PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Umea University Library]On: 26 February 2009Access details: Access Details: [subscription number 781079107]Publisher Psychology PressInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Child NeuropsychologyPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713657840

Sense of Time and Executive Functioning in Children and AdultsMaria Grazia Carelli a; Helen Forman a; Timo Mäntylä a

a University of Umeå, Umeå, Sweden

First Published on: 11 September 2007

To cite this Article Carelli, Maria Grazia, Forman, Helen and Mäntylä, Timo(2007)'Sense of Time and Executive Functioning in Childrenand Adults',Child Neuropsychology,14:4,372 — 386

To link to this Article: DOI: 10.1080/09297040701441411

URL: http://dx.doi.org/10.1080/09297040701441411

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

Child Neuropsychology, 14: 372–386, 2008http://www.psypress.com/childneuropsychISSN: 0929-7049 print / 1744-4136 onlineDOI: 10.1080/09297040701441411

© 2007 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

SENSE OF TIME AND EXECUTIVE FUNCTIONING IN CHILDREN AND ADULTS

Maria Grazia Carelli, Helen Forman, and Timo MäntyläUniversity of Umeå, Umeå, Sweden

A number of patient studies suggest that impairments in frontal lobe functions are associatedwith disorders in temporal information processing. One implication of these findings is thatsubjective experience of time should be related to executive functions regardless of etiology. Intwo experiments, we examined sense of time in relation to components of executive functioningin healthy children and adults. In Experiment 1, children between 8 to 12 years completed sixexperimental tasks that tapped three components of executive functioning: inhibition, updat-ing, and mental shifting. Sense of time was examined in a duration judgment task in whichparticipants reproduced stimulus durations between 4 to 32 s. In Experiment 2, adult partici-pants completed the time reproduction task under varying concurrent task demands. Bothexperiments showed selective effects in that time reproduction errors were related to the inhibi-tion and updating, but not to the shifting, components of executive functioning. However, theobserved effects were modulated by task demands and age-related differences in cognitive com-petence. We conclude that individual differences in executive functioning are only weaklyrelated to time reproduction performance in healthy children and adults.

Most cognitive control functions, including planning, task initiation, and coordination,are time related in that they require compliance with temporal constraints (Fuster, 1993,2002; Ingvar, 1985). For example, Fuster proposed a general theory of prefrontal function-ing in which temporal organization and integration of cognition and behavior plays a centralrole: “The enactment of a goal-directed sequence of actions is a continuous process of tem-poral integration. At the root of this process is the mediation of cross-temporal contingenciesbetween the action plan, the goal, and the acts leading to the goal” (Fuster, 2002, p. 96).

Consistent with this view, patient studies, brain-imaging studies, and neuropharma-cological evidence suggest that cognitively controlled timing is closely related to prefron-tally mediated executive functions (e.g., Barkley, Koplowitz, Anderson, & McMurray,1997; Janowsky, Shimamura, & Squire, 1989; Kerns, McInerney & Wilde, 2001; Lewis &Miall, 2006; Rammsayer, 1999; Willcutt, Doyle, Nigg, Faraone, & Pennington, 2005). Forexample, Barkley et al. (1997) asked school-aged children with attention deficit/hyperac-tivity disorder (ADHD) to reproduce varying time durations. They found that childrenwith ADHD reproduced stimulus durations less accurately than healthy children. Thesefindings were consistent with Barkley’s (1997) model of ADHD and suggest that ADHD-related problems in executive functioning are related to biases in time perception.

This research was supported by a grant from the Swedish Research Council.Address correspondence to Maria Grazia Carelli, Department of Psychology, University of Umeå, S-901

87, Umeå, Sweden. E-mail: Grazia.Carelli@psy.umu.se

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 373

A general implication of these patient studies (see also Toplak, Dockstader, &Tannock, 2006, for an overview) is that individual and developmental differences in exec-utive functions should be related to sense of time regardless of etiology. In other words,healthy individuals with low performance in executive functioning tasks should also makegreater timing errors than better functioning individuals. Furthermore, it is important toexamine these issues in healthy children and adults, because “empirical findings andmethodological issues challenge the etiologic primacy of inhibitory and executive deficitsin ADHD” (Castellanos, Sonuga-Barke, Milham, & Tannock, 2006, p. 116).

Mäntylä, Carelli, and Forman (2007) examined developmental differences in execu-tive functions in relation time-based prospective memory performance. School-aged chil-dren and adults completed executive functioning tasks and a prospective memory task. Inthe latter task, participants indicated the passing of time every 5 minutes, while watching amovie. Executive functioning had selective effects on time-based prospective memoryperformance in that children with low performance in the updating and inhibition tasks ofexecutive functioning checked the clock more frequently and made greater timing errorsthan did children with more efficient control functions. Their findings suggest that indi-vidual and developmental differences in executive functions are related to time-based pro-spective memory performance (see also Kerns, 2000; Mäntylä, 2003; Mäntylä & Nilsson,1997). These findings provided indirect support for the hypothesis that executive controlfunctions mediate time keeping in healthy children and adults.

This hypothesis is also consistent with the attentional-gate model of Block andZakay (1996; Zakay & Block, 2004; see also Block & Zakay, 2006). According to theirmodel, prospective duration judgments depend on attention-demanding processes thatoccur concurrently with the processing of nontemporal information. To the extent that aperson’s attentional resources are limited, fewer resources can be allocated to temporalinformation, and fewer time signals accumulate in the cognitive counter. Duration judg-ments are assumed to reflect the total number of accumulated signals, and experienceddurations decrease as the difficulty of the secondary task increases (Block, 1992; Zakay &Block, 2004; see also Block & Zakay, 1997).

Zakay and Block (2004) reported a study in which they examined the impact ofexecutive control functions on prospective timing. In one experiment, participants wererequired to time the duration of reading sentences that varied in syntactic ambiguity. Con-sistent with the attentional-gate model, resolving syntactic ambiguity reduced attentionalresources for timing, yielding shorter reproductions in the semantic-ambiguity conditionthan in the no-ambiguity condition.

Taken together, past research provides some behavioral evidence for the hypothesisthat executive control functions mediate temporal information processing in healthy chil-dren and adults. However, past work has been based on a unitary view of executive func-tioning (cf. Baddeley, 1996; Duncan, 1995; Norman & Shallice, 1986). From thatperspective, different forms of executive functioning tasks, such as the Stroop task or theWisconsin Card Sorting test, are assumed to reflect the same underlying executive func-tioning construct, which in turn is expected to mediate sense of time. The issue concerningthe unity and diversity of the executive functioning construct is open to debate (Duncan,1995; Salthouse, 2005; Teuber, 1972), but one influential view suggests that executivefunctions reflect independent (but correlated) subcomponents of cognitive control(Miyake, Friedman, Emerson, Witzki, Howerter, & Wager, 2000).

The aim of this study was to examine sense of time in relation to individual anddevelopmental differences in components of executive functioning. To this end, we used a

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

374 M. G. CARELLI ET AL.

latent variable approach (e.g., Miyake et al., 2000; Salthouse, Atkinson, & Berish, 2003),in which school-aged children (Experiment 1) and adults (Experiment 2) completed sixexperimental tasks that tapped three basic components of executive functioning (inhibi-tion of prepotent responses, updating of working memory contents, and mental shifting).Furthermore, participants completed a duration judgment task in which they reproducedstimulus durations between 4 to 32 s. In Experiment 2, adult participants completed thetime reproduction task under varying concurrent task demands.

Our primary hypothesis was that individual differences in executive functioningwould be associated with time reproduction performance, so that participants with diffi-culties in executive functioning tasks would make greater timing errors than participantswith more efficient control functions. We also reasoned that timing ability might be mod-ulated by an overall level of executive functioning, so that children’s timing performance,but not necessarily adult participants’ timing performance, would be related to individualdifferences in executive functioning under low task demands. However, by increasingoverall task demands, the role of individual differences in executive functions might beaccentuated also in cognitively competent adult (undergraduate) participants.

Furthermore, we expected selective effects in that individual differences in theupdating component of executive functioning were expected to play a central role in pro-spective timing performance. Most duration judgment tasks, including time reproduction,involve working memory functions (e.g., maintaining and updating stimulus durationswhile inhibiting task irrelevant information), and these demands increase with increasingstimulus durations. From that perspective, prefrontally mediated working memory func-tions should play a more central role in prospective duration judgments than more posteri-ori control functions, such as mental shifting.

EXPERIMENT 1

Method

Participants. Fifty-one children between 8 to 12 years (mean age=10.2 years,SD=0.96 years) participated in the study. Twenty-six children were between 8 to 10 years(7 boys and 19 girls) and 25 children between 11 to 12 years (7 boys and 18 girls). Partic-ipants were recruited from an elementary public school in Umeå, a medium-sized town inthe North East of Sweden. All children spoke Swedish as a first language. Parental reportsindicated that none of the children had any obvious behavioral or educational problems.Parental consent for participation was obtained for all children.

Task Characteristics

Duration judgments were based on a time reproduction task, in which a simplestimulus picture (“smiley”) appeared on the computer screen for 4, 8, 12, 24, and 32 s,respectively (see also Meaux & Chelonis, 2003). Except for three practice items (3, 5, and10 s), each duration was presented twice, in the same random order for both blocks. Aftereach stimulus presentation, participants reproduced its duration on the computer screen bypressing a designated key. The experimenter did not provide any specific feedback duringthe task.

Executive functioning was assessed by using six tasks that were assumed to tap theinhibition, updating, and shifting components of the construct. Inhibition of prepotent

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 375

responses was assessed with the Stroop and stop signal tasks. In the stop signal task, theletters X or O appeared on the computer screen, and participants indicated the identity ofthe stimulus letter by pressing a specific key as fast as possible. They were also instructedto withhold a response when they heard a beep (i.e., a stop signal) immediately after thetarget letter. A fixation point appeared on the screen 1000 ms before target presentation,and the stop signal appeared between 400 to 600 ms after the target (see also Logan, 1994;Salthouse et al., 2003), and responses faster than 400 ms were not included in the analysis(< 5%). Task performance was examined in terms of stop signal reaction time, omissionerrors, and comission errors, but only the error data are reported here.

The Stroop task (Stroop, 1935) comprised 36 colored rectangles and color words.Participants read aloud the colors of the rectangles (neutral condition), followed by the listof color words, printed in incongruent colors (inconsistent condition). For each partici-pant, the difference in reading times between the neutral and incongruent conditions wasthe primary measure of the Stroop task.

Task shifting was based on two experimental tasks, referred as the connections andcategory fluency tasks, respectively. The connections task (see also Salthouse et al., 2000)consisted of three pages with 49 circles containing two colors (pink and yellow) and num-bers. In the same condition, participants drew lines between items from the same category(i.e., only between numbers or between colors) as quickly as possible. In the different con-dition, they connected the items by alternating between two sequences (e.g., 1–yellow,2–pink, 3–yellow, etc.). A difference in completion time between the same and differentconditions was assumed to reflect shifting ability.

In the category fluency task (see also Parkin, Hunkin, & Walter, 1995), participantsfirst generated instances for two separate categories (“animals” and “things to eat ordrink”) during a one-minute period, followed by a combined condition in which they gen-erated instances by alternating between the two categories (e.g., cat, pizza, horse,milk,…). Participants provided verbal responses that were recorded for subsequent analy-sis. For both shifting tasks, the primary dependent measure was the difference in numberof responses between the same and mixed conditions.

The updating component of executive functioning was assessed with the n-back andmatrix monitoring tasks. In the n-back task, the experimenter presented a capital conso-nant, with one item on each page of a booklet. The pages were turned at an even pace, atthe rate of 2–3 s per item. The children tapped the page when they recognized a letter asthe same as the one two-items back. The task comprised four sequences of 25 letters, withseven targets in each sequence.

In the matrix monitoring task (Salthouse et al., 2003), a 3 × 3 matrix appeared on acomputer screen, with black dot in one of the nine cells. The matrix disappeared after 3 s,followed by a sequence of three arrows that indicated the movement of the dot in the (imag-inary) matrix. Finally, the matrix reappeared on the computer screen with the dot in one ofthe cells. Participants were instructed to decide whether the position of the dot was the sameor different by pressing a specific key on the computer keyboard. Children completed 12trials, and number of incorrect responses was the primary measure of the matrix monitoringtask. Number of incorrect responses was the dependent measure of both updating tasks.

Procedure

Participants were tested individually during two 40-minute sessions in a quiet roomat their elementary school. During the first session, each child completed three executive

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

376 M. G. CARELLI ET AL.

functioning tasks (Stroop, 2-back, and connections, respectively), followed by the timereproduction task. During the second session, about 7–10 days later, they completed theremaining three executive functioning tasks (stop signal, matrix monitoring, and categoryfluency, respectively). Some of the participants also completed a prospective memory taskin the final phase of the experiment, but these data are not reported here. The order of taskcompletion was the same for all participants. For each task, the experimenter attempted toverify that the child had understood the instructions, and in case of difficulties, the specifictask instruction was further clarified. Furthermore, all the tasks included a practice phase,during which the experimenter illustrated the task instructions.

Results and Discussion

Time Reproduction Data. Duration judgments were examined in terms of abso-lute and relative errors. The former measure, referred to as the absolute error score,reflects timing errors, regardless of their direction. The latter measure, referred to as therelative error score, was obtained by dividing each participant’s time reproduction by thetime duration of the sample interval presented on that trial. This measure provides astandard score across the different time intervals, with coefficients above 1.0 reflectingoverproductions, and coefficients below 1.00 reflecting underproductions.

Figure 1 shows the timing data as a function of stimulus duration and type of mea-sure. Consistent with previous studies, the magnitude of absolute errors increased acrossstimulus duration, F(1, 50)=29.30, MSe=1108536, p < .01. Contrast test indicated signifi-cant effects of the linear, F(1, 50)=68.72, MSe=1785789, p < .01, and quadratic,F(1, 50)=5.82, MSe=1221325, p < .02, components. Furthermore, the relative error datashowed a significant main effect of stimulus duration, F(1, 50)=2.77, MSe=0.08, p < .05.As can be seen in Figure 1, children’s reproductions were rather well calibrated, withminor underproductions for all durations (> 0.96), except for the shortest duration (< 0.93,F(1, 50)=7.36, MSe=0.08, p < .01). Contrast tests showed that this condition was signifi-cantly different from the remaining stimulus durations, with no significant differencesamong them.

Executive Functioning Data. Descriptive statistics for the six executive func-tioning tasks are summarized in Table 1. These data show reasonable distributions, andthe subsequent analyses were based on nontransformed data. For all six measures, a largevalue refers to poor task performance.

The executive functioning data were submitted to a principal component analysis(promax rotation). These analyses yielded a two-component solution. As can be see inTable 2, the two inhibition tasks (stop signal and Stroop) and the two updating tasks(matrix monitoring and n-back) constituted Factor 1 (referred to as the Supervision com-ponent), and the two shifting tasks constituted Factor 2 (referred to as the Shifting compo-nent). These two factors were the only ones with eigenvalues greater than one, and thecorrelation between the two factors was −.14. It should be noted that positive factor scores(i.e., individual z scores) reflected low task performance for both constructs.

Executive Functioning and Duration Judgments. To examine individualand developmental differences in executive functioning in relation to duration judgments,each participant’s factor scores were related to time reproduction data. In these analyses,time reproduction performance was based on absolute and relative errors. Table 3 shows

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 377

Figure 1 Time reproduction performance as a function of stimulus duration and type of measure in Experiment 1.

0

500

1000

1500

2000

2500

3000

3500

Abs

olut

e E

rror

0.75

0.80

0.85

0.90

0.95

1.00

1.05

Relative E

rror

Absolute errorRelative error

4 sStimulus Duration

32 s24 s12 s8 s

Table 1 Descriptive Statistics for the Executive Functioning Tasks in Experiments 1 and 2.

Experiment 1 (n=51) Experiment 2 (n=53)

Task M (SD) Skewness M (SD) Skewness

Stroop 30.04 (10.79) 0.77 12.80 (5.33) 0.63Stop signal 9.16 (7.03) 0.96 3.61 (3.15) 1.16Matrix monitoring 1.06 (1.31) 1.26 4.17 (1.55) 0.41n-back 5.64 (3.52) 1.21 1.33 (1.47) 1.59Connections 2.80 (3.83) −0.08 9.59 (4.62) −0.16Category fluency 2.77 (3.81) −0.10 4.20 (6.88) −0.30

Table 2 Factor Loadings for the Principal Factor Analysis of the Inhibition, Updating, and Shifting Tasks ofExecutive Functioning in Experiments 1 and 2.

Experiment 1 Experiment 2

Task Supervision Shifting Supervision Shifting

Stroop .61 .14 .78 .34Stop signal .74 −.21 .72 −.02Matrix monitoring .53 −.03 .61 −.20n-back .69 −.32 .53 .08Connections −.08 .79 −.07 .80Category fluency −.10 .77 .14 .65

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

378 M. G. CARELLI ET AL.

that the Supervision component of executive functioning correlated significantly withabsolute timing error. This result suggests that children with low performance in theupdating and inhibition tasks made greater absolute timing errors than did children withmore efficient control functions. These effects were independent of age in that the partialcorrelation between Supervision and absolute timing error was significant even after con-trolling for age (r=.30, p < .05). However, separate analyses of short (4 and 8 s) and long(24 and 32 s) durations revealed nonsignificant correlations for short durations (r=.14, p <.35) and a marginally significant correlation for longer durations (r=.29, p < .06). Further-more, as shown in Table 3, Shifting was not related to timing error. As shown in Table 3,relative errors were not related to age or executive functioning.

We also carried out a multiple regression analysis with absolute timing error as aregressor and age, Supervision, and Shifting, respectively, as predictors. This analysisshowed that only Supervision contributed significantly to timing error (beta=.34, p < .05),suggesting that children with good performance in the updating and inhibition tasks (butnot in the shifting tasks) reproduced stimulus durations more accurately than children withless efficient control functions.

EXPERIMENT 2

The results of Experiment 1 indicated that school-aged children’s time reproductionperformance is mediated by individual differences in executive control functions. Chil-dren with difficulties in executive functions made greater timing errors than did childrenwith more efficient control functions. More important, individual differences in executivefunctions had selective effects on timing performance in that children with difficulties inupdating and inhibition functions, but not in mental shifting, made greater errors thanchildren with more efficient working memory functions.

To extend the generality of these findings, we examined the relation between dura-tion judgments and executive functioning in young adults. By focusing on a selectedgroup of university students we attempted to clarify whether time reproduction perfor-mance would be related to individual differences in cognitive control functions in cogni-tively competent adults. A possibility exists that cognitive timing is related to executivefunctions at a lower (e.g., developmentally early) level of competence. Thus, adult partici-pants’ timing errors might not be related to individual differences in executive functioningin that participants with low performance in executive functioning tasks would reproducestimulus durations with the same accuracy as participants with more efficient controlfunctions. However, these effects might be accentuated under more demanding task

Table 3 Pearson Correlation Coefficients for Age, Executive Functioning, and Timing Error in Experiment 1.

Measure 1 2 3 4 5

1. Age – −.40** .13 −.04 −.102. Supervision .40** – −.13 .32* .083. Shifting .13 −.13 – −.04 −.144. Absolute error −.04 .32* −.04 – −.065. Relative error −.10 .08 −.14 −.06 –

*p < .05, **p < .01.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 379

conditions. To examine this hypothesis, we manipulated concurrent task load in Experi-ment 2. Specifically, participants also completed the time reproduction task under twotask-load conditions. In the study-load condition, they observed stimulus durations whilecounting backwards, followed by a test phase in which they reproduced each observedstimulus duration (i.e., load during study only). In the test-load condition, they reproducedstimulus durations while counting backwards (i.e., load during test only).

Method

Participants. Fifty-three Umeå university undergraduates (27 females and 26males) participated in the experiment for payment (approximately US$6). They werebetween 20 to 28 years of age (mean age=24.3 years, SD=3.05 years), and none of theparticipants had prior experience of similar experiments.

Tasks and Procedure. Participants were tested individually during two 40-minutesessions in a quiet room at the university campus. During the first session, participantscompleted three executive functioning tasks and a time reproduction task (without a con-current task). During the second session, about one week later, they completed the threeremaining executive functioning tasks and a time reproduction task under concurrent taskload. The latter task was completed both under study and test load. Specifically, in thestudy-load condition, participants started counting backwards in steps of three (e.g., 291,288, 285, etc.) before the stimulus (smiley) appeared on the computer screen. At test, theyreproduced stimulus durations without concurrent task load. In the test-load condition,participants started counting before the smiley appeared in the reproduction phase. In bothconditions, the experimenter confirmed that the participant counted correctly before pre-senting each stimulus. The experimenter clarified the task instructions during a practicephase in which participants observed and reproduced stimulus durations while countingbackwards. In addition to the practice phase, participants completed two sets of five dura-tions (4, 8, 12, 24, and 32 s, respectively). The durations appeared in a random order inboth sets. Order of task load was counterbalanced, so that half of the participants startedwith study-load condition and the remaining participants started with the test-load condition.

The executive functioning tasks were similar to those of Experiment 1, except thatthe adult participants completed the original version of the connections task (Salthouseet al., 2000), and that the n-back task involved or computerized (E-Prime) version of thetwo-back task. Furthermore, in the matrix monitoring task, participants were presentedwith two 4 × 4 matrices, followed by two parallel sequences of arrows. As in Experiment1, all the executive functioning tasks included a practice phase, during which the experi-menter illustrated the task instructions.

Results and Discussion

The executive functioning and timing data were analyzed in the same way as inExperiment 1. Principal component analysis of the executive functioning data yieldedagain a two-component solution (see Table 1). As in Experiment 1, the two inhibitiontasks (stop signal and Stroop) and the two updating tasks (matrix monitoring and n-back)constituted Factor 1 (Supervision), and the two shifting tasks constituted Factor 2(Shifting). These two factors were the only ones with eigenvalues greater than one, and thecorrelation between the two factors was .08.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

380 M. G. CARELLI ET AL.

Figure 2 summarizes the time reproduction data as a function stimulus duration andtask condition. As in Experiment 1, the magnitude of absolute errors increased acrossstimulus duration, F(4, 188)=84.38, MSe=641154603, p < .01. Furthermore, as can beseen in Figure 2, concurrent task load during study and test had similar effects on absoluteerrors. Specifically, the two concurrent task load conditions produced greater absoluteerrors than the no-load condition, F(2, 94)=38.89, MSe=798060160, p < .01, andthese effects were accentuated for longer durations, F(8, 376)=6.74, p < .01. Contrasttests showed that the task × duration interaction was significant for the linear × linear,F(1, 48)=19.58, p < .01, and quadratic × linear, F(1, 48)=13.71, p < .01, components,respectively.

Figure 3 shows the relative error data as a function of stimulus duration and taskcondition. Again the main effects of task condition, F(2, 94)=34.50, MSe=0.23, p < .01,and duration, F(4, 188)=17.78, MSe=0.08, p < .01, and their interaction, F(8, 376)=7.59,MSe=0.09, p < .01, were significant. Contrast tests indicated that the main effect of taskcondition was due to a significant difference between the study-load and test-load condi-tions, F(1, 48)=40.47, p < .01, whereas their differences with the no-load condition wereonly marginally significant (ps < .10). Contrast tests of the task condition × duration inter-action showed significant effects for the linear × linear, F(1, 48)=43.94, p < .01, andlinear × quadratic, F(1, 48) = 6.39, p < .01, components, respectively. As shown in Figure2, these data suggest that the shorter duration were overestimated, whereas the longerdurations were underestimated in both load conditions. Compared with the no-load condi-tion, only the two longest durations produced significantly greater underestimations in thestudy-load condition. By contrast, in the test-load condition, the reproduction of two long-est durations was relatively well calibrated and not significantly different from the no-loadcondition. Although this pattern of bias is similar to Vierordt´s law (Allan, 1979; Vierordt,1868), it is reasonable to assume that this bias is mediated by the nature of the concurrenttask.

Figure 2 Absolute timing error as a function of stimulus duration and task condition in Experiment 2.

0

2000

4000

6000

8000

10000

Stimulus Duration

Abs

olut

e E

rror

No loadStudy loadTest load

4s 32s24s12s8s

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 381

As shown in Figures 1 and 2, both children (Experiment 1) and adult participants(the no-load condition of Experiment 2) showed similar patterns of timing error. A sepa-rate 2 (age) x 5 (stimulus duration) mixed ANOVA confirmed this observation by yieldingno other effects than significant main effects of stimulus duration. In other words, therewere no age-related differences in time reproduction performance, measured both in termsof absolute or relative errors.

To examine executive functioning in relation to duration judgments, each partici-pant’s factor scores were related to time reproduction performance, measured in terms ofabsolute and relative errors. As can be seen in Table 4, these data suggest that individualdifferences in executive functioning were related to time reproduction performance inadult participants. However, these effects were limited in that only the Supervision com-ponent of executive functioning was related to absolute errors in the study-load condition.In other words, participants with low performance in the inhibition and updating tasksmade greater absolute errors than participants with more efficient control functions, butthese effects were only observed for absolute errors in the study-load condition. Consis-tent with Experiment 1, the Shifting component of executive functioning was not related

Figure 3 Relative timing error as a function of stimulus duration and task condition in Experiment 2.

0.50

0.70

0.90

1.10

1.30

1.50

1.70

Stimulus Duration

Rel

ativ

e E

rror

No loadStudy loadTest load

4s 32s24s12s8s

Table 4 Pearson Correlation Coefficients for Age, Executive Functioning, and Timing Error in Experiment 2.

Absolute Error Relative Error

Condition Age Supervision Shifting Age Supervision Shifting

No load .14 .07 .14 −.06 .02 .16Study load −.03 .28* −.06 −.03 .15 −.04Test load −.15 .11 −.09 −.10 .16 −.09

*p < .05.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

382 M. G. CARELLI ET AL.

to absolute or relative errors. We also examined these data separately for long and shortdurations. These analyses showed no significant effects, except that the correlationbetween long durations and supervision scores was marginally significant in the study-load condition (r=.23, p < .10). Finally, as shown in Table 4, relative error scores were notrelated to the Supervision and Shifting components of executive functioning.

GENERAL DISCUSSION

The aim of the study was to examine duration judgments in relation to components ofexecutive functioning in school-aged children and adults. The main findings of Experiment 1showed that children with difficulties in executive functions made greater timing errors thanchildren with more efficient control functions. However, these effects were selective in thatchildren with difficulties in updating and inhibition functions, but not in mental shifting, madesomewhat greater errors for long durations than did children with better updating and inhibi-tion functions.

In Experiment 2, we attempted to extend the generality of these findings by involv-ing young adults as participants. We reasoned that a strong version of the hypothesisshould imply that sense of time is related to individual differences in executive functions,not only within a limited range of variability (e.g., due to development or impairment infrontal lobe functions) but also within a selected group of cognitively competent individuals(e.g., university students).

The results of Experiment 2 did not support this hypothesis in that individual differ-ences in executive functions were not related to time reproduction performance under nor-mal (i.e., no-load) test conditions. Furthermore, a similar pattern of results was observedunder more demanding (test-load) conditions. Finally, individual differences in executivefunctions were related to time reproduction performance only in that the Supervision com-ponent of executive functioning was related to absolute errors in the study-load condition.

It should also be noted that these marginal effects were limited to absolute timingerrors in both experiments. By contrast, relative errors were not related to individual ordevelopmental differences in executive functioning in healthy children or adults. This pat-tern of results is inconsistent with the hypothesis that ADHD patients make greater under-estimations than healthy individuals because they have difficulties in executive controlfunctions (e.g., Barkley et al., 1997; see also Castellanos et al., 2006; Toplak et al., 2006).

In both experiments, the analyses of the executive functioning data yielded a two-component solution in which the inhibition and updating tasks constituted one factor andthe two shifting tasks the second factor. These findings are consistent with the notion thatmental shifting is a distinct component of executive functioning (Baddeley, 1996; Lehto,Juujärvi, Kooistra, & Pulkkinen, 2003; Miyake et al., 2000). Our findings are also consis-tent with neuropsychological studies (Gehring & Knight, 2002) and neuroimaging find-ings (see Collette & Van der Linden, 2002, for a review) suggesting that parietal areasplay a more basic role in shifting processes than prefrontal areas.

However, this pattern of results is inconsistent with the three-component model ofMiyake et al. (2000). Although our study involved relatively small samples of partici-pants, and a large study might have produced a different outcome, it should be emphasizedthat updating and inhibition are related constructs. Most updating tasks involve inhibitionin that they require revising the items held in working memory by replacing less relevantinformation with newer, more relevant information (Jonides & Smith, 1997). Consistentwith this view, the findings of Miyake et al. (2000) supported a three-component model of

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 383

executive functioning, but the correlation between their updating and inhibition constructswas .63 (see Lehto et al., 2003, for similar correlations for school-aged children; see alsoPennington, 1994; Van der Sluis, Dolan, & Stoel, 2005).

Taken together, our findings suggest that updating and inhibition functions areinvolved in duration judgments, but that these effects were limited and modulated by taskdemands and age-related differences in overall level of cognitive competence. In school-aged children (with less developed control functions than university students), variabilityin executive functioning tasks was related to absolute, rather than relative, timing error forlonger stimulus durations.

Adult participants’ timing errors were not related to individual differences in execu-tive functioning in that participants with low performance were able to reproduce (both shortand long) stimulus durations with the same accuracy as did participants with more efficientcontrol functions. This pattern of results suggests that time reproduction performance is onlyweakly mediated by individual differences in executive functioning and that these effectsreflect a combination of task-specific and subject-specific (developmental) effects.

The present study is consistent with the attentional-gate model of prospective timing(Block & Zakay, 1996, 2006; Zakay & Block, 2004) in that children with limited atten-tional resources made greater timing errors than children with more efficient control func-tions. Furthermore, counting backwards while encoding durations increased themagnitude of underestimations (of longer durations) in Experiment 2. Although our find-ings are consistent with the attentional-gate model, the predictions of the model are ratherunspecific in that any task manipulation that increases attentional demands is assumed tomodulate duration judgments (i.e., the amount of temporal signals that are accumulated inthe counter). Thus, the model is neutral with respect to qualitative (rather than quantita-tive) differences in executive functioning: Two equally demanding tasks that reflect dif-ferent aspects of executive functioning (e.g., task shifting and inhibition) are expected tohave similar effects on experienced durations.

By contrast, our findings showed selective effects in that children with low perfor-mance in the updating and inhibition, but not in the shifting, tasks made greater timingerrors than better performing children. These selective effects of executive functioningmay reflect overall task demands as well as more specific demands on temporal mainte-nance of information. Concerning overall task demands, one might argue that the shiftingtasks were less demanding than the updating and inhibition tasks. In the absence of anindependent index of task load it is not meaningful to compare these tasks in terms of taskdifficulty or complexity (i.e., to argue that, for example, the n-back task was moredemanding than the connections task, or vice versa). However, it should be noted that chil-dren and adults completed different versions of tasks. Yet, separate analysis of children’sand adults’ data showed similar patterns of selective effects.

Concerning more specific demands on temporal maintenance of information, work-ing-memory-related updating and inhibition tasks can be assumed to play a central role intemporal processing. Compared with shifting tasks, most updating tasks require mainte-nance of dynamic event information, which provides a temporal coherence for theobserved event (Mäntylä et al., 2007). Furthermore, updating of working memory repre-sentation is assumed to involve “temporal tagging” to keep track of old and no longer rel-evant information (Jonides & Smith, 1997), and this retaining a sequence of events inworking memory is assumed to lead a sense of temporal continuity.

This notion is also consistent with evidence suggesting that the dorsolateral prefron-tal cortex is integral to both cognitive timing and working memory. For example, Lewis

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

384 M. G. CARELLI ET AL.

and Miall (2006) suggested that cognitively controlled timing is based on the same dorso-lateral prefrontal cells that are known to be involved in working memory (see alsoRammsayer, 1997, 1999). Their perspective is somewhat different from more traditionalmodels of the scalar expectancy theory, including the attentional-gate model. Instead ofmerely keeping track of the progress of a separate time keeper (i.e., an accumulator pro-cess that collects quantized ticks from a hypothetical neural pacemaker), these workingmemory processes might actually constitute the time-dependent process itself.

The present findings suggest that the time reproduction task and related psycho-physical tasks of interval timing are rather insensitive to individual and developmental dif-ferences in higher order cognitive control functions. Although the time reproduction taskis more “cognitive” than most tasks of motor timing (Lewis & Miall, 2006; Rammsayer,1999), its demands on executive control functioning are still rather low. Clinical reportsand observations suggest that individuals with ADHD (and related frontal lobe disorders)have marked difficulties in temporal information processing. However, compared to thedominating psychophysical methods of interval timing these observations and anecdotesappear to refer to rather different time frames and contexts. Our findings suggest thatexperimental and clinical assessment of temporal dysfunctions should be based on tasksthat reflect the complexities of everyday cognition.

Original manuscript received February 26, 2007Revised manuscript accepted May 8, 2007First published online September 11, 2007

REFERENCES

Allan, L. G. (1979). The perception of time. Perception & Psychophysics, 26, 340–354.Baddeley, A. (1996). Exploring the central executive. The Quarterly Journal of Experimental

Psychology, 49A, 5–28.Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive functions: Con-

structing a unifying theory of ADHD. Psychological Bulletin, 121, 65–94.Barkley, R. A., Koplowitz, S., Anderson, T., & McMurray, M. B. (1997). Sense of time in children

with ADHD: Effects of duration, distraction, and stimulant medication. Journal of the Interna-tional Neuropsychological Society, 3, 359–369.

Block, R. A. (1992). Prospective and retrospective duration judgment: The role of information pro-cessing and memory. In F. Macar, V. Pouthas, & W. J. Friedman (Eds.), Time, action and cog-nition: Towards bridging the gap (pp. 141–152). Dordrecht: Kluwer Academic.

Block, R. A., & Zakay, D. (1996). Models of psychological time revisited. In H. Helfrich (Ed.),Time and mind (pp. 171–195). Kirkland, WA: Hogrefe & Huber.

Block R. A., & Zakay, D. (1997). Prospective and retrospective duration judgments: A meta-analyticreview. Psychonomic Bulletin & Review, 4, 184–197.

Block, R. A., & Zakay, D. (2006). Prospective remembering involves time estimation and memoryprocesses. In J. Glickson & M. Myslobodsky (Eds.), Remembering on time: The case for atime-based prospective memory (pp. 25–49). New York: World Scientific Publishing.

Castellanos, F. X., Sonuga-Barke, E. J. S., Milham, M. P., & Tannock, R. (2006). Characterizingcognition in ADHD: Beyond executive dysfunction. Trends in Cognitive Science, 10, 117–122.

Collette, F., & Van der Linden, M. (2002). Brain imaging of the central executive component ofworking memory. Neuroscience & Biobehavioral Reviews, 26, 105–125.

Duncan, J. (1995). Attention, intelligence, and the frontal lobes. In M. S. Gazzaniga (Ed.), The cog-nitive neurosciences (pp. 721–733). Cambridge, MA: MIT Press.

Fuster, J. (1993). Frontal lobes. Current Opinion in Neurobiology, 3, 160–165.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

TIME CONTROL IN CHILDREN AND ADULTS 385

Fuster, J. (2002). Physiology of executive functions: The perception-action cycle. In D. T. Stuss &R. T. Knight (Eds.), Principles of frontal lobe function. New York: Oxford University Press.

Gehring, W. J., & Knight, R. T. (2002). Lateral prefrontal damage affects processing selection butnot attention switching. Cognitive Brain Research, 13, 262–279.

Ingvar, D. H. (1985). Memory of the future: An essay on the temporal organization of consciousawareness. Human Neurobiologia, 31, 503–524.

Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Memory and metamemory: Comparisonsbetween patients with frontal lobe lesions and amnesic patients. Psychobiology, 17, 3–11.

Jonides, J., & Smith, E. E. (1997). Working memory: A view from neuroimaging. Cognitive Psy-chology, 33, 5–42.

Kerns, K. A. (2000). The CyberCruiser: An investigation of development of prospective memory inchildren. Journal of the International Neuropsychological Society, 6, 62–70.

Kerns, K. A., McInerney, R. J., & Wilde, N. J. (2001). Time reproduction, working memory, andbehavioral inhibition in children with ADHD. Child Neuropsychology, 7, 21–31.

Lewis, P., & Miall, R. C. (2006). Remembering the time: A continuous clock. Trends in CognitiveSciences, 10, 401–406.

Lehto, J. E., Juujärvi, P., Kooistra, L., & Pulkkinen, L. (2003). Dimensions of executive functioning.British Journal of Developmental Psychology, 21, 59–80.

Logan, G. D. (1994). On the ability to inhibit thought and action: A user’s guide to the stop signalparadigm. In D. Dagenbach & T. H. Carr (Eds.), Inhibitory processes in attention, memory, andlanguage (pp. 189–239). San Diego, CA: Academic Press.

Mäntylä, T. (2003). Assessing absentmindedness: Prospective memory complaint and impairment inmiddle-aged adults. Memory & Cognition, 31, 15–25.

Mäntylä, T., & Carelli, M. G., & Forman, H. (2007). Time monitoring and executive functioning inchildren and adults. Journal of Experimental Child Psychology, 96, 1–19.

Mäntylä, T., & Nilsson, L.-G. (1997). Remembering to remember in adulthood: A population-basedstudy. Aging, Neuropsychology, and Cognition, 4, 81–92.

Meaux, J. B., & Chelonis, J. J. (2003). Time perception differences in children with and withoutADHD. Journal of Pediatric Health Care, 17, 64–71.

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: A latent variable analysis. Cognitive Psychology, 41, 49–100.

Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behavior.In R. J. Davidson, G. E. Schwartz, & D. Shapiro (Eds.), Consciousness and self-regulation:Advances in research and theory (Vol. 4, pp. 1–18). New York: Plenum.

Parkin, A. J., Hunkin, N. M., and Walter, B. M. (1995). Relationships between normal aging, frontallobe function, and memory for temporal and spatial information. Neuropsychology, 9, 304–312.

Pennington, B. F. (1994). The working memory function of the prefrontal cortices. In M. M. Haith,J. B. Benson, R. J. Roberts Jr., & B. F. Pennington (Eds.), The development of future-orientedprocesses (pp. 243–289). Chicago, IL: University of Chicago Press.

Rammsayer, T. H. (1997). Are there dissociable roles of the mesostriatal and mesolimbocortical dopam-ine systems on temporal information processing in humans? Neuropsychobiology, 35, 36–45.

Rammsayer, T. H. (1999). Neuropharmacological evidence for different timing mechanisms inhumans. The Quarterly Journal of Experimental Psychology, B52, 273–286.

Salthouse, T. A. (2005). Relations between cognitive abilities and measures of executive function-ing. Neuropsychology, 19, 532–545.

Salthouse, T. A., Atkinson, T. M., & Berish, D. E. (2003). Executive functioning as a potentialmediator of age-related cognitive decline in normal adults. Journal of Experimental Psychology:General, 132, 566–594.

Salthouse, T. A., Toth, J., Daniels, K., Parks, C., Pak, R., Wolbrette, M., & Hocking, K. J. (2000).Effects of aging on efficiency of task switching in a variant of the Trail Making Test. Neurop-sychology, 14, 102–111.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009

386 M. G. CARELLI ET AL.

Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of ExperimentalPsychology, 18, 643–662.

Teuber, H.-L. (1972). Unity and diversity of frontal lobe functions. Acta Neurobiologiae Experimen-talis, 32, 615–656.

Toplak, M. E., Dockstader, C., & Tannock, R. (2006). Temporal information processing in ADHD:The findings to date and new methods. Journal of Neuroscience Methods, 151, 15–29.

Van der Sluis, S., Dolan, C. V., & Stoel, R. D. (2005). A note on testing perfect correlations in SEM.Structural Equation Modeling, 12, 551–577.

Vierordt, K. V. (1868). Der Zeitsinn nach Versuchen (pp. 4–191). Tubingen: H. Laupp.Willcutt, E. G., Doyle, A. E., Nigg, J. T., Faraone, S. V., & Pennington, B. F. (2005). Validity of the

executive function theory of Attention-Deficit/Hyperactivity Disorder: A meta-analytic review.Biological Psychiatry, 57, 1336–1346.

Zakay D., & Block, R., A. (2004). Prospective and retrospective duration judgments: An executive-control perspective. Acta Neurobiologiae Experimentalis, 64, 319–328.

Downloaded By: [Umea University Library] At: 15:21 26 February 2009