tionnaire (i.e., higher levels of perceived com-petence, trust, and task cohesion in the AdaptiveTeam Environment condition [  M     4.067, SD    0.458;  M     4.133,  SD    0.640;  M    4.333,  SD    0.617, respectively] as comparedwith the Maladaptive Team Environment con-dition [  M     2.667,  SD    0.617;  M     2.800, SD    0.676;  M     2.600,  S  D    0.828, respec-tively]) (Miller et al., 2012). To try and ensure that participants felt comfortable respondinghonestly to the questionnaire, we told them thattheir responses would not be revealed to eitherteammate. Experimental Protocol On entering the testing preparation room,participants completed informed consent and aTetris experience questionnaire (participants’lifetime Tetris experience ranged from havingplayed   10 h to having played   50 h). Aftercompleting the paperwork, we explained theexperimental protocol and introduced the team-mates to the participants. We told participantsthat they were competing against eight otherparticipants in the study. We told them that if their composite score (Tetris score combinedwith oddball score) placed them first among allparticipants, they would receive $40; if they 81TEAM ENVIRONMENT, ATTENTION, AND COGNITIVE WORKLOAD      T     h     i   s     d   o   c   u   m   e   n    t     i   s   c   o   p   y   r     i   g     h    t   e     d     b   y    t     h   e     A   m   e   r     i   c   a   n     P   s   y   c     h   o     l   o   g     i   c   a     l     A   s   s   o   c     i   a    t     i   o   n   o   r   o   n   e   o     f     i    t   s   a     l     l     i   e     d   p   u     b     l     i   s     h   e   r   s .     T     h     i   s   a   r    t     i   c     l   e     i   s     i   n    t   e   n     d   e     d   s   o     l   e     l   y     f   o   r    t     h   e   p   e   r   s   o   n   a     l   u   s   e   o     f    t     h   e     i   n     d     i   v     i     d   u   a     l   u   s   e   r   a   n     d     i   s   n   o    t    t   o     b   e     d     i   s   s   e   m     i   n   a    t   e     d     b   r   o   a     d     l   y .
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks