Content uploaded by Martin Heil
Author content
All content in this area was uploaded by Martin Heil on Dec 03, 2015
Content may be subject to copyright.
Original Paper. Biomedical Human Kinetics, 3, 18 – 22, 2011
DOI: 10.2478/v10101-011-0005-6
The influence of juggling on mental rotation performance in children
Petra Jansen 1, Léonie F. Lange 2, Martin Heil 3
1 Institute of Sport Science, University of Regensburg; 2 Institute of Anaesthesiology, University of Medicine,
Berlin; 3 Institute of Experimental Psychology, University of Düsseldorf, Germany
Summary
Study aim: To assess the influence of juggling training on mental rotation performance in children.
Material and methods: Two groups of girls aged 6 – 14 years were studied: experimental (EG; n = 26) and control
(CG; n = 24). All girls solved a mental rotation task with 3-D block figures on computer screen (pre-test). After the
initial test, EG girls participated in juggling training for 3 months; the CG girls participated in light strength training
with theraband stretch bands. After 3 months, all girls solved the mental rotation task again (post-test). The post-
p
re
differences in the mental rotation performance were recorded.
Results: Children who learned juggling performed the mental rotation task significantly (p<0.05 – 0.01) faster, in
terms of reaction time, at non-zero angular disparity than their mates who were strength-trained.
Conclusions: Since mental rotation skills enhance spatial imagination, problem solving and mathematical skills, it may
be assumed that juggling training enhances also other cognitive domains and is worth implementing in the education
process.
Key words: Motor behaviour – Spatial performance – Juggling – Mental rotation – Children
Introduction
Research in the developmental, as well as in cognitive
science, is focused on the relations between motor abili-
ties and cognitive performance. Since Piaget’s work [19],
it has been assumed that the sensorimotor system is im-
portant for mental representation. Evidence supporting
this assumption came recently from modality theories
which emphasise the importance not only of a central
control executive function, but also of modal functions
[7]. It was shown that a dysfunction in motor develop-
ment is often associated with a dysfunction in cognitive
development and vice versa [8]. This relates to the spe-
cific assumption [3] that motor development and move-
ment experience are relevant factors for cognitive per-
formance, especially for spatial abilities [1]. Spatial abili-
ties are cognitive processes that involve visualisation,
orientation, and mental rotation [17]. Thereby, mental
rotation is the ability to imagine how an object would
look if rotated away from the orientation in which it is
actually presented [21].
The influence of motor processes on solving a men-
tal rotation task in adults was already demonstrated [22].
Mental rotation tests result in faster times and fewer errors
when manual and mental rotations were compatible. In
another study with adults [24], it could be shown that men-
tal rotation could be trained with the help of a manual
rotation programme, where mental rotation was trained
via motor rotation consisting of moving a joystick.
The effects of motor processes on solving a mental
rotation task were also investigated in children. A sam-
ple of 5, 8, and 11-year old children and of adults per-
formed a mental rotation task while simultaneously ro-
tating their hand, guided with a handle, in the direction
compatible/ incompatible with the mental rotation task.
The 5 and 8-year old children solved the mental rotation
task, compatible with the motor task, easier than the 11-
year old or adults, showing association between motor
and mental rotations in younger children [6]. In contrast
to that, a motor effect during mental transformations of
body parts was only found in one of two experiments
(performed with children aged 5 - 6 years, 7 years, and
adults) and this effect was less pronounced in younger
children [16].
An experimental study of body-oriented motor train-
ing, without using a handle or a joystick, on visual spa-
tial tasks in children is missing until now. In a study on
adults it was shown that juggling training over a period
Author’s address Prof. Petra Jansen, Institute of Sport Science, Universitätsstr. 31, 93053 Regensburg, Germany
petra.jansen@psk.uni-regensburg.de
P. Jansen et al. 19
of 3 months improved mental rotation performance com-
pared with a control, untrained group [12]. That result is
valid from a neurological point of view because increase
in brain plasticity was shown after juggling training [4]
in exactly that brain area (intraparietal sulcus) which is
involved in mental rotation [15].
It is not yet known if motor and imagery processes
in children and adults are alike [6,16]. The main goal of
this study was to find out if coordinative motor training
(here: juggling training) would lead to an enhancement
of mental rotation performance in children as shown earlier
for adults [12]. Beyond that, the present study expands
on that former study. Because a purely motor benefit
could also explain the reported differences in the reaction
velocity in juggling subjects compared to the control
ones [12], the control group in this study was subjected
to another conditional form of motor training, namely
strength training with theraband stretch-bands. The study
included only girls for the well-known gender differences
in mental rotation performance, even in children [18],
were not our objective.
Material and Methods
Participants: A group of 57 girls aged 6 – 14 years
participated in this study; 29 of them were randomly as-
signed to the juggling group and 28 to the strength train-
ing group. Seven girls did not participate in the post-test,
thus the experimental group (EG) counted 26, and the
control group (CG) 24 subjects. The girls were trained
in gymnastics, in Wesel, Germany, during their regular
gym classes. The girls from each group were trained by
two instructors at different corners of the gym. The par-
ents of children gave their informed consent for children’s
participation and data utilisation after having been made
familiar with the objective and protocol of the study which
was approved by the local Committee of Ethics. In addi-
tion, the club committee and the gymnastics instructors
agreed to carry out the experiments during the lessons.
The children received juggling balls or the theraband
stretch bands as gifts and the entire gymnastics group
received €300 for their participation. The girls had nor-
mal or appropriately corrected visual acuity. None of
the children were able to juggle before participating in
this study.
Procedure: All participants were subjected to timed
pre- and post-study mental rotation tasks which took place
in a quiet locker room or in a first-aid room, in two dif-
ferent gyms in Wesel, in groups of three children. Each
girl worked with her own laptop (15-inch monitor at about
50-cm distance) not seeing her mates and had her personal
test leader. The experimental stimuli consisted of 18 per-
spective line drawings of three-dimensional forms (each
one composed of 10 cubes) [13,21] and were 3D-rotated
around the horizontal or vertical axis. Each form had a
maximum size of 7 × 7 cm on the screen spaced by 14
cm. In each trial, two drawings of the same form were
presented together. The stimulus presented on the right
side of the monitor was either identical to the left side or
mirror-reversed. The angular disparity between the two
stimuli was 0, 90 or 180° (see Fig. 1 as an example). The
subjects responded “same” by pressing the left touchpad
button with index finger and “different” by pressing the
right touchpad button with the middle finger.
Individual sessions lasted about 50 min each; every
trial started with a grey-background screen and after 500
ms the pair of stimuli appeared and the subject had to de-
cide whether the two stimuli were “same” or “different”
(mirror-reversed). The stimuli remained on the screen
until the subject responded by pressing a touchpad-button.
The participants were instructed to react as quickly and
as accurately as possible. The “+” or “–” sign was dis-
played for 500 ms in the centre of the screen to indicate
the correctness of given response. The trials were spaced
by 1500-ms grey-background screen intervals. After a
block of 27 trials, the participants could choose to take a
short break. They were asked to start the new block by
pressing the space key. Each combination of objects (18
different drawings), normal or mirror-reversed, and an-
gular disparity (0, 90 or 180°) was presented 4 times
during the test, which resulted in a total number of 432
trials. In order to make the participants familiar with the
test sequence, 54 unrecorded test trials were performed
at the beginning of the experiment. The stimulus items
were the same in the pre- and post-study mental rotation
tests separated by a 3-month interval. In that time period,
the children of the experimental group (EG, juggling)
were subjected to juggling training twice weekly, 15 min
at the beginning and 15 min at the end of regular gym-
nastics classes. The training was based on the Rehoruli’s
method and designed by a German juggler, Stephan Eh-
lers [5]. The children practiced various throw-and-catch
tasks, their difficulty increasing every week. The children
were instructed to train at home for about 10 min daily.
At every session, the exercise time and the numbers of
successful throws were recorded – how often a ball was
successfully thrown from one hand to the other during
the juggle performance.
Control children subjected to strength training (CG)
were underwent a light strength training with theraband
stretch bands. The following exercises were applied: “Bi-
ceps curls”– the girl stood with one foot on the theraband,
its ends being grasped with one hand each and wrapped up
until tight, then both arms were bent alternately, and
20 Mental rotation training of children
“Knee bend” – the girl stood on both feet on the theraband,
shoulder length apart, then bent down and wrapped the
ends of the theraband round her hands; after that, she sat
up and stretched up her arms simultaneously. Control
girls were also asked to train at home for 10 min a day
and the numbers of repetitions of the “Knee bend” exer-
cise they were able to perform were recorded during the
training sessions.
The motor training effect in the juggling group (EG)
was expressed as the difference between the first and
the last training sessions in the numbers of successful
throws. That effect in the strength-training group (CG)
was expressed as the difference in the numbers of “knee
band” exercise.
Data analysis: Only trials with correct response were
used for reaction time (RT) analysis which was further
restricted to “same” responses only, for angular dispar-
ity could not be unequivocally defined for “different”
responses [14]. Moreover, all RT values outside the
mean ± 2SD interval were eliminated prior to the analy-
sis. Nevertheless, the analysis of all data combined re-
vealed identical results.
The percent differences in the error rate scores (dif-
ference score error) and the post-pre RT differences (dif-
ference score RT) were dependent variables [24]. The
data were subjected to one-way ANOVA for each group
separately, the significance levels being corrected ac-
cording to Huynh and Feldt [11] in order to compensate
for non-sphericity of the data. Pearson’s coefficients of
correlation were computed for age, mental rotation per-
formance and motor behaviour (difference score between
the first and last training session for juggling or strength
performance). The level of p≤0.05 was considered sig-
nificant.
Results
No significant between-group difference in the reac-
tion time (mental rotation task) was found at baseline. In
the experimental group, juggling performance signifi-
cantly (p<0.01) increased following the 3-month training
compared with the pre-training status (baseline). The same
was true for the control group performing “knee bend”
exercises (Table 1). An example of the items of the
mental rotation task is presented in Fig. 1.
The results of the training-induced differences in the
reaction time are presented in Fig. 2. Significantly (p<0.01)
better results were attained by girls from the experimen-
tal group (EG) than from the control one (CG) at angu-
lar disparities equal to 90 or 180º. Moreover, mean RT
improvement in EG was significantly (p<0.01) highest
at 90º. No significant angle-related differences were
noted in CG. Regarding the error rate scores, no sig-
nificant differences either between groups or between
angular disparities were found.
Table 1. Mean values (±SD) of age, reaction time pre-
training and of training effects
Group
Variable Juggling
n = 26 Strength
n = 24
Age (years) 10.4 ± 2.2 10.5 ± 2.4
Reaction time (ms) pre 3704 ± 1055 3297 ± 1061
Motor training effect
Range 16.8 ± 30.8**
0 – 117 2.33 ± 1.76**
0 – 6
** p<0.01
Fig. 1. An example of the items of the mental rotation
task
Fig. 2. Post-pre differences in reaction time in the men-
tal rotation task (Means ±SE)
** Significant (p<0.01) difference between groups; ºº Signifi-
cantly (p<0.01) different from mans at 0 and 180º
Training effects were significantly (p<0.05) corre-
lated with age in both groups but no significant correla-
tions were noted with either difference scores in RT or
with error rates (Table 2). Multiple regression calculus
applied to data in the experimental group showed that a
significant contribution to the mental rotation task had
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0º 90º 180º
EG
CG
ms
Angular disparity
**
**
oo
P. Jansen et al. 21
only age (p<0.05), that of motor performance remaining
non-significant.
Table 2. Coefficients of correlation of training effects
with age or difference scores in the mental rotation task
Group
Correlated variable Juggling
n = 26 Strength
n = 24
Age 0.466* 0.443*
Difference score (RT) -0.012 0.160
Difference score (Error) -0.142 0.216
* p<0.05
Discussion
This study was a continuation of our earlier report
on juggling [12] and showed the relations between jug-
gling and the mental rotation performance in school-
children. The results indicated that the training effect was
specific for juggling and not for a general motor effect.
The juggling training affected the reaction time (RT) in
trials, in which the angular disparity was 90 or 180°, that
at 0° remaining unchanged. Of course, no mental rota-
tion is required to compare the two stimuli at 0° dispar-
ity, that condition serving thus as a kind of reference for
the other two. In contrast to RT, there was no effect of
juggling training on error rates. This finding is in accor-
dance with many other studies showing that reaction
time is the substantially more sensitive measurement for
mental rotation effects than error rates [10].
This study confirmed the results of an earlier study
[23] that manual rotation training can improve the men-
tal rotation performance of school-aged children but the
results depended on age [6]. It was demonstrated that
only those aged 5 or 8 years but not the 11 years old had
shorter RT in mental rotation tasks when the spinning of
a hand crank matched the assumed path of mental rota-
tion, rather than the other way round. When mental rota-
tion of pictures showing body parts were studied, motor
effects in mental rotation were not stable [16] and it was
concluded that imagery and motor planning shared com-
mon resources. This is in accordance with the view that
mental representations such as mental rotation are con-
trolled by different modal functions connected to sen-
sory and action systems [16,19]. When interpreting the
results of mental rotation processes in terms of embodi-
ment [25] and modal functions, the experimental design
has to be very precise. It might be speculated why jug-
gling and not a strength task is effective in mental rota-
tion. Both motor tasks imply a performance feedback;
girls from both groups received feedback from their
training improvement, although the feedback in the jug-
gling group was coordinative and that in the strength
group was a strength one; it could be assumed that jug-
gling and mental rotation shared common features, since
both juggling and mental rotation required a cyclic ac-
tivity and had temporal and spatial constraints [12]. In
juggling, the hands move along more or less elliptical
trajectories while throwing and catching balls move in a
regular fashion [20]. In mental rotation, one object is
brought along a cyclic trajectory around the three axes
to bring it in the position of the standard object. While
juggling may be thought of as a “spatial clock” [20], men-
tal rotation seems to be a covert manual rotation [26].
The latter authors showed that rotational hand move-
ments interfered with mental rotation and vice versa.
Furthermore, juggling requires mirrored movements of
hands while mental rotation requires the decision of
whether two rotated objects are superimposable or mir-
rored. Further studies are needed to compare the effects
of mental rotation and of other motor tasks involving
cyclic arms or body activity, e.g. swimming or discus
throwing.
The here presented data suggest that the age, not mo-
tor development, is the key factor in the improvement.
This assumption has to be investigated in further studies
with cognitive development monitoring in order to more
precisely assess the effects of the cognitive and motor
processes on mental rotation performance. That latter is
not influenced automatically by the motor system but
since it shares common features with higher motor pro-
cesses, such as juggling, it might be influenced by a spe-
cific motor task. Another issue to be investigated into
are the known gender-related differences (cf. [13]); a
stronger activation of motor cortex was noted in women
than in men when observing hand vs. dot movement, in
men the situation being reversed [2].
Furthermore, the role of hormones ought to be dis-
cussed in more detail. The girls in this study were 6 – 14
years old and differed in their hormonal status. Because
at least the effect of testosterone level on mental rotation
performance has been shown [9], the varying hormonal
status before and during puberty has to be regarded in
further studies.
Summing up, a relation between juggling and mental
rotation performance and, in a broader sense, between
motor and cognitive performance was demonstrated.
Further studies are needed to evaluate, in detail, the ef-
fects of age, gender and training duration on how cogni-
tive the motor tasks need to be to have a positive influ-
ence on other higher-level cognitive tasks such as problem-
solving, as well the stability of such effects.
22 Mental rotation training of children
References
1. Campos J.J., D.I.Anderson, M.A.Barbu Roth, E.M.Hubbard,
M.J.Hertenstein, S.Witherington (2000) Travel broadens the
mind. Infancy 1:149-219.
2. Cheng Y.W, O.J.L.Tzeng, J.Decety, T.Imada, J.C.Hsieh
(2006) Gender differences in the human mirror system: a mag-
netencephalography study. Neuroreport 17:1115-1119.
3. Diamond A. (2007) Interrelated and interdependent.
Dev.Sci. 10:152-158.
4. Draganski B., C.Gaser, V.Busch, G.Schuierer, U.Bogdahn,
A.May (2004). Nuroplasticity: changes in grey matter induced
by training. Nature 427:311-312.
5. Ehlers S. (2005) REHORULI – Jonglieren lernen mit Er-
folgsgarantie. Books on Demand.
6. Frick A., M.M.Daum, S.Walser, F.W.Mast (2009) Motor
processes in children`s mental rotation. J.Cogn.Dev. 10:18-40.
7. Gallese V., M.Lakoff (2005) The brain`s concept. The
role of the sensory-motor system in conceptual knowledge.
Cogn.Neuropsychol 22:455-479.
8. Gillberg C. (2003) Deficits in attention, motor control,
and perception: a brief review. Arch.Dis.Child. 88:904-910.
9. Hausmann M, O.Güntürkün (2000) Steroid fluctuations
modify functional cerebral asymmetries: the hypotheses of
progesterone-mediated interhemispheric decoupling. Neuro-
psych. 38:1362-1374.
10. Heil M., F.Rösler, M.Link, J.Bajric (1998) What is im-
proved, if a mental rotation task is repeated - efficiency of
memory access, or the speed of a transformation routine?
Psychol.Res. 61:99-108.
11. Huynh H., L.S.Feldt (1976) Estimation of the Box cor-
rection for degrees of freedom from sample data in random-
ized block and split-plot designs. J.Educ.Stat. 1:69-82.
12. Jansen P., C.Titze, M.Heil (2009) The influence of jug-
gling on mental rotation performance. IJSP 4:351-359.
13. Jansen-Osmann P., M.Heil (2007) Suitable stimuli to ob-
tain (no) gender differences in the speed of cognitive processes
involved in mental rotation. Brain Cognit. 6:217-227.
14. Jolicoeur P., S.Regehr, L.B.Smith, G.N.Smith (1985)
Mental rotation of representations of two-dimensional and three-
dimensional objects. Can.J.Psychol. 39:100-129.
15. Jordan K., H.J.Heinze, K.Lutz, K.M.Kanowski, L.Jäncke
(2001) Cortical activations during the mental rotation of dif-
ferent visual objects. NeuroImage 13:143-152.
16. Krüger M., H.Krist (2009) Imagery and motor processes
- when are they connected? The mental rotation of body parts
in development. J.Cogn.Dev, 10:239-261.
17. Linn M.C., A.C.Petersen (1985) Emergence and charac-
terization of sex differences in spatial ability: A meta-analysis.
Child Dev. 56:1479-1498.
18. Neuburger, S.P.Jansen, M.Heil, C.Quaiser-Pohl (in press)
Gender differences in pre-adolescents‘mental rotation per-
formance: do they depend on grade and stimuli? Personality
and Individual Differences.
19. Piaget J. (1954) Construction of Reality in the Child.
Basic Books, New York.
20. Post A.A., A.Daffertshofer, P.J.Beek (2000) Principal
components in three-ball cascade juggling. Biol.Cyber. 82:143-
152.
21. Shepard R.N., J.Metzler (1971) Mental rotation of three-
dimensional objects. Science 171:701-703.
22. Wexler M., S.M.Kosslyn, A.Berthoz (1998) Motor proc-
esses in mental rotation. Cognition 68:77-94.
23. Wiedenbauer G., P.Jansen-Osmann (2008) Manual train-
ing of mental rotation in children. Learn.Instr. 18:30-41.
24. Wiedenbauer G., J.Schmid, P.Jansen Osmann (2007) Man-
ual training of mental rotation. Eur.J.Cogn.Psychol. 19:17-36.
25. Wilson M. (2002) Six views of embodied rotation and
the planning of hand movements. Psychon.Bull.Rev. 9:625-636.
26. Wohlschläger A., A.Wohlschläger (1998) Mental and
manual rotation. J.Exp.Psychol.Hum. 24:397-412.
Received 19.01.2011
Accepted 19.02.2011
© University of Physical Education, Warsaw, Poland