Journal of Experimental Child Psychology 114 (2013) 295–305

Contents lists available at SciVerse ScienceDirect

Journal of Experimental ChildPsychology

journal homepage: www.elsevier .com/locate/ jecp

Anticipatory action planning increases from 3 to 10 yearsof age in typically developing children

Marjolein Jongbloed-Pereboom a,⇑, Maria W.G. Nijhuis-van der Sanden b,Nicole Saraber-Schiphorst b, Céline Crajé a,1, Bert Steenbergen a

a Behavioural Science Institute, Radboud University Nijmegen, 6500 HE Nijmegen, The Netherlandsb Department of Rehabilitation, Pediatric Physiotherapy, Scientific Institute for Quality of Healthcare, Radboud UniversityNijmegen Medical Centre, 6500 HB Nijmegen, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history:Received 4 May 2012Revised 20 August 2012Available online 29 September 2012

Keywords:Action planningAnticipatory planningChildChild, preschoolReliabilityValidity

0022-0965/$ - see front matter � 2012 Elsevier Inhttp://dx.doi.org/10.1016/j.jecp.2012.08.008

⇑ Corresponding author. Fax: +31 24 361 6211.E-mail address: m.jongbloed-pereboom@pwo.ru

1 Current address: Research Group in Health CareNetherlands.

The primary aim of this study was to assess the development ofaction planning in a group of typically developing children aged3 to 10 years (N = 351). The second aim was to assess reliabilityof the action planning task and to relate the results of the actionplanning task to results of validated upper limb motor perfor-mance tests. Participants performed an action planning task inwhich they needed to grasp an object (a wooden play sword) andplace it into a tight-fitting hole. Our main dependent variablewas the grip type that participants used; that is, we measuredwhether initial grip was adapted in such a way that childrenreached a comfortable posture at the end of the action (the end-state comfort effect). Older children planned their actions moreoften in line with the end-state comfort effect compared withyounger children. Test–retest and interrater reliability of the actionplanning task were good, with intraclass correlation coefficients(ICCs) of .90 and .95, respectively. We compared the action plan-ning task with manual dexterity tests in a subset of participants(n = 197). We found a marginal relation with the manual dexteritytests, indicating that the action planning task measures differentprocesses. In sum, our study showed that action planning increasesfrom 3 to 10 years of age and that the experimental task we used isreliable in assessing anticipatory planning. Therefore, it may beused as a reliable additional test to investigate the degree to which

c. All rights reserved.

.nl (M. Jongbloed-Pereboom).and Nursing, Hanze University of Applied Sciences, 9714 CE Groningen, The

296 M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305

motor behavior is affected at the cognitive level of anticipatoryplanning.

� 2012 Elsevier Inc. All rights reserved.

Introduction

Action planning is an important feature of manipulative skills. Action planning can be defined asthe ability to take task demands of the movement and goal of the action into account when first takinghold of an object (e.g., Johnson-Frey, McCarty, & Keen, 2004). For example, when coffee needs to bepoured into a cup that is upside down on a shelf, most individuals will grasp the cup with the thumbdown. Consequently, at the end of the task, after the cup is turned, the thumb is pointing upward.Thus, the action starts with an uncomfortable posture in order to end the movement in a comfortableposture. This effect has been labeled the end-state comfort effect (Cohen & Rosenbaum, 2004; Rosen-baum, Vaughan, Barnes, & Jorgensen, 1992) and exemplifies that the end of the task is taken into ac-count when first taking hold of the cup, even when this necessitates an initial uncomfortable posture.Ending in a comfortable posture allows more precision to be exerted at the end of the task (Short &Cauraugh, 1999). Importantly, the grip types used to start and end movements can thus be used asdependent variables to assess action planning (Crajé, Aarts, Nijhuis-van der Sanden, & Steenbergen,2010; Rosenbaum, Meulenbroek, Vaughan, & Jansen, 2001).

The end-state comfort effect is a well-known phenomenon in adults (e.g., Cohen & Rosenbaum,2004; Rosenbaum et al., 2001; Short & Cauraugh, 1999). In children, however, research on the devel-opment of the end-state comfort effect is limited, and the results are equivocal. In two studies, nodevelopmental effect was found in children aged 2 to 6 years (Adalbjornsson, Fischman, & Rudisill,2008; Manoel & Moreira, 2005). Adalbjornsson et al. (2008) asked two groups of children (aged 2–3and 5–6 years) to pick up a cup that was placed upside down. Children needed to turn it over and pourwater into the cup. Only a few children displayed the end-state comfort effect, but there were no dif-ferences between the two age groups with respect to the number of children that used the end-statecomfort effect. In line with this, Manoel and Moreira (2005) did not find a developmental effect in 2- to6-year-olds in a bar grasping task. Smyth and Mason (1997) used a handle rotation task and a bargrasping task to assess action planning in 4- to 8-year-olds and found a developmental effect onthe first task but not on the second task. In contrast to these studies, more recently in older childrenan age effect was shown by an increased occurrence of the end-state comfort effect (Janssen & Steen-bergen, 2011; Stöckel, Hughes, & Schack, in press; Thibaut & Toussaint, 2010; Van Swieten et al., 2010;Weigelt & Schack, 2010). Three of these studies used a bar grasping task in 3- to 10-year-olds (Stöckelet al., in press; Thibaut & Toussaint, 2010; Weigelt & Schack, 2010), whereas Van Swieten et al. (2010)used a handle rotation task and did find an age effect in three age groups: 5 to 8 years, 9 to 14 years,and adults. Janssen and Steenbergen (2011) used a continuous measure of end-state comfort whereparticipants needed to grasp a cylinder and transport it to a platform of which the height could bemanipulated. In this study, an age effect on action planning was observed in 7- to 12-year-olds. Insum, the majority of studies show developmental trends in action planning.

In the studies described above, different experimental paradigms were used to assess action plan-ning. Not all of these tasks are suitable to add in an upper limb performance assessment in a clinicalsetting. Therefore, recently we developed a simple and quick-to-administer task to measure actionplanning in a study in children with unilateral cerebral palsy aged 3 to 6 years (Crajé et al., 2010).The task was specifically designed to be interesting and playful for children. It consists of swinginga small wooden sword into a tight-fitting hole in a wooden block. No age effect for action planningwas observed for the patient group. However, an age effect was observed for the control group of typ-ically developing children, although action planning probably had not yet reached an adult level at theage of 6 years.

The first aim of the current study was to validate the previously observed developmental trends inaction planning in a large group of typically developing children aged 3 to 10 years (N = 351). Based on

M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305 297

the previous studies, we expected a developmental trend in action planning, observed as a steady in-crease of the use of comfortable end postures in children in this age range.

The second aim of this study was to assess the test–retest and interrater reliability of the actionplanning task and to relate the results of the action planning task to results of validated upper limb mo-tor performance tests: the Box and Block Test for manual dexterity (Jongbloed-Pereboom, Nijhuis-vander Sanden, & Steenbergen, in press; Mathiowetz, Federman, & Wiemer, 1985; Mathiowetz, Volland,Kashman, & Weber, 1985) and the subtests of the Movement Assessment Battery for Children, secondedition, Dutch version (MABC-2 NL), which assess manual dexterity (Henderson, Sugden, Barnett, &Smits-Engelsman, 2010). By comparing the action planning task with these two manual dexterity tests,we were able to examine the relation between action planning and gross and fine manual dexteritytasks (divergent validity).

Method

Participants

Participants were 351 children aged 3 to 10 years. To assess hand dominance, children were askedto draw a boy or girl on a piece of paper, and 87% had a right hand preference. This corresponded to thepercentage of right-handed people in the general population (Perelle & Ehrman, 1994). Children wererecruited from five primary schools and three playgroups in The Netherlands. The schools and play-groups were selected by convenience in terms of logistics and contacts with the schools/playgroupsbut were representative of the general population for socioeconomic status and ethnicity. Exclusioncriteria were medical conditions and/or diagnosed behavioral problems and uncertainty about handdominance. In Table 1, the number of children per age group is shown subdivided into the numberof boys and girls in each age group.

Assessments

Action planningAction planning was assessed by the task that was previously described in Crajé et al. (2010). This

task was especially designed to measure action planning in children and was presented as a playfulgame. The task was to pick up a wooden sword (length 18.0 cm, width 2.0 cm, height 1.2. cm, handlelength 9.5 cm) from the table and subsequently place it into a tight-fitting hole in a wooden block(27.0 � 13.0 � 13.0 cm, hole 2.0 � 0.8 cm). As shown in Fig. 1A, the sword could be placed in six dif-ferent starting orientations. Four of these orientations served as the control orientations, and twoserved as the critical orientations. In these critical orientations, children needed to sacrifice comfortof the start posture in order to be able to end the task in a comfortable posture on fitting the swordinto the tight-fitting hole (Orientations 2 and 3 for the right hand, Orientations 5 and 6 for the lefthand). Children were instructed to pick up the sword with a power grip (i.e., a whole hand grip)and subsequently place it in the tight-fitting hole in the wooden block (Fig. 1B). The experiment

Table 1Numbers of boys and girls as a function of age.

Age (years) Total Boys Girls

3 27 13 144 36 16 205 44 23 216 50 22 287 52 26 268 46 21 259 51 24 27

10 45 28 17

Total 351 173 178

Fig. 1. (A) Experimental setup of the action planning task, with the sword on starting Orientation 3 in this particular trial.Starting orientations were numbered 1 to 6 in a clockwise direction. The sword orientation with the blade toward the targetlocation was designated as Orientation 1. (B) Example of a required end orientation.

298 M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305

always started with an ‘‘easy’’ trial that did not require any sword rotation (Orientation 1). After suc-cessful completion of this trial, the experiment started. Every orientation was repeated three times ina random order, resulting in a total of 18 trials per child. The task was performed with the dominanthand. The percentage of comfortable end postures in both the critical orientations and the control ori-entations served as the dependent variable of interest.

Manual dexterityA convenient sample of children of three schools and one playgroup participated in the validity

assessments (n = 197). In this subset, all age groups were represented. We aimed to include a mini-mum of 20 children per age group, with a similar number of boys and girls. Only in the youngest group(3 years) did we have slightly fewer children (n = 19 for the MABC-2 NL, n = 15 for the Box and BlockTest) in the study. Manual dexterity was tested with the Box and Block Test according to the instruc-tions in the test manual (Jongbloed-Pereboom et al., in press; Mathiowetz, Volland, et al., 1985). In thistask, participants were required to move as many blocks as possible from one box to the other in 60 s.The test merely measures gross manual dexterity. The type of grip that children use is free of choice;that is, blocks can be picked up with all fingers of the hand but can also be grasped with a pincer grip.Both hands were tested separately, starting with the dominant/preferred hand. The number of blockstransferred in 60 s served as the dependent variable (Mathiowetz, Volland, et al., 1985). Fine manualdexterity was tested with the MABC-2 NL (Henderson et al., 2010). For 3- to 6-year-olds (Age Band 1)the MABC-2 NL contained the subtests Posting Coins, Threading Beads, and Drawing Trail 1. For 7- to10-year-olds (Age Band 2), it contained the subtests Placing Pegs, Threading Lace, and Drawing Trail 2.The standard scores of the subtests and of the total manual dexterity score of the MABC-2 NL werecalculated for each participant according to the instructions in the test manual. Both the Box and BlockTest and the MABC-2 NL have good validity and reliability (Henderson et al., 2010; Jongbloed-Pere-boom et al., in press; Mathiowetz, Volland, et al., 1985).

ReliabilityTest–retest reliability was determined by reassessing the action planning task by the same tester

after 1 week at one school for a subset of the children (n = 20) aged 5 to 10 years. Test–retest reliabilityof the action planning task was excellent, with an intraclass correlation coefficient (ICC) score of .90.There were no significant differences between the first and second assessments.

Interrater reliability was assessed by comparing scores of three raters: one test assistant who vid-eotaped the assessment of 21 children, with the assessment of the videotape by two independent rat-ers. These 21 videos were pseudo-randomly chosen from the videos of the subset of 197 children (�1in 10 videos) who participated in the validity analysis, taking into account that all ages were repre-sented in this subgroup. Interrater reliability was excellent, with an ICC score of .95.

M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305 299

For 141 children, we also tested performance on the action planning task with the nondominanthand. However, we found no differences in the percentage of comfortable end postures in the criticalor control conditions for the dominant and nondominant hands (tested with the Wilcoxon signed-rank test); therefore, these data are not further discussed.

Procedure

Three testing stations were set up in a classroom: one with the Box and Block Test, one with theMABC-2 NL, and one with the action planning task. Three children were tested at the same time bythree assessors, and children changed stations in between tests. Children faced the wall of the class-room so that they could not see each other. Each child was seated on furniture that was suitable for hisor her height and that was installed such that the child sat comfortably and the forearms rested on thetable. The assessment of the action planning task was videotaped for off-line analysis. Total assess-ment duration was approximately 25 min per child.

Data analysis

Differences between boys and girls, and between right- and left-handed children, were analyzedwith independent t tests for normally distributed data (manual dexterity tasks) and with a Mann–Whitney U test for data lacking a normal distribution (action planning task). A Kruskal–Wallis testwas used to analyze the age effect of the action planning task; differences between age groups werefurther analyzed with Mann–Whitney U tests. Results of dominant and nondominant hands perform-ing the Box and Block Test were analyzed with two one-way analyses of variance (ANOVAs) with age,gender, and hand dominance as fixed factors. Post hoc tests of differences between age groups wereperformed using least significant difference (LSD) analyses. Differences of the mean standard scoreof the MABC-2 NL between the norm group and our group were tested with a one-sample t test,and differences in distribution of the standard score were tested with a v2 test. Differences betweenboys and girls were analyzed with independent-samples t tests. The correlations between the actionplanning task and the manual dexterity tests were tested with Spearman’s correlation coefficients.Differences with a p value of less than .05 were considered as statistically significant.

Results

Development of action planning

The mean percentages of comfortable end postures in the action planning task as a function of ageare shown in Fig. 2. First, we found an age effect for the percentage of comfortable end postures in thecritical orientations, as evidenced by the finding that older children more often showed the end-statecomfort effect, H(7) = 29.06, p < .001. As can be seen in Fig. 2, the age effect steadily increased between3 and 10 years of age. Post hoc analysis (LSD) further revealed significantly more critical orientationswith a comfortable end posture in 4-year-olds compared with 3-year-olds, z = �2.21, p < .05, r = .27,and in 10-year-olds compared with 9-year-olds, z = �2.24, p < .05, r = .22. Second, results in the criticalorientations were different for boys than for girls. In the total group, the girls were shown to end moreoften with a comfortable end posture than boys in the critical orientations, z = �2.80, p < .01, r = .15.Third, we also found a small, but significant, age difference in the control orientations, H(7) = 14.67,p < .05. This effect was significant between 8 and 9 years of age, z = �2.20, p < .05, r = .22, and between9 and 10 years of age, z = �2.46, p < .05, r = .24. The 9-year-olds ended a control orientation less oftenwith a comfortable end posture than both the 8- and 10-year-olds.

Manual dexterity

Results of the Box and Block Test are shown in Table 2. Scores of 4 children in the youngest agegroup are missing because these children were unable to follow instructions properly. Older children

Table 2Results of the Box and Block Test as a function of age.

Age (years) Box and Block Test score dominant hand

n M SD Range

3 15 24.2 7.4 15–394 22 36.8 8.1 16–495 21 40.5 7.0 27–566 22 45.9 7.1 28–577 35 51.1 4.9 44–648 27 56.2 7.0 39–679 21 57.4 7.0 45–71

10 30 57.9 5.4 48–68

Total 193 48.1 12.0 15–71

Fig. 2. Results of the action planning task. The percentages of comfortable end postures are shown for the control condition(Orientations 1, 4, 5, and 6 for right-handed children, Orientations 1, 2, 3, and 4 for left-handed children) and for the criticalcondition (Orientations 2 and 3 for right-handed children, Orientations 5 and 6 for left-handed children). Error bars represent95% confidence intervals.

300 M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305

transferred significantly more blocks in 60 s than younger children when using their dominant hand,F(7,167) = 44.92, p < .001, r = .70, and their nondominant hand, F(7,167) = 57.51, p < .001, r = .72. Posthoc analysis (LSD) revealed significant age effects (p < .05) for the scores in the dominant and nondom-inant hands between all age groups of the 3- and 8-year-olds except for the score on the dominanthand between the 4- and 5-year-olds (p = .07). Between 8 and 10 years, the age effect was no longerobserved. Results were similar for boys and girls and for right- and left-handed children.

Results of the MABC-2 NL manual dexterity score are shown in Table 3. The mean manual dexteritystandard score was slightly, but significantly, higher in our sample (11.0) than in the Dutch normgroup (10.0), t(194) = 5.08, p < .001, r = .34, whereas the distribution of scores in our sample wasnot significantly different from that in the Dutch norm group (analyzed with v2). Boys scored signif-icantly lower than girls on the total standard score, t(193) = 1.21, p < .001, r = .34, and on the subteststandard scores: Subtest 1 (Posting Coins/Placing Pegs), t(193) = 0.59, p < .05, r = 0.18; Subtest 2(Threading Beads/Threading Lace), t(195) = 7.73, p < .001, r = .29; Subtest 3 (Drawing Trail),t(195) = 13.59, p < .001, r = .25. When correcting for multiple testing (Bonferroni correction,p < .0125), the total standard score and the scores of Subtests 2 and 3 were still significantly higherfor girls.

Table 3MABC-2 NL manual dexterity standard scores.

Age (years) n Total Boys Girls

M SD M SD M SD

3 19 MD total 10.8 2.9 9.7 2.7 12.5 2.3Posting Coins 10.7 2.0 10.4 2.1 11.2 2.0Threading Beads 10.5 3.0 9.3 3.1 12.2 2.1Drawing Trail 1 10.1 1.8 9.6 1.9 11.0 1.4

4 22 MD total 10.3 2.6 9.1 2.7 11.2 2.4Posting Coins 10.8 2.2 10.3 2.3 11.1 2.1Threading Beads 9.5 2.2 8.3 1.8 10.3 2.1Drawing Trail 1 10.5 1.7 10.0 2.2 10.9 1.3

5 21 MD totala 9.5 2.0 9.0 1.5 10.3 2.6Posting Coinsa 9.5 2.6 9.0 2.6 10.1 2.6Threading Beads 10.0 2.4 9.3 2.5 10.9 2.1Drawing Trail 1 10.2 1.2 10.3 1.2 10.0 1.2

6 22 MD total 10.7 2.5 10.0 2.6 11.0 2.5Posting Coins 9.1 2.9 8.9 2.5 9.2 3.2Threading Beads 11.4 2.3 10.7 2.9 11.7 1.9Drawing Trail 1 11.0 0.2 10.9 0.4 11.0 0.0

7 35 MD totala 11.4 2.8 10.4 2.4 12.2 2.9Placing Pegsa 11.4 2.2 10.8 2.2 11.9 2.2Threading Lace 11.3 1.9 10.7 1.8 11.8 1.8Drawing Trail 2 9.7 2.5 9.3 2.4 10.2 2.6

8 27 MD total 11.3 2.9 9.5 3.4 12.6 2.9Placing Pegs 10.7 2.6 9.0 3.1 11.8 1.5Threading Lace 11.0 1.8 10.1 2.2 11.6 1.0Drawing Trail 2 10.6 2.2 9.7 2.8 11.3 1.4

9 21 MD total 11.2 3.0 10.2 3.3 12.4 2.4Placing Pegs 11.2 2.0 11.1 2.3 11.3 1.7Threading Lace 10.8 2.4 10.3 2.7 11.4 2.1Drawing Trail 2 10.1 3.1 8.8 3.7 11.6 1.4

10 30 MD total 11.9 2.3 11.3 2.4 12.9 1.5Placing Pegs 11.9 2.4 11.4 2.2 12.8 2.4Threading Lace 11.3 2.1 11.2 2.4 11.5 1.3Drawing Trail 2 10.5 1.9 10.2 2.4 11.0 0.0

Total 197 MD total 11.0 2.7 10.0 2.7 11.9 2.4Subtest 1 10.8 2.5 10.3 2.5 11.2 2.5Subtest 2 10.8 2.3 10.2 2.5 11.4 1.8Subtest 3 10.3 2.0 9.8 2.4 10.8 1.5

Note. MD total, manual dexterity total standard score; Subtest 1, Posting Coins/Placing Pegs; Subtest 2, Threading Beads/Threading Lace; Subtest 3, Drawing Trail 1/Drawing Trail 2.

a Scores on the manual dexterity total score and Subtest 1 were missing for 1 girl (5 years) and 1 boy (7 years).

M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305 301

Relation between action planning and manual dexterity

We calculated the correlation between the critical conditions of the action planning task and thescores of the two dexterity tasks. For the Box and Block Test, we used the scores of the dominant handbecause the action planning task was performed with the dominant hand. We found a significant, butlow, positive correlation, Spearman’s q = .28, p < .001. An increased number of blocks transported inthe Box and Block Test in 60 s was positively related to ending more often in a comfortable end pos-ture in the critical conditions of the action planning task. For the MABC-2 NL, we used the raw scoresof the subtests. These MABC-2 NL scores had no significant correlation with the critical conditions inthe action planning task except for Subtest 3 of Age Band 1 (Drawing Trail 1, ages 3–6 years), Spear-man’s q = �0.34, p < .01. Young children, who made few errors on the drawing trail, had a higher num-ber of comfortable end postures in the critical conditions of the action planning task.

302 M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305

Discussion

This study confirmed the previously found age effect in action planning in a large group of typicallydeveloping 3- to 10-year-olds. Specifically, we found a consistent developmental trend in action plan-ning. The development of anticipatory motor planning increased steadily between 3 and 10 years ofage, as older children planned their actions more often according to the end-posture comfort effectcompared with younger children. Our findings are in accordance with five other recent studies (Jans-sen & Steenbergen, 2011; Stöckel et al., in press; Thibaut & Toussaint, 2010; Van Swieten et al., 2010;Weigelt & Schack, 2010).

In our study, the percentages of comfortable end postures observed in the youngest children werelower than those observed in the previous study with a similar action planning task (Crajé et al., 2010).However, in the previous study, the number of participants was quite small. Moreover, in the currentstudy, we made the instructions more univocal. The children needed to use a power grip and were notallowed to change the sword’s orientation before grasping it. The mean percentage of comfortable endpostures in the critical orientations was 61% at 10 years of age. We expect the percentage of comfort-able end posture to increase further after this age because we did not yet observe a ceiling effect in thepercentage of comfortable end postures. In the study of Van Swieten et al. (2010), 9- to 14-year-oldsperformed a handle rotation task more often according to the end-state comfort effect than 5- to 8-year-olds (35–82 and 19–50%, respectively), although not nearly as often as adults (100%). On theother hand, Thibaut and Toussaint (2010) reported that action planning reached the same level ofend-state comfort as adults at 10 years of age. The percentage of comfortable end postures wasapproximately 90% at age 10 in their study. Similar results were found in the study of Stöckel and col-leagues (in press) in 9-year-olds. Both of these studies used a bar transport task (Stöckel et al., in press;Thibaut & Toussaint, 2010), and it seems that a ceiling effect is reached in this task at around 10 yearsof age. We assume that the differences between these two studies and the study of Van Swieten et al.(2010) and our study are due to task differences. Most likely, the bar transport task is easier than ouraction planning task and/or the handle rotation task used by Van Swieten and colleagues from a bio-mechanical point of view. The bar transport task has a limited range of possibilities to pick up andhandle the bar. There are only two ways and two orientations to pick up the bar, overhand and under-hand, and the bar does not provoke in hand manipulation. Our action planning task, on the other hand,has several orientations for beginning the task. Furthermore, for our action planning task and the han-dle rotation task, it is more difficult or awkward in some orientations to reach the end of the move-ment, whereas it is less difficult for the bar transport task to reach the end position. Therefore, thehandle rotation task and our action planning task may be more suitable to detect differences in antic-ipatory planning during childhood and (early) adolescence. An additional advantage of our actionplanning task may be that it is more ecologically relevant. Sword play may be less abstract than thebar transport and handle rotation tasks. In a previous study, we showed that a more functional taskleads to an increase in anticipatory planning (Steenbergen, Meulenbroek, & Rosenbaum, 2004).

We observed a decrease in action planning at 9 years of age. Remarkably, the 9-year-olds not onlyperformed worse in the critical orientations than the 10-year-olds but also performed worse in thecontrol orientations of the action planning task compared with 8- and 10-year-olds. Thibaut andToussaint (2010) reported a similar drop in action planning in their study, with 8-year-olds perform-ing less well than 6-year-olds. Thibaut and Toussaint argued that this effect can be caused by motorreorganization at around 8 or 9 years of age, where children hold more cues into account in perform-ing the action planning task than younger children. It is assumed that 8-year-olds do not yet suffi-ciently integrate these cues into their action as compared with 10-year-olds. This assumption issupported by neuroimaging studies. In these studies, it was shown that the brain undergoes markedchanges in white and gray matter, especially in the prefrontal cortex, in this age range (Casey,Tottenham, Liston, & Durston, 2005). The prefrontal cortex is an important structure for sensorimotorprocesses and planning (Casey et al., 2005; Wagner, Koch, Reichenbach, Sauer, & Schlösser, 2006).These empirical data may be explained from the perspective of the dynamic systems theory (DST; The-len, 1995). The motor reorganization and neural changes may lead to an increase in variability in plan-ning at 10 years of age onward. At the time of reorganization, planning may be hampered by these

M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305 303

changes and increasing input (Cox & Smitsman, 2006), in turn leading to a decline in anticipatory plan-ning at 9 years of age. However, it should be noted that this explanation is speculative at this point.

Interestingly and unexpectedly, we observed a gender difference in the action planning task. Girlsended with a comfortable posture in the critical orientations more often than boys. This difference inplanning was also observed in cognitive assessment of planning and might be related to differences inmaturation of the prefrontal cortex (Naglieri & Rojahn, 2001). In addition to this, we found similar gen-der differences in fine manual dexterity, where girls scored better than boys. This difference was alsoobserved in the norm reference group of the MABC-2 NL (Henderson et al., 2010) and has been re-ported in several other studies (van Mier, 2006). These gender differences in motor execution areprobably due, to a large extent, to environmental differences (Thomas & French, 1985; van Mier,2006). However, Piek, Gasson, Barrett, and Case (2002) also found physical gender differences duringinfancy with more synchrony in the joints of the arms of girls, and Soucie and colleagues (2011) foundgreater joint mobility for both arms and legs in females across age groups from 2 to 69 years. To ourknowledge, studies at school age on biological gender differences regarding fine manual dexterity arenot available.

The relation between planning and performance as assessed by manual dexterity tasks was gener-ally marginal. Manual dexterity, assessed with the Box and Block Test, was significantly related to theaction planning task, but the correlation value was low. Fine manual dexterity, assessed with theMABC-2 NL, was not significantly related to the action planning task except for drawing a trail inthe youngest age range. This correlation was again low. Therefore, we can assume that the action plan-ning task measures predominantly planning processes and only to a small extent motor executionprocesses. More important, the action planning task measures a different process than the generalmanual dexterity tests; that is, it is a specific measure of the preparatory processes related to actionperformance, that is, anticipatory action planning. Several studies have shown that anticipatory plan-ning is compromised in children and adolescents with cerebral palsy (Steenbergen & Gordon, 2006). Inchildren with unilateral cerebral palsy, our action planning task can be a valuable additional tool toinvestigate whether compromised motor behavior has a cognitive component related to anticipatoryplanning as well. Furthermore, our data could serve as reference data to compare anticipatory plan-ning in these children with that in typically developing peers. If planning is indeed compromised,rehabilitation programs may be adapted to train this specific motor planning aspect (Steenbergen &Gordon, 2006).

Reliability of the planning task was good, with both test–retest reliability and interrater reliabilitybeing excellent. This means that the action planning task can be reliably used as a clinical test to assessanticipatory planning in children. However, the test–retest reliability result should be interpretedwith some caution because the two youngest age groups were not represented in this sample.Although the ICC score was high (.90), this may have been somewhat lower if the youngest childrenwould have also participated in this sample.

Finally, we compared performance on the task by both the dominant and nondominant hands andfound no differences between the hands. Therefore, in children with normal motor function, it is suf-ficient to measure action planning in only one of the hands. Therefore, we suggest testing only thedominant hand. In children with a motor disorder (e.g., unilateral cerebral palsy), we recommend test-ing the nonaffected or less affected side of the body to decrease the influence of performance disabil-ity. Finally, we recommend testing the control orientations less often, for example, one trial perorientation instead of three trials per orientation. In this way, the task could be administered evenquicker while no essential information on action planning would be lost.

Conclusions

We found a consistent developmental trend in action planning in a large group of children aged 3 to10 years, with a drop in action planning at 9 years. This is the first study in which an experimental taskmeasuring anticipatory planning was tested for reliability and compared with manual dexterity tasks.The action planning task was found to be reliable to test anticipatory planning in young children. Fur-thermore, the action planning task related marginally to two manual dexterity tasks.

304 M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305

We conclude that the action planning task can be used as a valuable addition to measure anticipa-tory planning in a clinical setting. It can be used to investigate the extent to which compromised mo-tor behavior in preschoolers and school-age children is caused by cognitive processes related toanticipatory planning.

Acknowledgments

We thank Marga Huntelaar, Monique Hermanns, Roos Schoonderbeek, Lotte Verschragen, AngèlePeeters, Mandy Bentvelsen, and Ilja van Lierop for their help in collecting the data. We also thankall of the children and their parents for participating in this study. Finally, we thank the participatingschools for their help in organizing the data collection.

References

Adalbjornsson, C. F., Fischman, M. G., & Rudisill, M. E. (2008). The end-state comfort effect in young children. Research Quarterlyfor Exercise and Sport, 79, 36–41.

Casey, B. J., Tottenham, N., Liston, C., & Durston, S. (2005). Imaging the developing brain: What have we learned about cognitivedevelopment? Trends in Cognitive Sciences, 9, 104–110.

Cohen, R. G., & Rosenbaum, D. A. (2004). Where grasps are made reveals how grasps are planned: Generation and recall of motorplans. Experimental Brain Research, 157, 486–495.

Cox, R. F. A., & Smitsman, A. W. (2006). Action planning in young children’s tool use. Developmental Science, 9, 628–641.Crajé, C., Aarts, P., Nijhuis-van der Sanden, M., & Steenbergen, B. (2010). Action planning in typically and atypically developing

children (unilateral cerebral palsy). Research in Developmental Disabilities, 31, 1039–1046.Henderson, S. E., Sugden, D. A., Barnett, A. L., & Smits-Engelsman, B. C. M. (2010). Movement assessment battery for children (2nd

ed.). Amsterdam, The Netherlands: Pearson Assessment and Information [Dutch translation].Janssen, L., & Steenbergen, B. (2011). Typical and atypical (cerebral palsy) development of unimanual and bimanual grasp

planning. Research in Developmental Disabilities, 32, 963–971.Johnson-Frey, S. H., McCarty, M. E., & Keen, R. (2004). Reaching beyond spatial perception: Effects of intended future actions on

visually guided prehension. Visual Cognition, 11, 371–399.Jongbloed-Pereboom, M., Nijhuis-van der Sanden, M. W. G., & Steenbergen, B. (in press). Norm scores of the Box and Block Test

for children. American Journal of Occupational Therapy.Manoel, E. J., & Moreira, C. R. P. (2005). Planning manipulative hand movements: Do young children show the end-state comfort

effect? Journal of Human Movement Studies, 49, 93–114.Mathiowetz, V., Federman, S., & Wiemer, D. (1985a). Box and Block Test of manual dexterity: Norms for 6–19 year olds.

Canadian Journal of Occupational Therapy, 52, 241–245.Mathiowetz, V., Volland, G., Kashman, N., & Weber, K. (1985b). Adult norms of the Box and Block Test of manual dexterity.

American Journal of Occupational Therapy, 39, 386–391.Naglieri, J. A., & Rojahn, J. (2001). Gender differences in planning, attention, simultaneous, and successive (PASS) cognitive

processes and achievement. Journal of Educational Psychology, 93, 430–437.Perelle, I. B., & Ehrman, L. (1994). An international study of human handedness: The data. Behavior Genetics, 24, 217–227.Piek, J. P., Gasson, N., Barrett, N., & Case, I. (2002). Limb and gender differences in the development of coordination in early

infancy. Human Movement Sciences, 21, 621–639.Rosenbaum, D. A., Meulenbroek, R. J., Vaughan, J., & Jansen, C. (2001). Posture-based motion planning: Applications to grasping.

Psychological Review, 108, 709–734.Rosenbaum, D. A., Vaughan, J., Barnes, H. J., & Jorgensen, M. J. (1992). Time course of movement planning: Selection of hand

grips for object manipulation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 1058–1073.Short, M. W., & Cauraugh, J. H. (1999). Precision hypothesis and the end-state comfort effect. Acta Psychologica, 100, 243–252.Smyth, M. M., & Mason, U. C. (1997). Planning and execution of action in children with and without developmental coordination

disorder. Journal of Child Psychiatry, 38, 1023–1037.Soucie, J. M., Wang, C., Forsyth, A., Funk, S., Denny, M., Roach, K. E., et al (2011). Range of motion measurements: Reference

values and a database for comparison studies. Haemophilia, 17, 500–507.Steenbergen, B., & Gordon, A. M. (2006). Activity limitation in hemiplegic cerebral palsy: Evidence for disorders in motor

planning. Developmental Medicine and Child Neurology, 48, 780–783.Steenbergen, B., Meulenbroek, R. G. J., & Rosenbaum, D. A. (2004). Constraints on grip selection in hemiparetic cerebral palsy:

Effects of lesional side, end-point accuracy, and context. Cognitive Brain Research, 19, 145–149.Stöckel, T., Hughes, C. M. L., & Schack, T. (in press). Representation of grasp postures and anticipatory motor planning in

children. Psychological Research. Retrieved from <http://www.springerlink.com/content/mp018p300611243r/fulltext.pdf>.Thelen, E. (1995). Motor development: A new synthesis. American Psychologist, 50, 79–95.Thibaut, J.-P., & Toussaint, L. (2010). Developing motor planning over ages. Journal of Experimental Child Psychology, 105,

116–129.Thomas, J. R., & French, K. E. (1985). Gender differences across age in motor performance. A meta-analysis. Psychological Bulletin,

98, 260–282.Van Mier, H. (2006). Developmental differences in drawing performance of the dominant and non-dominant hand in right-

handed boys and girls. Human Movement Science, 25, 657–677.Van Swieten, L. M., van Bergen, E., Williams, J. H. G., Wilson, A. D., Plumb, M. S., Kent, S. W., et al (2010). A test of motor (not

executive) planning in developmental coordination disorder and autism. Journal of Experimental Psychology, 36, 493–499.

M. Jongbloed-Pereboom et al. / Journal of Experimental Child Psychology 114 (2013) 295–305 305

Wagner, G., Koch, K., Reichenbach, J. R., Sauer, H., & Schlösser, R. G. M. (2006). The special involvement of the rostrolateralprefrontal cortex in planning abilities: An event-related fMRI study with the Tower of London paradigm. Neuropsychologia,44, 2337–2347.

Weigelt, M., & Schack, T. (2010). The development of end-state comfort planning in preschool children. Experimental Psychology,57, 476–482.