cognitive processes: effects on perseveration Dimitri van der Linden a,b,*, Michael Frese a,1, a Work and Organizational Psychology, University of Amsterdam, Roetersstraat 15, b Work and Organizational Psychology, University of Nijmegen, Montessorilaan 3, c Experimental and Work Psychology, University of Groningen, Grote Kruisstraat 2/1, Received 4 June 2002; received in revised form 19 November 2002; accepted 4 December 2002 We tested whether behavioural manifestations of mental fatigue may be linked to compro- mised executive control, which refers to the ability to regulate perceptual and motor processesfor goal-directed behaviour. In complex tasks, compromised executive control may becomemanifest as decreased flexibility and sub-optimal planning. In the study we use the WisconsinCard Sorting Test (WCST) and the Tower of London (TOL), which respectively measure flex-ibility (e.g., perseverative errors) and planning. A simple memory task was used as a controlmeasure. Fatigue was induced through working for 2 h on cognitively demanding tasks. Theresults showed that compared to a non-fatigued group, fatigued participants displayed moreperseveration on the WCST and showed prolonged planning time on the TOL. Fatigue did notaffect performance on the simple memory task. These findings indicate compromised executivecontrol under fatigue, which may explain the typical errors and sub-optimal performance thatare often found in fatigued people.
Ó 2002 Elsevier Science B.V. All rights reserved.
* Corresponding author. Address: Section Work and Organizational Psychology, University of Nijmegen, P.O. Box 9104, 6500 HE Nijmegen, The Netherlands. Tel.: +31-24-361-2743; fax: +31-24-361-5937.
E-mail address: (D. van der Linden).
1 The Author has since moved to the University of Giessen.
0001-6918/03/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0001-6918(02)00150-6 D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 PsycINFO classification: 2340Keywords: Mental fatigue; Executive control; Cognitive flexibility; Planning Working on cognitively demanding tasks for a considerable time often leads to mental fatigue, which can impact task performance. In industry, many incidentsand accidents have been related to mental fatigue as the result of sustained perfor-mance (Baker, Olson, & Morisseau, 1994). Therefore, in order to prevent or dealwith fatigue related errors it is important to understand the nature of mental fatigueand its specific effects on behaviour. However, despite many studies on fatigue, it isremarkably difficult to get a grip on mental fatigue and the cognitive processes un-derlying its behavioural manifestations (Broadbent, 1979; Desmond & Hancock,2001; Hockey, 1997; Holding, 1983). The current study was conducted to providesome additional insights into fatigue and its underlying processes. In this study wedefine mental fatigue as a change in psychophysiological state due to sustained per-formance (Desmond & Hancock, 2001; Job & Dalziel, 2001). This change in psycho-physiological state has subjective and objective manifestations, which include anincreased resistance against further effort (Meijman, 2000), an increased propensitytowards less analytic information processing (Sanders, 1998), and changes in mood(Broadbent, 1979; Holding, 1983). Sustained performance, in this definition, doesnot necessarily involve the same task but can extend over different tasks that requiremental effort, such as fatigue because of a day in the office (which often also involvesseveral different tasks).
One of the interesting questions in fatigue research is in what way cognitive con- trol of behaviour changes under fatigue. Some researchers proposed that mental fa-tigue affects those control processes that are involved in the organization of actionsand that play a major role in deliberate and goal-directed behaviour (Bartlett, 1943;Hockey, 1997; Lorist et al., 2000; Sanders, 1998). Already 60 years ago, Bartlett(1943) reported observations that support this Ôcontrol viewÕ of fatigue. Specifically,after more than 2 h of skilled work, pilots in a flight simulator (the famous Cam-bridge Cockpit studies) were still able to perform individual actions well, but itwas the overall organization of these actions that seemed to suffer. Bartlett stated‘‘. . . all the time the general drift in the operator is towards a less closely organizedand effective central control.’’ (p. 256).
Another relevant finding in fatigue research that supports the Ôcontrol viewÕ is that performance on simple or well-learned tasks, which can be executed in a more or lessautomatic way, can be upheld over long periods of time, after sleep deprivation, orafter (mentally) demanding activities. On the other hand, complex tasks that requirethe deliberate control of behaviour are generally difficult to perform under such cir-cumstances (Broadbent, 1979; Hockey, 1993; Holding, 1983; Sanders, 1998).
These typical effects on different levels of information processing that are found in several fatigue studies and the specific disorganization of behaviour that tends to oc-cur under fatigue both suggest that mental fatigue is mainly characterized by deteri- D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 oration of executive control. Executive control refers to the ability to regulate percep-tual and motor processes in order to respond in an adaptive way to novel or chang-ing task demands (Baddeley & Logie, 1999; Miller & Cohen, 2001; Norman &Shallice, 1986). In the current study we examine whether fatigued people indeedshow deficits in task performance that indicate that their executive control on behav-iour is compromised.
Through executive control humans are able to ‘‘. . . transiently couple almost any response to almost any stimulus, even when there are neither innate nor acquiredconnections between stimulus and response.’’ (Goschke, 2000, p. 331).
The literature shows that there is some debate about the nature of executive control processes, for example controlled attention (Engle, Kane, & Tuholski, 1999), inhibi-tion of irrelevant information (Fuster, 1989; Miyake et al., 2000), task set mainte-nance, task set switching (Engle et al., 1999; Fuster, 1989; Miyake et al., 2000;Rogers & Monsell, 1995), and working memory updating (Miyake et al., 2000) haveall been proposed as core aspects of executive control. On the other hand, several re-searchers have also proposed that all these processes have a ‘‘. . . underlying common-ality’’ (Kimberg & Farah, 1993, p. 415), which indicates that there may be some morebasic process responsible for the behavioural manifestations of control. In the currentstudy we use a conceptualization of executive control that is assumed to be such abasic control process (Braver et al., 2001; De Jong, 2000, 2001; Duncan, Emslie,Williams, Johnson, & Freer, 1996; Kimberg & Farah, 1993), namely that the controlof goal-directed behaviour depends on the ability to keep goals and goal-reated infor-mation active in mind (Anderson, 1993; Braver et al., 2001; De Jong, 2000, 2001; Dun-can et al., 1996; Kimberg & Farah, 1993). Goals and goal-related information refer toall information regarding the conditions under which certain actions are executed(some researchers refer to this as task context, Braver et al., 2001; Kimberg & Farah,1993). Such information can be considered a set of end-states and task rules (e.g.,when the task is X, then when A and B are both present, do Y) which, when held ac-tively in mind, can indirectly exert their influence on the selection of actions, therebybiasing behaviour towards goal-attainment (Anderson, 1993; Duncan et al., 1996;Kimberg & Farah, 1993). For clarity, we henceforward refer to the activation levelof goals and goal-related information as goal-activation (Duncan et al., 1996).
During compromised executive control, it is not the mental representation of the goal itself that is affected. Instead it is the activation level through which a goal caninfluence the selection of actions, that is reduced (De Jong, 2000; Duncan et al., 1996;Kimberg & Farah, 1993). During periods of reduced goal-activation, actions areguided by more automatic processes, which are triggered by situational or externalcues, even when this is inappropriate. Duncan et al. (1996) referred to such periodsas goal-neglect, which may be underlying many of the problems of executive control.
In general, Duncan et al. (1996) argued that when executive control is compromised‘‘. . . in different contexts the patient (which has difficulties with executive control)may appear perseverative or distractible, rigid or inappropriate, passive or impulsive D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 and disinhibited’’ (p. 258). We assume that fatigued people display a tendency to re-duce executive control, and consequently show similar deficits in task performance.
1.2. Mental fatigue and executive control There are only few studies that explicitly investigated the effects of mental fatigue from an executive control perspective. One set of studies investigated the effects of fa-tigue on response planning and task switching, both being important aspects of exec-utive control. Lorist et al. (2000) used behavioural and EEG-data to study the effectsof time-on-task (i.e., mental fatigue) on planning and task switching. The EEG-dataof their study showed that with increasing time-on-task there was a reduced involve-ment of those brain areas that are associated with the exertion of executive control(the frontal lobes). This result supported their initial expectations on the effects ofmental fatigue. In addition fatigue led to an increased number of errors and an in-crease in reaction time (Lorist et al., 2000). However, the study did not reveal differ-ential effects of fatigue on switch and non-switch trials, nor did it show effects offatigue on response planning. Thus, at the behavioural level, a specific effect of mentalfatigue on task switching (executive control) was not found.
De Jong (2000) also studied the effects of fatigue (time-on-task) on task switching and response planning. He investigated whether reaction time costs of task switchingwere due to periods in which participants did not engage in response planning eventhough they had the opportunity to do so. Moreover, he assessed whether fatigue in-fluenced the number of periods in which participants did not seem to engage in plan-ning. In this study, fatigue did not affect response planning.
Thus, even though fatigue literature suggests that fatigue affects high-level infor- mation processing, the studies by De Jong (2000) and Lorist et al. (2000) did not un-ambiguously showed an effect of mental fatigue on executive control. There areseveral possible explanations. In the studies by Lorist et al. and de Jong, mental fa-tigue was operationalized as the time spent on the same task. Thus, after some time,participants in the task switching studies (De Jong, 2000; Lorist et al., 2000) mighthave had so much practice that some of the processes of task switching could havebeen executed automatically. However, executive control on behaviour is particu-larly important when a task is novel (Dias, Robbins, & Roberts, 1997; Duncanet al., 1996). For example, Dias et al. (1997) found that inhibition problems in taskswitching were mainly found in situations where a switch was novel and not wellpracticed. Hence, with well-practiced participants it would be much more difficultto detect effects of fatigue on executive control processes. Moreover, in a task-switch-ing paradigm, participants are told exactly what to do, which reduces the need todevelop own strategies and to engage in complex problem solving. However, devel-oping strategies in a complex task and reacting to changes in task circumstances aresituations that typically put heavy demands on executive control (Duncan et al.,1996; Fuster, 1989; Miller & Cohen, 2001) and thus these types of behaviour maybe vulnerable when mentally fatigued.
In the current study we investigated the effects of mental fatigue with a different design. First, we induced fatigue by using tasks that are different from the experimen- D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 tal tasks. Thus, we measured the effects of mental fatigue between-tasks instead ofwithin-tasks. The advantage of this approach is that the tasks after the fatigue ma-nipulation were novel and can be expected to put heavy demands on executive con-trol. Second, we used tasks that are not strongly structured but that required theparticipants to develop their own strategies and to adequately process feedback. Spe-cifically, we expect that in such complex tasks, fatigued people show deficits in twomajor aspects of problem solving that are considered hallmarks of executive control,namely, flexibility and planning (Fuster, 1989; Gazzaniga, Ivry, & Mangun, 1998;Norman & Shallice, 1986; Shallice, 1982). A deficit in flexibility often manifest itselfin behaviour as a tendency to perseverate or to hold on to an ineffective strategy(Milner, 1963; Norman & Shallice, 1986), whereas deficits in planning can be ob-served by a tendency to initiate actions without considering a strategy beforehand,by ineffective plans, or by increased planning time (Oaksford, Frances, Grainger,& Williams, 1996; Owen et al., 1995; Shallice, 1982). To test whether fatigue leadsto these specific changes in task behaviour we used tasks that have been used exten-sively in executive control research, namely, the Wisconsin Card Sorting Test(WCST) (Milner, 1963) and the Tower of London (TOL) (Shallice, 1982).
WCST was used to test whether fatigue concurs with increased perseveration, which would indicate compromised executive control. In the WCST, participantshave to discover how to sort cards that hold geometrical figures. Sorting rules in thistask are based on the colour, shape, or number of figures on the cards. However,because no detailed instructions are given, participants have to discover the sortingrules by themselves through systematic exploration. In the WCST, such explorationis supported by feedback after each trial. Once participants discovered the currentlyactive sorting rule (which in the WCST is operationalized as 10 correct responsesafter each other), the rule changes without notice. Subsequently, participants haveto use the feedback to notice that the sorting rule has changed and then they haveto discover the new sorting rule.
Many clinical studies showed that the most common measures to assess executive control in the WCST are the number of perseverative errors and the number of dis-covered sorting rules (Heaton, 1981; Milner, 1963; Norman & Shallice, 1986; Som-sen, van der Molen, Jennings, & van Beek, 2000). Perseveration in the WCST meansthat people tend to continue applying previous sorting rules that are no longer valid.
In accordance with the conceptualization of executive control we employ in the cur-rent study, Kimberg and Farah (1993) used cognitive modelling to show that persev-eration in the WCST can be ascribed to decreased goal activation. When feedbackinformation about the invalidity of the current sorting rule is not held sufficiently ac-tive in mind, actions continue to be guided by previous sorting rules, which alreadyhad a high activation level (Kimberg & Farah, 1993). As a result of perseverationand the use of inflexible strategies to search for the sorting rule, sub-optimal execu-tive control has also been associated with a low number of discovered sorting rules(Milner, 1963).
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 Besides perseverative errors, there are also other types of non-perseverative errors that people make on the WCST (Heaton, 1981; Milner, 1963). Several underlyingreasons have been proposed for non-perseverative errors. For example, incorrectguessing when trying to discover the sorting rule, or difficulties in holding the currentsorting rule active in mind (lapse in task set maintenance, Hartman, Bolton, &Fehnle, 2001; Paolo, Troster, Axelrod, & Koller, 1995). Studies that also examinednon-perseverative errors in the WCST, showed somewhat mixed results. Some stud-ies report that compromised executive control did not only lead to perseverative er-rors but also to non-perseverative errors. Paolo et al. (1995) reported that in theelderly (who show deficits on the WCST) perseverative and non-perseverative errorswere positively related (r ¼ 0:64). However, in most studies, an increased numberof perseverative errors were the most frequent result (cf. Heaton, 1981). Thus, weexpect that mental fatigue will mainly coincide with increased perseveration.
Somsen et al. (2000) argued that the WCST consists of two qualitative different types of problem solving, namely rule application and rule search. Rule applicationmeans that, once the sorting rule is discovered, participants have to remember bywhich rule to sort and to perform some relatively simple cognitive operations(e.g., match the cards on colour). On the other hand, when the current sorting ruleis unknown, participants have to engage in Rule search, which involves flexible reac-tions to task feedback and conceptualization of new task rules. There are several(psychophysiological) studies showing that rule search put heavy demands on exec-utive control processes, whereas rule application does not (Barcelo, Munoz-Cespedes, Pozo, & Rubia, 2000; Konishi et al., 1999). The response times thatparticipants show during Rule search and rule application may provide additionalinsight into the processes underlying task behaviour. Specifically, several studiesshowed that automatic response selection takes less time than response selectionthat involves executive control (Monsell & Driver, 2000; Sanders, 1998; Shiffrin &Scheider, 1977). Thus, shortened response times during rule search and increasedperseveration would support the idea of compromised executive control during con-ceptualization of sorting rules. On the other hand, prolonged response times andincreased perseveration would indicate that executive control is exerted yet is ineffec-tive in preventing inappropriate actions. In the current study, we analyse responsetimes during Rule search and rule application to examine which pattern of responsetime and errors in the WCST may occur under fatigue.
We used the TOL, (Shallice, 1982) to test planning processes under fatigue. The TOL is a puzzle in which participants have to rearrange coloured beads over pegsuntil they match a goal-state. The TOL measures planning because effective perfor-mance requires goals and sub-goals to be determined before one starts to act (Hodg-son, Bajwa, Owen, & Kennard, 2000; Owen et al., 1995; Shallice, 1982). In addition,the TOL also assesses flexibility as each new TOL-trial requires the development ofnew strategies and the ÔinhibitionÕ of previous strategies that are no longer valid inthe current task context (Hodgson et al., 2000). Because planning and flexibility de- D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 pend on the ability to let behaviour be guided by task goals (Duncan et al., 1996) andbecause the TOL was designed as a relatively pure measure of executive control(Shallice, 1982), we expect that fatigued people show planning deficits on theTOL. Quality of planning is assessed by the combination of reaction time and per-formance measures. An important reaction time measure in the TOL is the time be-tween the presentation of a trial and the first move. This time is generally consideredto reflect planning processes (Anderson, 1993; Hodgson et al., 2000; Owen et al.,1995; Shallice, 1982). In addition, the number of moves to solve a TOL-trial is animportant performance measure that reflects how effective the initial planning was(Oaksford et al., 1996; Shallice, 1982).
In the TOL task, fatigued participants may become more impulsive, meaning that they will minimize or skip planning. However, because of poor planning they wouldneed more moves and may also need more time to solve the trials. In the TOL, suchperformance deficit manifests itself in short first-move times and increased number ofmoves. Similar patterns of performance deficits on the TOL are found in people withfrontal lobe damage, who are impaired on executive control (Goel & Grafman, 1995;Hodgson et al., 2000; Owen et al., 1995; Shallice, 1982). Compared to such groups ofpatients, it can be expected that the effects of fatigue on executive control in normalparticipants are much less severe and may even arise from different underlying (neu-rological) processes. Consequently, we expect that fatigued people may still attemptto plan their behaviour, yet such planning may be inefficient. If this is so, then actualperformance on the TOL in terms of number of errors and problem-solving timemay not show strong deficits yet planning time would be prolonged. Such specificeffects on the TOL are sometimes reported in studies on the effects of mood or instudies on Parkinson disease patients, who are impaired on executive control (Cools,Stefanova, Barker, Robbins, & Owen, 2002; Hodgson et al., 2000; Owen et al., 1995).
Although the main emphasis in the current study is on the WCST and the TOL, we also used a forward digit span as a control measure. Compromised executive con-trol under fatigue implies that not all aspects of cognitive performance are affectedunder fatigue but only those aspects that involve flexibility, planning, and the delib-erate regulation of actions (Norman & Shallice, 1986; Riccio et al., 1994; Shallice &Burgess, 1991). The forward digit span task requires to keep information in mind fora short period and to reproduce that information, which does not heavily rely on ex-ecutive control (Baddeley & Logie, 1999; Norman & Shallice, 1986). Norman andShallice (1986) argued that the digit span is relatively insensitive to compromised ex-ecutive control because the task relies on ‘‘.maintenance rehearsal schemas, which inmost people is a well-learned routine skill’’ (p. 15). Moreover, Kimberg and Farah(1993) argued that simple memory tasks are not affected by compromised executivecontrol because these tests do not involve different sub-sets of goals that may inter-fere with each other (as in the WCST or the TOL). Hence, holding goals and goal-related information in mind and updating this information in the light of changingtask context is not an issue in these tasks.
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 When digit span performance stays unaffected by mental fatigue this indicates that effects of fatigue on the WCST and TOL may not be ascribed to a ÔsimpleÕ in-ability to hold information in mind or to general lack of compliance of fatigued par-ticipants.
Fifty-eight undergraduate college students participated in this study (15 males and 43 females, Mean age ¼ 21 years, SD ¼ 2:5). Participants were paid €20 for sessionsthat lasted approximately 4 h. Participants were randomly assigned to a fatigue(n ¼ 31) or a non-fatigue (n ¼ 27) condition.
2.2.1. The Wisconsin Card Sorting Test (Milner, 1963) We used a computerized version of the WCST in which four target cards were pre- sented at the upper half of the computer screen. These target cards differed from eachother on the sorting dimensions (colour, shape, and number) and remained visible ateach trial. Each trial, a new sorting card was presented at the lower half of the com-puter screen. Participants sorted a card by pressing a button on the keyboard thatcorresponded to a target card (buttons Ô1Õ, Ô2Õ, Ô3Õ, and Ô4Õ on the keyboard). Aftera sorting response, a big plus sign with the word ÔCorrectÕ in it was presented whenthe sort was correct or a big minus sign with the word ÔWrongÕ when the sort wasincorrect. The feedback stayed on the computer screen until participants pressedthe ENTER-button, after which the next card was presented.
When participants had discovered the sorting rule, the rule switched without no- tice. The WCST had six rule-switches built in. Every sorting rule occurred twice. TheWCST ended after a participant discovered all six sorting rules (corrected six time10 correct sorts) or after a maximum of 128 trials.
Dependent measures. Performance measures of the WCST, were rated by the com- puter using the algorithms as proposed by Heaton (1981) and comprised persevera-tive errors, unique errors in which cards were sorted in a way that did not match anyof the sorting dimensions (shape, colour, or number), and miscellaneous errorswhich consists of all errors that were not perseverative or unique errors. In theWCST, unique errors are generally very rare. A large proportion of unique errorsindicate that participants did not adopt a reasoning strategy. Therefore, similarlyto Somsen et al. (2000) we adopted a criterion of 30% unique errors to exclude par-ticipants from further analyses. In the current study, only one participant matchedthis criterion (this participant had 42% unique errors) and was excluded from furtheranalyses.
Rule Application RT was operationalized as the median RT of all trials that fell within a sequence of 10 correct responses plus the first response thereafter. Rule ap- D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 plication was considered a baseline reaction time to perform some simple cognitiveoperations (e.g., match features of the cards).
Rule Search RT was operationalized as the median RT of those trials that pre- ceded a sequence of 10 correct trials (discovered rule).
2.2.2. The Tower of London (Shallice, 1982) The TOL consists of three pegs on which three coloured beads have to be config- ured in the same way as in a given goal-state. We used a computerized version of theTOL in which the pegs and beads were presented in the middle of the screen. Thegoal-state was presented in smaller format at the upper right corner of the screen.
Participants could manipulate beads by dragging and dropping them with themouse. Restrictions during the task were that the maximum number of beads on apeg was determined by the length of a peg. Furthermore, it was not possible to dragbeads that had other beads on top of them. The entire TOL-test consisted of 18 dif-ferent configuration problems (18 trials).
The main dependent measure of the TOL to assess planning is the median First Move RT from the beginning of TOL-trials. This is the time from the initial presen-tation of a TOL-trial to the time of the first response. Other measures to assess TOLperformance were the average number of moves needed to perform the trials, the me-dian time per TOL-trial, the total number of rule violations (e.g., trying to drag abead that has another bead on top).
We used a computerized version of the digit span. Each trial, every second, a digit was presented on the screen. After the presentation of the digits, the participant wasprompted to fill in the digits on the screen. The tasks started out with a four-digitsequence trial. If the participant correctly answered a trial, the next trial consistedof a sequence with one digit more. Otherwise the next trial had an equal numberof digits. The task consisted of 10 trials. The digit span was measured at the begin-ning of the experimental session and right after the manipulation.
Fatigue was induced through a so-called scheduling task on the computer (Taat- gen, 1999). In this task, participants had to assign work to fictional employees. Theduration of the work and the availability of employee hours differed per trial. Fur-thermore, in each trial there was a set of conditions, which had to be fulfilled (e.g.,tasks B and E had to be performed before A). A limited amount of time was avail-able for each scheduling trial, depending on the number of variables and difficulty ofthat trial (time ranged from 5 to 12 min). There was no information on intermediateresults on the computer screen and no external memory aids were allowed; thus thetask required a high degree of mental effort. Moreover, sustained performance onthis task has been shown to induce mental fatigue (van der Linden, Frese, & Sonnen-tag, submitted).
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 Subjective fatigue was measured with the Rating Scale Mental Effort (Zijlstra, 1993) which consists of seven 150-point answer scales that refer to several aspectsof fatigue. The Rating Scale Mental Effort is used as a single measure of fatigue(Mean CronbachÕs Alpha, pre- and post-manipulation measure ¼ 0:91). The RatingScale Mental Effort was filled out before and after the manipulation.
We included a measure of motivation to continue with the experiment and to do oneÕs best in the experimental tasks. We constructed four items in a seven-pointLikert format in which participants were asked how much effort they were willing toput into the tasks and how much they wanted to do their best. The motivation scalewas given directly after the manipulation (CronbachÕs alpha ¼ 0:88).
Because fatigue is generally found to affect mood, we also measured mood states with four sub-scales of the short version of the translated Profile of Mood States(Wald & Mellenbergh, 1990). The sub-scales measured anger, depression, tensionand vigour. The sub-scales of the Profile of Mood States were given before anddirectly after the manipulation.
As an additional control we measured general intelligence to examine whether IQ was related to performance on any of the experimental tasks. Intelligence was mea-sured with RavenÕs (1962) Advanced Progressive Matrices. We used a paper and penversion and gave the participants a maximum of 30 min to work on the test beforethe manipulation.
Participants were tested individually in sessions that lasted about 4 h. At the be- ginning of the session, participants filled out the Rating Scale Mental Effort and theProfile of Mood States. Then they worked respectively, on the digit span task and for30 min on the Advanced Progressive Matrices. The manipulation followed directlyafter the Progressive Matrices. Participants in the fatigue group had to work onthe scheduling task for 2 h (this implied that participants in the fatigue conditionworked on cognitively demanding tasks for 2.5 h: Advanced Progressive Matricesplus scheduling task). The participants in the control group were told they had tobridge 2 h. Within this time they had to stay in the laboratory and were allowedto read some magazines or otherwise spend their time as they wanted (care was takenthat they did not engage in any cognitively demanding tasks such as studying).
After the 2-h manipulation, participants filled out the Rating Scale Mental Effort, Profile of Mood States, and the motivation questionnaire. Then participants worked D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 respectively on the digit span task, the TOL (18 trials), and the WCST. Due to tech-nical reasons we decided not to counterbalance the order of presentation of the task.
Reported fatigue (Rating Scale Mental Effort) was submitted to analysis of vari- ance with time of measurement (before vs. after the manipulation) as a within-subject factor and condition (fatigue vs. not) as between-subject factor. This analysissuggests that our manipulation was successful. We found a significant interaction be-tween time of measurement and condition (F ð1; 55Þ ¼ 42:31, p < 0:001), see Table 1for means. Post-hoc tests showed that the fatigue and non-fatigue groups did not dif-fer on reported fatigue before the manipulation (F ð1; 55Þ ¼ 0:46, p > 0:05) but sig-nificantly differed after the manipulation (F ð1; 55Þ ¼ 17:14, p < 0:001). Moreover,the participants in the fatigue group increased in fatigue after the manipulation(tðpairedÞ ¼ À4:14, p < 0:001), whereas the control group did not differ in theirpre- and post-measures of fatigue (tðpairedÞ ¼ 0:68, p > 0:05).
Willingness to exert effort on the experimental tasks and to do ones best on these tasks, as measured directly after the manipulation, was significantly lower for the fa-tigued participants than for the non-fatigued participants (tð54Þ ¼ 2:53, p < 0:05).
After the manipulation, participants in the fatigue and non-fatigue conditions sig- nificantly differed in feelings of anger (F ð1; 55Þ ¼ 15:07, p < 0:001). With fatiguedparticipants reporting higher levels of anger. Before the manipulation they did notsignificantly differ on anger (F ð1; 55Þ ¼ 1:40, p > 0:05). Before and after the manip-ulation the groups did not significantly differ in levels of tension, depression, andvigour.
The groups did not significantly differ on the RavenÕs Advanced Progressive Ma- trices, which was given before the manipulation (tð56Þ ¼ À0:55, p > 0:05). The mean Table 1Means (and SD) of pre- and post-manipulation measures of subjective fatigue D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 number of correct items for the fatigued group was 24.44 (SD ¼ 3:80) before the ma-nipulation, and for the non-fatigue group 23.93 (SD ¼ 5:76) before the manipula-tion. Thus, any differences in performance on the experimental tasks could not beattributed to pre-existing differences in general intelligence between the groups.
We expected fatigue to coincide with increased perseveration and a low number of discovered sorting rules. The analyses of the WCST confirmed these hypotheses asfatigued participants showed higher percentages of perseverative errors (F ð1; 55Þ ¼5:01, p < 0:05, see Table 2 for the means) and discovered less sorting rules(F ð1; 55Þ ¼ 7:82, p < 0:01) than non-fatigued participants.
Although fatigued participants tended to have higher percentages of miscella- neous and unique errors, these differences did not reached significance levels(F ð1; 55Þ ¼ 3:35, p ¼ 0:08 and F ð1; 55Þ ¼ 3:07, p ¼ 0:09 for unique and miscella-neous errors respectively).
Response RTs for Rule application and Rule search were submitted to an ana- lyses of variance (ANOVA) with type of RT (Rule application vs. Rule Search) aswithin subject factor, and condition (Fatigue vs. not) as between subject factor. Thisanalysis revealed a significant main effect of type of RT (F ð1; 51Þ ¼ 42:50, p < 0:001)(see Fig. 1).
à p < 0:05, Ãà p < 0:01 for differences fatigue vs. non-fatigue group.
a Range from 0 to 6.
b Proportion score (¼ divided by number of trials, max 128).
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 Fig. 1. Means of fatigue and non-fatigue groups on Rule application and Rule search during the WCST.
Post-hoc within-subject t-tests showed that both the fatigue and the non-fatigue group took significantly longer to give a response during Rule search than duringRule application tðpairedÞ ¼ 5:21, p < 0:001, for the fatigue group and tðpairedÞ ¼3:88, p < 0:001, for the non-fatigue group (see Table 2). However, there was alsoa significant interaction between type of RT and condition (F ð1; 51Þ ¼ 5:23, p <0:05). This interaction showed that the increase in RT from Rule application to Rulesearch was less pronounced in the fatigue group compared to the non-fatigue group.
Post-hoc between group comparisons showed that fatigued and non-fatigued partic-ipants did not significantly differ in baseline RT during Rule application(F ð1; 52Þ ¼ 0:51, p > 0:05). However, between-subjects comparison of Rule searchRT in which we controlled for Rule application RT (as covariate) showed thatfatigued participants took significantly less time to respond after rule switches(F ð1; 51Þ ¼ 4:90, p < 0:05). This analysis suggests that the significantly increasednumber of perseveration in the fatigue group (see description of the analyses above)concurred with a decreased time allotted to giving a response during Rule search.
The fatigued and non-fatigued participants did not significantly differ in the aver- age number of moves per TOL-trial (F ð1; 54Þ ¼ 0:73, p > 0:05) or on the mean timeper TOL-trial (F ð1; 54Þ ¼ 0:78, p > 0:05, see Table 2 for the means. Nor were thereany significant differences in the number of rule errors (violations of the rules in theTOL, F ð1; 54Þ ¼ 1:11, p > 0:05). Thus, the level of induced fatigue did not affectplanning accuracy. However, fatigued and non-fatigued participants differed signif-icantly in the mean reaction time for the first move (F ð1; 54Þ ¼ 4:85, p > 0:05), whichreflects initial planning time, with fatigued participants being slower to initiate thefirst move.
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 The digit span was measured before and after the manipulation. We expected fatigue not to affect digit span performance. To test this we submitted digit spanperformance to an ANOVA with time of measurement (before vs. after the manip-ulation) as a within-subject factor and condition (fatigued vs. not) as between-subjectfactor. We found a significant main effect of time of measurement (F ð1; 53Þ ¼ 9:98,p ¼ 0:01), which showed that both the fatigue and non-fatigue groups performedbetter on the second digit span measure than on the first measure. This indicateda learning effect. However, in line with our expectations there was no significant in-teraction between time of measurement and condition (F ð1; 53Þ ¼ 0:04, p > 0:05),showing that the level of induced fatigue did not affect short-term memory perfor-mance.
3.5. Covariance analyses with mood and motivation Because the fatigue and non-fatigue groups did differ significantly in anger and task motivation, additional analyses were conducted to examine whether moodand motivation could explain the observed differences on the experimental tasks.
Therefore we submitted the results of the WCST, TOL, and digit span to additionalanalyses of covariance (ANCOVAs) in which we controlled for anger and motiva-tion. With these analyses, all the main results stayed the same or even became morereliable. With the covariance analyses the fatigue and non-fatigue groups still signif-icantly differed in perseveration (F ð1; 50Þ ¼ 4:10, p ¼ 0:048) and number of discov-ered sorting rules (F ð1; 50Þ ¼ 6:53, p ¼ 0:01). However, unique and miscellaneouserrors did no longer reach marginal significance (respectively, p ¼ 0:24 and 0.52).
These results are in accordance with our expectation that perseveration and numberof discovered sorting rules are the strongest indications of fatigue effects in theWCST. The interaction between response time type (Rule search vs. Rule applica-tion) and condition was also maintained and even showed a marked increased in sig-nificance level (F ð1; 50Þ ¼ 9:0, p ¼ 0:01).
On the TOL, first-move RT differences also stayed significant in the covariance analysis (F ð1; 50Þ ¼ 12, 95, p ¼ 0:001) whereas the other measured did not reachedsignificance nor was there an effect of the covariates on the results of the digit span.
Thus, these analyses showed that group differences on the WCST and TOL could notbe explained by the different scores on the mood and motivation questionnaires.
In this study, we examined whether mental fatigue coincides with compromised executive control. We used the idea that executive control depends on the abilityto hold goals and goal-related information active in mind so that they can exert theirinfluence on the selection of actions (Braver et al., 2001; De Jong, 2000; Duncanet al., 1996; Kimberg & Farah, 1993). We expected compromised executive control D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 under fatigue to become apparent in lowered flexibility and sub-optimal planning.
The overall results support this idea as fatigued participants showed more perfor-mance deficits than non-fatigued participants on tasks that required to flexibly gen-erate and test hypotheses (WCST) and to plan (TOL). There was no differencebetween fatigued and non-fatigued participants in the forward digit span, which doesnot tap executive control processes but relies on the maintenance and reproductionof information (Baddeley & Logie, 1999; Kimberg & Farah, 1993; Norman & Shal-lice, 1986). Fatigued participants did not perform worse on all tasks or task aspectsbut particularly performed worse than non-fatigued participants in those task as-pects that were related to executive control, which makes it unlikely that the resultsare due to general non-compliance or general loss of motivation of the fatigued par-ticipants. In contrast, the results suggest that the deficit in task performance of fa-tigued participants were caused by difficulties in upholding sufficient levels ofexecutive control during the tasks.
4.1. Fatigue and flexible reactions to changing task circumstances In the WCST, lowered flexibility in task behaviour is operationalized as persever- ation. Compared to non-fatigued participants, fatigued participants made signifi-cantly more perseverative errors in the WCST, which implied that they repeatedlytried to sort cards according to a rule that already proved faulty in earlier sortingattempts (Heaton, 1981; Milner, 1963). The goal-activation perspective states thatduring perseveration the representations of task goals themselves may be unaffected,yet their activation may be too low to exert influence on the selection of actions(Duncan et al., 1996; Kimberg & Farah, 1993). This means that people may persev-erate even if they are aware that the current actions are no longer appropriate (DeJong, 2000; Duncan et al., 1996; Kimberg & Farah, 1993). Although we could notdirectly determine to which extent our fatigued participants were aware of their in-appropriateness of actions during perseveration, we can assume that they perceivedthe feedback after a sorting attempt. Namely, after each trial, the computer screenwas completely cleared and obvious feedback (a big plus or minus sign) was pre-sented in the middle of the screen. This feedback stayed on the screen until partici-pants decided to continue with the next trial (by pressing a button). Thus, evenfatigued participants must have noticed that their sorting action had not been suc-cessful; nevertheless, they showed perseveration.
In general, perseveration in the WCST arises from non-cognitive rigid patterns of behaviour and inadequate integration of task feedback for the selection of responses(Heaton, 1981; Norman & Shallice, 1986; Somsen et al., 2000), which may also beresponsible for the low number of discovered sorting rules for fatigued participants(Milner, 1963; Somsen et al., 2000).
Although fatigued participants showed a trend towards an increased number of non-perseverative errors compared to non-fatigued participants the results showedthat perseverative errors were the most reliable effects of fatigue whereas non-persev-erative errors were mainly linked to decreased motivation and increase in anger. Spe-cifically, in analyses in which we controlled for motivation (willingness to do oneÕs D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 best) and mood, trends towards differences in non-perseverative errors between thegroups were no longer present whereas group differences on perseverative errorsstayed significant.
The reaction time data of the WCST gave additional insight into the lowered flex- ibility under fatigue. Specifically, response-time data provided some converging ev-idence that responses of the fatigued participants were more strongly guided byautomatic cognitive processes. For fatigued and non-fatigued participants alike,we found an increase in response RT after a rule switch (Rule search RT). As weargued, rule search puts more demands on executive control than rule application(Barcelo et al., 2000). Thus, it is reasonable to assume that the long Rule searchRT reflects this deployment of executive control processes. However, in Rule search,the median RT of fatigued participants showed a less pronounced increase in re-sponse time than non-fatigued participants. As automatic response selection (basedon previous activated response tendencies) generally demands less time than goal-directed, deliberate response selection (Monsell & Driver, 2000; Shiffrin & Scheider,1977), the RT data indirectly support the idea of insufficient goal-activation.
4.2. Fatigue, planning, and processes underlying task deficits The results on the TOL suggest that fatigued participants were inefficient on the planning aspect of behaviour regulation. This was in accordance with our expecta-tions. Mental fatigue did not affect overall performance on the TOL. The resultsshowed that, compared to non-fatigued participants, fatigued participants did notneed significantly more moves or more time to solve the TOL-trials, nor did they dis-play more violations of TOL-rules. However, we found a significantly prolongedfirst-move RT for fatigued participants. The first-move RT in the TOL reflects initialplanning time (generating sequences of goals and sub-goals, Anderson, 1993; Oaks-ford et al., 1996; Shallice, 1982).
One question that needs to be addressed when considering the total pattern of re- sults in this study (the TOL and WCST results) is why fatigued people showed in-creased perseveration and a less pronounced increase in response times duringRule search in the WCST, yet show unimpaired performance but prolonged plan-ning times in the TOL. We have to note that the traditional TOL and WCST taskthat we used allow the assessment of deficits in flexibility and planning but doesnot allow detailed insight into the processes underlying such deficits. Thus, a conclu-sive answer to this question cannot be provided. Nevertheless, comparison of the re-sults of our study with other studies in which similar patterns of results were foundmay be informative. Specifically, there are studies in which frontal lobe patients andpatients with ParkinsonÕs disease showed different deficits on the TOL yet displayedsimilar impairment on the WCST (Cools et al., 2002; Fournet, Moureaud, Roulin,Naegele, & Pellat, 2000; Hodgson et al., 2000; Owen et al., 1995). Both type of pa-tients are assumed to be impaired on executive control and both groups showincreased perseveration on the WCST compared to control groups (Gazzanigaet al., 1998). However, frontal lobe patients typically show unimpaired (or evenshorted) initial thinking and time yet perform rather poorly on the TOL. This is gen- D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 erally interpreted as a deficit in planning accuracy (Goel & Grafman, 1995; Owenet al., 1995). On the other hand, patients with mild ParkinsonÕs disease perform nor-mal on the TOL, which indicates that their planning accuracy is not impaired. Yet,their initial planning time is prolonged compared to control groups (Owen et al.,1995). Thus, at the level of observable behaviour on the TOL and the WCST, ourresults more closely seem to resemble the pattern of performance deficits in Parkin-sonÕs disease patients than performance deficits of frontal lobe patients. It is impor-tant to note that executive control deficits in ParkinsonÕs disease patients are ascribedto changes in sub-cortical dopamine systems that affect higher cortical levels, such asthe pre-frontal cortex (Cools et al., 2002; Harrison, Stow, & Owen, 2002; Owen et al.,1995). Specifically, in ParkinsonÕs disease patients, nigrostriatal, and to a lesser ex-tent mesocorticolimbic dopamine is depleted.
The similarity between the results of our study with results on the WCST and the TOL in Parkinson disease patients may suggest an important role for dopamine inthe effects of mental fatigue and executive control. Other theories and findings sup-port this idea, for example, recent theories on the biological substrates of executivecontrol state that dopamine plays a major role in the activation (stability) of goalrepresentations (Braver et al., 2001; Cohen & Servan-Schreiber, 1992; Miller &Cohen, 2001; Robbins et al., 2000). Moreover, dopamine activity has been associatedwith intrinsic motivation and response readiness (Tucker & Williamson, 1984), bothconcepts are strongly related to mental fatigue. Finally, it is generally known thatcoffee intake, which enhances dopamine release, reduces both the subjective and ob-jective effects of mental fatigue (Lorist, 1998). Although it would go beyond thescope of the current study to discuss the possible role of dopamine in detail, this ideaposes a direction for future studies on the relationship between mental fatigue andexecutive control.
4.3. Limitations and suggestions for future studies Although the current study provides some insight into the cognitive processes of performance regulation under fatigue there were also some limitations. One of theselimitations relates to the tasks we used. Both the WCST and the TOL are relativelycomplex tasks in which different cognitive deficits can lead to similar manifestationson the tasks (as may be apparent from our discussion of the WCST and TOL in theprevious section). Thus, future studies might want to aim at a more direct assessmentof the processes that are assumed to underlie loss of flexibility and inefficient plan-ning under fatigue. On the other hand, an advantage of using the WCST and theTOL is that the WCST and the TOL have been used in many studies and clinical set-tings to study executive control (Fuster, 1989; Heaton, 1981; Shallice, 1982). More-over, there are many neuropsychological studies that directly showed that these tasksyield activation of brain structure that are deemed to subserve the translation ofgoals into action (Barcelo et al., 2000; Duncan & Owen, 2000).
Another limitation is that we cannot answer specific questions about the motiva- tional issues involved in cognitive performance under fatigue. Executive controlstrongly overlaps with motivation in the sense that adequate control of behaviour D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 is only exerted when some importance is assigned to task goals (Derryberry & Reed,2001; Monsell & Driver, 2000; Tucker & Williamson, 1984). For example, peoplewith impaired executive control seem to lack the ÔdriveÕ to engage in self-directedbehaviour and to initiate actions (Duncan et al., 1996). Such lack of drive or actioninitiation is also typical for fatigued people (Meijman, 2000).
In the current study, fatigued participants reported a higher level of resistance against further effort and had lower task motivation than non-fatigue participants.
However, the results of the current study did not change when we used ANCOVAsin which we controlled for the motivation to perform well on the task. Moreover, weshould not conclude from the motivational (questionnaire) measures that all changesin behaviour under fatigue are caused by deliberate decisions not to comply with taskgoals (Ônot to do oneÕs best). Alternative to the idea of deliberate non-compliance, wecan expect that the loss of intrinsic motivation under fatigue may have caused fa-tigued participants to experience difficulties in the exertion of executive control evenwhen they, at a conscious level, wanted to do well. Hence future studies may want todifferentiate more clearly between ‘‘. . .an involuntary failure to marshal adequateeffort’’ and ‘‘. . .deliberate non-compliance or laziness’’ (Douglas, 1999, p. 106).
Despite the limitations mentioned above, the current study supports the view that compromised executive control underlies behavioural manifestations of mental fa-tigue. Although there are several studies showing that fatigue seems to affect high-level cognitive processes (Hockey, 1997; Holding, 1983; Sanders, 1998), to ourknowledge the current study is one of the first to explicitly approach fatigue froman executive control perspective. Such a perspective has important implications.
For example, compromised executive control under fatigue does not imply that cer-tain basic cognitive processes can no longer be executed at all. Moreover, it also doesnot imply that cognitive processes are fundamentally changed under fatigue. How-ever, from the (goal-activation) view we adopted in the current study, compromisedexecutive control under fatigue does imply a reduced probability that actions will beguided by task goals or by changing task context (Braver et al., 2001; Duncan et al.,1996). Subsequently there would be an increased tendency for more automatic reg-ulatory processes to guide action selection, even when this is inappropriate. Suchlapses in the exertion of executive control may be responsible for the typical slipsof action and intrusion errors that are often found in fatigued people (Hockey,1997; Holding, 1983; Sanders, 1998).
This study was supported by a grant from The Netherlands Concerted Research action ‘‘Fatigue at Work’’ of The Netherlands Organization of Scientific Research(NWO). Furthermore, we would like to thank Riek Somsen of the University of Am-sterdam for the discussion on performance measures in the WCST and for giving us D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 the computerized version of the WCST. We would also like to thank Michiel Kom-pier, Toon Taris, and two anonymous reviewers for useful comments on earlier ver-sions of this manuscript.
Anderson, J. R. (1993). Rules of mind. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Baddeley, A. D., & Logie, R. H. (1999). Working memory: The multiple-component model. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechansims of active maintenance and executive control(pp. 28–61). Cambridge: Cambrige University Press.
Baker, K., Olson, J., & Morisseau, D. (1994). Work practices, fatigue, and nuclear power plant safety performance. Human Factors, 36(2), 244–257.
Barcelo, F., Munoz-Cespedes, J. M., Pozo, M. A., & Rubia, F. J. (2000). Attentional set shifting modulates the target P3b response in the Wisconsin Card Sorting test. Neuropsychologica, 38, 1342–1355.
Bartlett, F. C. (1943). Fatigue following highly skilled work. Proceedings-of-the-Royal-Society B, Vol. 131, Braver, T. S., Barch, D. M., Keys, B. A., Carter, C. S., Cohen, J. D., Kaye, J. A., Janowsky, J. S., Taylor, S. F., Yesavage, J. A., Mumenthaler, M. S., Jagust, W. J., & Reed, B. R. (2001). Context processing inolder adults: evidence for a theory relating cognitive control to neurobiology in healthy aging. Journalof Experimental Psychology: General, 130, 746–763.
Broadbent, D. E. (1979). Is a fatigue test now possible? Ergonomics, 22, 1277–1290.
Cohen, J., & Servan-Schreiber, D. (1992). Context, cortex, and Dopamine: A connectionist approach to behavior and biology in schizoprhenia. Psychological Review, 99, 45–77.
Cools, R., Stefanova, E., Barker, R. A., Robbins, T. W., & Owen, A. M. (2002). Dopaminergic modulation of high-level cognition in PakinsonÕ s disease: The role of the prefrontal cortex revealed byPET. Brain, 125, 584–594.
De Jong, R. (2000). An intention-activation account of residual switch costs. In S. Monsell & J. Driver (Eds.), Control of cognitive processes (Vol. XVIII, pp. 357–376). Cambridge: The MIT Press.
De Jong, R. (2001). Adult age differences in goal activation and goal maintenance. European Journal of Derryberry, D., & Reed, M. A. (2001). A multidisciplinary perspective on attentional control. In C. folk & B. Gibson (Eds.), Attention, distraction, and action: Multiple perspectives on attentional capture (pp.
325–347). London: Elsevier Science.
Desmond, P. A., & Hancock, P. A. (2001). Active and passive fatigue states. In P. A. Hancock & P. A.
Desmond (Eds.), Stress, workload, and fatigue (pp. 455–465). Mahwah, New J ersey: LawrenceErlbaum Associates.
Dias, R., Robbins, T. W., & Roberts, A. C. (1997). Dissociable forms of inhibitory control within prefrontal cortex with an analog of the Wisconsin Card sorting test: Restriction to novel situations andindependence from ‘‘on-line’’ processing. Journal of Neuroscience, 17, 9285–9297.
Douglas, V. I. (1999). Cognitive control processes in attention-deficit/hyperactivity disorder. In H. C.
Quay & A. Hogan (Eds.), Handbook of disruptive behavior disorders (p. 105). New York: KluwerAcademic/Plenum.
Duncan, J., Emslie, H., Williams, P., Johnson, R., & Freer, C. (1996). Intelligence and the frontal lobe: The organization of goal-directed behavior. Cogntive Psychology, 30, 257–303.
Duncan, J., & Owen, A. M. (2000). Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends in Neuroscience, 23, 475–483.
Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual differences in working memory capacity and what they tell us about controlled attention, general fluid intelligence, and functions of theprefrontal cortex. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechansims of activemaintenance and executive control (pp. 102–134). Cambridge: Cambrige University Press.
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 Fournet, N., Moureaud, O., Roulin, J. L., Naegele, B., & Pellat, J. (2000). Working memory functioning in medicated ParkinsonÕs disease patients and the effect of withdrawal of dopaminergic medication.
Neuropsychology, 14, 247–253.
Fuster, J. M. (1989). The prefrontal cortex: Anatomy, physiology, and neuropsychology of the frontal lobe (second ed.). New York: Raven Press.
Gazzaniga, M. S., Ivry, R. B., & Mangun, G. R. (1998). Cognitive neuroscience: The biology of the mind.
Goel, V., & Grafman, J. (1995). Are the frontal lobes implicated in ‘‘planning’’ functions. interpreting data from the tower of Hanoi. Neuropsychologia, 33, 633–642.
Goschke, T. (2000). International reconfiguration and involuntary persistence in task set switching. In S.
Monsell & J. Driver (Eds.), Control of cognitive processes (Vol. XVII). Cambridge: MIT Press.
Harrison, J. E., Stow, I., & Owen, A. M. (2002). ParkinsonÕs disease. In J. E. Harrison & A. M. Owen (Eds.), Cognitive deficitis in brain disorders (pp. 197–215). London: Martin Dunitz.
Hartman, M., Bolton, E., & Fehnle, S. E. (2001). Accounting for age differences on the Wisconsin Card sorting test: Decreased working memory, not inflexibility. Psychology and Aging, 16, 385–399.
Heaton, R. K. (1981). Wisconsin card sorting test manual. Odessa: University of Colorado.
Hockey, G. R. J. (1993). Cognitive-energetical control mechanisms in the management of work demands and psychological health. In A. D. Baddeley & L. Weiskrantz (Eds.), Attention, selection, awarenessand control. A tribute to Donald Broadbent (pp. 328–345). Oxford: Oxford University Press.
Hockey, G. R. J. (1997). Compensatory control in the regulation of human performance under stress and high workload: A cognitive-energetical framework. Biological Psychology, 45, 73–93.
Hodgson, T. L., Bajwa, A., Owen, A. M., & Kennard, C. (2000). The strategic control of gaze direction in the Tower of London task. Journal of Cognitive Neuroscience, 12, 894–907.
Holding, D. (1983). Fatigue. In R. Hockey (Ed.), Stress and fatigue in human performance (pp. 145–164).
Job, R. F. S., & Dalziel, J. (2001). Defining fatigue as a condition of the organism and distinguishing it from habituation, adaptation, and boredom. In P. A. Hancock & P. A. Desmond (Eds.), Stress,workload, and fatigue (pp. 466–475). Mahwah, New Jersey: Lawrence Erlbaum Associates.
Kimberg, D. Y., & Farah, M. J. (1993). A unified account of cognitive impairments following frontal lobe damage: The role of working memory in complex organized behavior. Journal of ExperimentalPsychology: General, 122, 411–428.
Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M., & Miyashita, Y. (1999). Common inhibitory mechanism in human inferior prefrontal cortex revealed by event-related functional MRI.
Brain, 122, 981–991.
Lorist, M. M. (1998). Caffeine and information processing in man. In J. Snel & M. M. Lorist (Eds.), Nicotine, caffeine, and social drinking: Behaviour and brain function. Amsterdam: Harwood AcademicPublishers.
Lorist, M. M., Klein, M., Nieuwenhuis, S., De Jong, R., Mulder, G., & Meijman, T. F. (2000). Mental fatigue and task control: Planning and preparation. Psychophysiology, 37, 1–12.
Meijman, T. F. (2000). The theory of the stop-emotion: On the functionality of fatigue. In D. Pogorski & W. Karwowski (Eds.), Ergonomics and safety for global business quality and production (pp. 45–50).
Warschaw: CIOP.
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function. Annual Review of Milner, B. (1963). Efects of different brain lesions on card sorting. Archives of Neurology, 9, 90–100.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The unity and diversity of executive functions and their contributions to complex ‘‘frontal lobe’’ tasks: Alatent variable analysis. Cognitive Psychology, 41, 49–100.
Monsell, S., & Driver, J. E. (2000). Control of cognitive processes: Attention and performance, XVIII.
Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behavior. In R.
J. Davidson, G. E. Swartz, & D. Shapiro (Eds.), Consciousness and self-regulation: Advances in theoryand research (Vol. 4, pp. 1–18). New York: Plenum Press.
D. van der Linden et al. / Acta Psychologica 113 (2003) 45–65 Oaksford, M., Frances, M., Grainger, B., & Williams, J. M. G. (1996). Mood, reasoning, and central executive processes. Journal of Experimental Psychology: Learning, Memory and Cognition, 22, 476–492.
Owen, A. M., Sahakian, B. J., Hodges, J. R., Summers, B. A., Polkey, C. E., & Robbins, T. W. (1995).
Dopamine-dependent frontstriatal planning deficitis in early ParkinsonÕs disease. Neuropsychology, 9,126–140.
Paolo, A. M., Troster, A. I., Axelrod, B. N., & Koller, W. C. (1995). Construct validitiy of the WCST in normal elderly and persons with ParkinsonÕs disease. Archives of Clinical Neuropsychology, 10, 463–473.
Raven, J. C. (1962). Advanced Progressive Matrices, Set II. London: H. K. Lewis.
Riccio, C. A., Hall, J., Morgan, A., Hynd, G. W., Gonzalez, J. J., & Marshall, R. M. (1994). Executive function and the Wisconsin Card Sorting Test: Relationship with behavioral ratings and cognitiveability. Developmental Neuropsychology, 10, 215–229.
Robbins, T. W., Rogers, R. D., Miller, E. K., Petrides, M., Frith, C., Duncan, J., & Owen, A. M. (2000).
The neural substrate of control. In S. Monsell & J. Driver (Eds.), Control of cognitive processes:Attention and performance (Vol. XVIII, pp. 473–576). Cambridge, MA: The MIT Press.
Rogers, R. D., & Monsell, S. (1995). Cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207–231.
Sanders, A. F. (1998). Elements of human performance. London: Lawrence Erlbaum Associates.
Shallice, T. (1982). Specific impairments in planning. Philosophical Transactions of the Royal Society Shallice, T., & Burgess, P. (1991). Higher-order cognitive impairments and frontal lobe lesions in man. In H. S. Levin, H. M. Eisenberg, & A. L. Benton (Eds.), Frontal lobe function and dysfunction (pp. 125–138). Oxford: Oxford University Press.
Shiffrin, R. M., & Scheider, W. (1977). Controlled and automatic human information processing II: Perceptual learning, automatic attending, and a general theory. Psychological Review, 84, 127–190.
Somsen, R. J. M., van der Molen, M. W., Jennings, J. R., & van Beek, B. (2000). Wisconsin Card Sorting in adolescents: analysis of performance, response time, and heart rate. Acta Psychologica, 104, 27–257.
Taatgen, N. A. (1999). Learning without limits. From problem solving towards a unified theory of learning.
Unpublished dissertation, University of Groningen, Groningen, The Netherlands.
Tucker, D. M., & Williamson, P. A. (1984). Asymetric neural control systems in human self-regulation.
Psychological Review, 91, 185–215.
van der Linden, D., Frese, M., & Sonnentag, S. (submitted). Mental fatigue and exploration when performing a complex computer task: Systematic versus unsystematic behavior.
Wald, F. D. M., & Mellenbergh, G. J. (1990). De verkorte versie van de vertaling van de profile of Mood States (POMS). (The translation of the shortened version of the profile of Mood States). NederlandsTijdschrift voor de Psychologie, 45, 86–90.
Zijlstra, F. R. H. (1993). Efficiency in work behavior: A design approach for modern tools. Delft: Delft



CHM 6230 : Méthodes physiques en chimie de coordination Hiver 2011 : Christian Reber Description de l’annuaire Caractérisation des composés de coordination par des méthodes spectroscopiques, magnétochimiques et électrochimiques. Principes et applications à la détermination de la géométrie moléculaire, de la structure et de la dynamique électronique. Objectifs : S

Precautions Getting quitting right Nicotine is a mental and physical addiction, as powerful as those caused by heroin Nicotine nasal spray - The nicotine is inhaled into and cocaine. As with other addictive drugs, people can experience withdrawal when the persons nose from a pump bottle and absorbed they get less nicotine than they are used to. Symptoms can include irritability,

Copyright © 2010 Health Drug Pdf