Journal of Motor Behavior

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/vjmb20

The Relation of Mental Rotation and Postural

Stability

Philipp Hofmann & Petra Jansen

To cite this article: Philipp Hofmann & Petra Jansen (2021): The Relation of Mental Rotation and
Postural Stability, Journal of Motor Behavior, DOI: 10.1080/00222895.2021.1899113

To link to this article: https://doi.org/10.1080/00222895.2021.1899113

Published online: 21 Mar 2021.

Submit your article to this journal

Article views: 53

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=vjmb20
 Journal of Motor Behavior, 2021
# Taylor & Francis Group, LLC

RESEARCH ARTICLE
The Relation of Mental Rotation and Postural Stability
Philipp Hofmann, Petra Jansen
Faculty of Human Science, University of Regensburg, Regensburg, Germany

ABSTRACT. Main goal of this study was to investigate the is a rotated version of the left stimulus. The right stimu-
influence of mental rotation tasks on postural stability. 84 par- lus could be the same or a mirrored version of the left
ticipants were tested with two object-based mental rotation
tasks (cube vs. human figures), an egocentric mental rotation
stimulus. The participants have to decide whether both
task with one human figure, a math- (cognitive control) and a items are not mirror reversed, i.e., “same”, or mirror
neutral task, while standing on a force plate in a both-legged reversed, i.e., “different”. In an egocentric condition, a
narrow stance. Parameters related to the Center of Pressure common task is that only one human figure, raising one
course over time were used to quantify postural stability. The
simultaneous solution of mental rotation tasks has led to pos- arm, is presented and the participant has to decide if
tural stabilisation compared to the neutral condition. Egocentric either the right or the left arm was raised (Jansen &
tasks provoked more postural stability than object-based tasks Kaltner, 2014). One main difference between both types
with cube figures. Furthermore, a more stable stance was
observed for embodied stimuli than for cube figures. An of transformations is that in an object-based task the
explorative approach showed the tendency that higher rotation observer’s position in relation to the environment
angles of the object-based mental rotation task stimuli lead to remains fixed and the participant has to judge the mental
more postural sway. These results contribute to a better under- rotation stimuli in relation to each other. In contrast to
standing of the interaction between mental rotation and motor
skills and emphasize the role of type of task and embodiment this, participants have to imagine themselves rotating in
in dual task research. an egocentric mental rotation task. In this case the rela-
Keywords: mental rotation, postural stability, embodiment, tion between environment and object remains fixed.
dual task Egocentric transformations involve a simulative process
recruiting a representation of our own body (Kessler &
Rutherford, 2010).
These egocentric or perspective transformations are
Introduction performed faster and more accurately than object-based
rotations in the picture plane (Amorim & Stucchi, 1997;
tanding upright seems to be one of the easiest and
S standardized motor tasks for adults. Besides, it has
been shown that this easy and quite natural motor task
Keehner et al., 2006; Wraga et al., 1999; 2005).
Furthermore, the function of response time on angle of
rotation shows a flatter slope for egocentric transforma-
can be influenced by cognitive tasks, which have to be
tions compared to object-based transformations, because
completed during standing. However, a clear result pat-
tern of the relation between postural stability and the the increase of the slope in egocentric mental rotation
execution of cognitive tasks is not apparent. While in tasks appear only for angles above 60 or 90 degrees
some dual-tasks studies the parallel execution of different (Keehner et al., 2006; Michelon & Zacks, 2006). One
types of cognitive tasks leads to a deterioration of pos- reason for this might be that in egocentric tasks larger
tural stability (Mujdeci et al., 2016; Pellecchia, 2003; angles request greater mental effort because of the per-
Shumway-Cook et al., 1997; Simoneau et al., 1999), spective transformation whereas smaller angles can be
other studies show that the simultaneous performance of judged by a visual matching. The more classical increase
different cognitive tasks leads to enhanced postural sta- of response time with increasing angular disparity is
bilization (Andersson et al., 2002; Dault et al., 2001; more obvious in object-based tasks than in egocentric
Hunter & Hoffman, 2001; Potvin-Desrochers et al., mental rotation tasks (Jola & Mast, 2005; Michelon &
2017; Vuillerme et al., 2000). Zacks, 2006). The increase in reaction times and
decrease in hit accuracy with an increasing angle
Mental Rotation
(Shepard & Metzler, 1971) indicates that the mental rota-
Mental rotation is the cognitive process of imagining tion tasks become more difficult with increasing angle.
how an object would look like if it’s rotated from its ori- Approaches already exist to confirm this also with the
ginal position (Shepard & Metzler, 1971). There are two physiological measurement of pupillary diameters (Bauer
types of transformations in mental rotation: object-based et al., 2021).
spatial transformations and egocentric perspective trans-
formations (Zacks et al., 2000). An object-based mental Correspondence address: Philipp Hofmann, Faculty of Human
rotation task is often characterized through the presenta- Science, University of Regensburg, Universit€atsstraße 31, 93053
tion of two stimuli on a screen, where the right stimulus Regensburg, Germany. E-mail: philipp.hofmann@ur.de

1
P. Hofmann and P. Jansen

Influence of Mental Rotation on Motor Tasks With regard to the different perspectives in mental
rotation tasks, especially with embodied stimuli, there
A link between motor processes and the mental rota-
are similarities to motor imagery, which can be seen as
tion ability has been carved out in several studies
the inner representation of an action without performing
(Jansen et al., 2012; Pietsch & Jansen, 2012;
it physical (Di Rienzo et al., 2014). For motor imagery,
Steggemann et al., 2011; Voyer & Jansen, 2017). Also
Wexler et al. (1998) stated that the mental rotation pro- two perspectives can be distinguished: the kinesthetic
cess is a covert motor rotation and Wohlschl€ager and perspective, where a person imagines performing the
Wohlschl€ager (1998) assumed that motor and mental movement himself and the visual perspective, where a
rotations share common processes. Most of the experi- person imagines someone else performing the movement
mental studies, which are related to the influence of a (Stins et al., 2015). Therefore, an egocentric mental rota-
motor training on mental rotation, show a beneficial tion task with an embodied stimulus corresponds to the
result. The other way around, this means the question if kinesthetic perspective and an object based mental rota-
a mental rotation task influences a motor task, has not tion task is similar to the visual perspective. Rodrigues
been systematically investigated until now, even there et al. (2010) examined the two different sights of motor
are a lot of studies investigating the effect of mental imagery, kinesthetic and visual, with the imaging of a
training on motor task and postural stability. However, plantar flexion movement (rising on tiptoes) and found
in this study we are interested in the common processes that imaging oneself rising on tiptoes leads to a greater
of basic motor processes and mental rotation. This would postural sway than imaging someone else rising on tip-
give a hint that for mental rotation basic motor processes toes. They tested in a standardized feet position with
are important for the covert motor rotation. both feet together and measured the muscle activity at
Until now, only a few studies already examined the the Musculus gastrocnemius with a surface-EMG to con-
explicit influence of mental rotation on the postural trol for actual movement during the imagination. Stins
sway: For example, Dault et al. (2001) investigated with et al. (2015) investigated the imagination of five differ-
an experimental design the influence of a) different ent activities (2 lower body movements, 2 upper body
working memory tasks for different postural conditions movements and 1 neutral condition) from a kinesthetic
and b) an egocentric mental rotation task with a stickman and a visual perspective. They showed that the kines-
on the variability of the Center of Pressure (CoP) dis- thetic imagery of lower body movements caused a
placement in anterior-posterior and medio-lateral direc- greater postural sway than the kinesthetic imagery of
tion. In this latter task the results carved out a decrease upper body movements or the neutral condition. For vis-
in the amplitude of sway compared to a control condition ual imagery they found no significant differences
but there were no differences in response to the different between the conditions.
types of working memory tasks. Ganczarek et al. (2015)
investigated the influence of looking on a picture with
different depths and background areas on body sway but Goal of the Study
also stressed out significant positive correlations between It is the main goal of this study to investigate how dif-
body sway parameters and the performance of a psycho- ferent kinds of mental rotation tasks, egocentric tasks
metric mental rotation test, which was measured after the with embodied stimuli, object-based tasks with embodied
postural stance task. They suggested that body sway stimuli and object-based tasks with non-embodied stimuli
might be mediated by individual differences in mental influence a basic motor ability, i. e. postural stability.
imagery. Kawasaki and Higuchi (2013) found that only The results will give insight in the relation of mental
conducting an egocentric mental rotation task with feet
rotation and basic motor processes and contribute to the
but not with cars as stimuli results in a stabilization of
common process theory (Wohlschl€ager & Wohlschl€ager,
body sway. The body sway has been measured immedi-
1998). If mental rotation has an influence on postural
ately after the mental rotation intervention. This finding
stability, one might assume that both tasks share com-
was only found during unipedal standing but not during
mon processes. This might give a hint, that the common
a bipedal stance with the feet close together. For other
process is not a covert motor rotation but a basic
body parts, like for example hand stimuli no correlation
motor process.
between the mental rotation task and postural stability
Therefore, the following hypotheses will be
could be shown when measured in unipedal stance
investigated:
(Kawasaki et al., 2014). Also the up to 60 min lasting
effect of a mental rotation intervention on postural stabil- 1. First of all and according to the study of Dault et al.
ity, measured in unipedal stance, was more effective for (2001), we assume generally a minor body sway
feet stimuli than for hand stimuli (Kawasaki & after completing a mental rotation task compared to
Higuchi, 2016). looking at a fixation cross.

2 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

2. Second, the relation of body sway and mental declaration. All participants gave their written informed
rotation can be investigated due to the nature of the consent to participate in this study.
mental rotation task:
Material
a. More body sway in egocentric tasks than in Postural Stability Task
object-based tasks. Because an egocentric mental To quantify the postural sway of each person, the Center
rotation task is similar to the kinesthetic imagery of Pressure (CoP) course over time will be investigated,
and an object based mental rotation task is which is measured with a force platform (Rhea et al.,
similar to the visual imagery, we hypothesize that 2014). During bipedal standing the CoP is the point loca-
solving an egocentric mental rotation task causes tion of a weighted average of the sum of vertical ground
a bigger change in body sway than solving an reaction forces applied by the feet on the force plate
object based mental rotation task because the (Winter, 1995). A smaller CoP displacement is referred to
subject has to imagine rotating his/her own body as a more stable stance (Palmieri et al., 2002) and this con-
(Kessler & Rutherford, 2010). While previous clusion of a smaller CoP displacement as a more stable
studies have been able to show this relation for stance is common (Rhea et al., 2015). In this study an
lower body movements, we will investigate AMTI force plate (AMTI OR6-7-2000) with a sampling
whole body images, as these correspond to the frequency of 1000 Hz was used. The cognitive tasks were
classical mental rotational stimuli. run on a laptop (Dell Inspiron 1750, 1600  900) placed on
b. Less body sway with humanized stimulus eye-level at a self-selected distance in front of the partici-
material than with cube figures.In accordance pant, standing on the force plate. Postural stability in dual
with the results of Kawasaki and Higuchi (2013) task designs is very vulnerable to disturbances caused by
we assume that solving mental rotation tasks with different types of response, as for example verbal
human figures causes a stabilization of body responses, which have an influence on body sway (Conrad
sway compared to solving a mental rotation task & Sch€onle, 1979; Dault et al., 2003; Jeong, 1991; Yardley
with cube figures because of the embodiment
et al., 1999), and also the direct response at the laptop,
theory (Wilson, 2002). To control for this we
since any contact with an anchored object stabilizes body
examine the same object-based task and only
sway (Clapp & Wing, 1999; Jeka & Lackner, 1994).
vary the stimulus-material (cube figures vs
Because of this and similar to Huxhold et al. (2006), the
human figures).
test persons were therefore given one bluetooth mouse in
c. More body sway with increasing angular
each hand. However only the right mouse was switched on
disparity. As Pellecchia (2003) stated an
increasing body sway with an increasing and connected to the laptop for answering. Participants
difficulty of the concurrentcognitive task were asked to stand as still as possible in an erected pos-
approach it will be investigated, if the increasing ition with their arms at their sides. The palm of the hand
rotationangles of the stimuli in the mental faced towards the body without touching the body. The
rotation tasks will cause more body sway. feet were placed in a narrow stance position on either side
of a three cm wide tape and with the heels aligned in front
of another tape in order to maintain a standardized foot
Methods placement (Richer & Lajoie, 2019). The following Figure 1
shows a sketch of the test setup.
Participants All participants wore ultra-thin try-on socks to main-
84 students of the University of Regensburg (64 tain the barefoot condition but provide a better hygiene.
females and 20 males) participated in this study. With a The head faced the laptop. One trial lasted minimum
small to medium effect size f ¼ .15, an alpha-level of p 70 seconds and consisted out of several tasks of the same
¼ .05 and a power of 1-ß ¼ .95, a power analysis with condition. The force plate and laptop were aligned so
Gpower (Faul et al., 2007) for the repeated measures that the line of sight was directed to the wall. In add-
ANOVA resulted in N ¼ 84 to detect significant effects ition, the force plate was located on the open side of a
for the body sway between the five different types of three-part construction with 1,25 m (width) x 1,85 m
stimuli. The participants’ mean age was 21.1 years (SD (height) blue moveable walls to prevent unwanted visual
¼ 1.6 years), with a range from 18  27 years. The aver- stimuli. To characterize the course of the CoP, the
age height of the participants was 171.7 cm (SD ¼ parameters maximum width in anterior-posterior direc-
9.3 cm). None of the participants had a disease or an tion, maximum width in medio-lateral direction, mean
injury affecting the balance. The experiment was con- amplitude and mean sway velocity were calculated
ducted according to the ethical guidelines of the Helsinki (Hufschmidt et al., 1980; Palmieri et al., 2002).

3
P. Hofmann and P. Jansen

Object-based mental rotation task with embodied fig-

ures: Two pictures of the same female were shown next
to each other. The women on the picture raised one arm.
Instructions for solving the tasks were the same as in the
cube figures condition.
Egocentric mental rotation task with embodied figures:
The picture showed only one female human figure,
which either raised her right or left arm. Participants had
to click the left mouse button if the figure raised the left
arm and the right mouse button if the figure raised the
right arm. The female figure was the same for both con-
ditions and is the same female figure as used in Kaltner
and Jansen (2018).
For all mental rotation tasks, the mental rotation fig-
ures had a size of 400 px times 400 px and were only
rotated in picture plane clockwise by 60 -steps (0 , 60 ,
FIGURE 1. Exemplary test set-up. 120 , 180 , 240 , 300 ). During the practice session for
the mental rotation tasks there was feedback presented
for 1500 ms in font size 40 shown in the center of the
Cognitive Task screen. During the main trial no feedback was given but
For presenting the stimuli the software “Presentation” a fixation cross in the middle of the screen for 1500 ms.
(Version 20.1 Build 12.04.17, Neurobehavioral Systems, The math task revealed the final result during practice
Inc., Berkeley, CA, www.neurobs.com) was used. but there was no feedback given in the main trial.
Regarding the experimental design there were five differ-
ent conditions, two control conditions and three mental Cognitive and Physical Effort
rotation tasks, see Figure 2. As task difficulty or fatigue might be confounding fac-
All stimuli were presented on a white screen. In the tors the questionnaire ASS ("Effort scale in sport") by
first control condition participants had to look at a fix- B€usch et al. (2015) was used to assess subjective cogni-
ation cross without doing something else. In the second tive and physical effort during the experiments. The ASS
control condition they had to solve a math-task with sub- questionnaire is a German ten-level evaluation question-
tractions. For this, they got a randomly chosen number naire with additional color coding. Each level is provided
between 200 and 999 as starting number and then a ser- with a semantically one-dimensional sentence ("not exer-
ies of numbers between 1 and 9, which they had to sub- ting” to "so exerting that I have to stop"). To evaluate
tract from the last interim result. After solving the cognitive effort, the tenth level has been replaced by the
subtraction in mind, they clicked the right or left mouse sentence "so exerting that I cannot solve".
button and the next number was presented. When the
trial was finished, they informed the experimenter about Procedure
the final result. The three mental rotation tasks include
a) an object based mental rotation task with cube figures, The individual test sessions lasted about 45 minutes
b) an object based mental rotation tasks with embodied and took place in a quiet laboratory at the University of
stimuli and c) an egocentric mental rotation tasks with Regensburg. All participants were tested separately and
embodied stimuli. there was always the same experimenter. A test session
Object-based mental rotation task with cube figures: consisted of practice and 10 single trials (each of the
The cube figures consisted of ten alternating black and five cognitive tasks appeared two times). After this they
white cubes and were created with the software R, based had to complete a demographic questionnaire. To carry
on the work of Jost and Jansen (2020). The alternation out one trial, one of the five cognitive tasks had to be
of colors defines a bottom and a top of the cube figures completed for 70 seconds while the postural stability task
and makes them therefore more similar to the human fig- on the force plate was being performed simultaneously.
ures (Amorim et al., 2006). It was the task of the partici- A condition could exceed 70 seconds, because the experi-
pants to click either the left mouse button if the two ment was programmed in a way that the last task of a
presented stimuli were the same, i.e., the right stimulus trial could be solved by the test person, so that the trial
was not mirror reversed to the left one, or the right was not aborted while solving a task. The cognitive tasks
mouse button if the stimuli were different, i.e., the right (two control conditions: fixation cross task, math task;
stimulus was mirror reversed to the left stimulus. The three experimental tasks: object-based mental rotation
left stimulus always remained the same figure. task with cube figures, object-based mental rotation task

4 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

FIGURE 2. Two control conditions and three different mental rotation tasks.

with human figures, egocentric mental rotation task with sway was compared at the different rotation angles of
human figure) were pseudo-randomized for each partici- the mental rotation tasks. For this purpose, the force
pant. This means, the first five trials consisted of the plate data were synchronized with the mental rotation
five cognitive tasks in randomized order and so did the data in Matlab, so that for each period while a stimulus
second five trials. There was a 90-second break between (without fixation cross and pause) was displayed, a value
the trials, during which the subjects sat down and, using for the sway parameters (maximum width in anterior-
the ASS questionnaire, evaluated the previous trial in posterior direction, maximum width in medio-lateral dir-
terms of cognitive and physical effort. At the beginning ection, mean amplitude and mean sway velocity) was
of each trial, the test person was positioned on the force given. A two-way repeated measures analysis of variance
plate by the experimenter. After this, standardized using the factors "stimulus type" and "rotation angle"
instructions were given for the respective tasks. was then calculated for each parameter. Additionally, the
Participants were allowed to start the trial themselves by mean slope of each parameter in the respective measure-
clicking the mouse, which started the trial immediately. ment unit per 60 angle change. (This results in four
Simultaneously with this mouse click, the experimenter mean values for clockwise and anti-clockwise measure-
started the CoP recording of the force plate. After com- ments, 0 ; 60 /300 ; 120 /240 ;180 ) was calculated for
pletion of each trial, the experimenter stopped the CoP each stimulus type and checked with a one-way repeated
recording and asked the test person to step down from measures analysis of variance. Only correct responses to
the force plate for taking a seat at a table to fill out the non-mirrored stimuli completed within the 60 seconds
ASS questionnaire. Before the 10 trials were completed, were included in the analysis of the individual rotation
there was a round of practice in which all five cognitive angles, as angular disparity is not clearly defined for mir-
tasks were processed for 30 seconds in randomized order rored-reversed stimuli and this is the common way of
while performing the postural stability task. After the analysis in mental rotation studies (Jolicoeur et al., 1985;
practice trials, the test persons were given a sufficient Kaltner et al., 2017). The Friedman test was calculated
break and the ASS questionnaire was explained to them. to evaluate the ordinal data of the ASS questionnaire. To
correct for violations of sphericity the Greenhouse-
Statistical Analysis Geisser adjustment was used. Post hoc tests were
Bonferroni-corrected.
The processing of the CoP data was performed in
Matlab. The statistical analysis was performed using
SPSS. Only the CoP data of seconds 5 to 65 were eval- Results
uated, to avoid movements where the participant antici-
Mean Sway Parameter Values per Trial
pated the beginning or end of the trial (Hunter &
Hoffman, 2001) and still have a sufficiently long meas- A mean value was calculated over time for each par-
urement for getting reliable CoP-data (Carpenter et al., ameter per trial. Due to the above-mentioned irregular-
2001). All irregularities during the trial, such as speak- ities 4.2% of the data was missing. To handle missing
ing, coughing, scratching or similar, were noted and the data the mean value of the respective task was input for
respective trials were excluded from the analysis further analysis. To get one value for each condition, the
(Woollacott & Vander Velde, 2008). In a first analysis, a mean value for the two trials of each condition was cal-
mean value over time was calculated for each parameter culated. Table 1 shows the four different parameters for
per trial and then the course of the CoP was compared the five stimuli.
between different cognitive tasks using one-way repeated Regarding the mean amplitude of the CoP course the
measures analyses of variance. repeated measures ANOVA with “stimulus type” as fac-
To investigate the body sway during different mental tor revealed a statistically significant difference of the
rotation tasks in more depth (hypothesis 2), the body different stimulus types, (F(2.836, 235.392) ¼ 107.205, p

5
P. Hofmann and P. Jansen

TABLE 1. Mean CoP-parameters for the five cognitive tasks

Stimulus
Parameter Fixation cross Cube figures Embodied figures Egocentric Task Math Task
MA 5.44 (1.28) 3.97 (0.87) 3.94 (0.87) 3.87 (0.83) 4.01 (0.88)
SV 196.95 (32.30) 198.03 (32.36) 197.60 (33.11) 198.12 (31.34) 196.25 (32.72)
Range AP 25.00 (5.54) 19.27 (4.78) 18.83 (4.56) 18.45 (4.02) 19.87 (4.73)
Range ML 19.34 (4.32) 17.15 (3.60) 16.86 (3.48) 17.07 (3.94) 17.14 (3.95)

Note. Mean value (SD) for the two trials of each condition. All values are reported in [mm]. MA ¼ Mean amplitude,
SV ¼ Sway Velocity, Range AP ¼ Maximum range of CoP in anterior-posterior direction, Range ML ¼ Maximum range of
CoP in medio-lateral direction.

< .001, partial g2 ¼ .564). Bonferroni adjusted post hoc ASS Questionnaire
analysis showed that the mean deviation from the arith-
Figure 3 shows the results for the ASS-questionnaires.
metic mean point for the task where participants looked
The left side of the figure displays the cognitive effort
at the fixation cross was significantly higher than in all
and the right side the physical effort. Since each condi-
other tasks (all p < .001). All other tasks did not differ
tion was evaluated twice per person, the median of these
significantly from each other.
two evaluations has been calculated. Regarding the cog-
Regarding the sway velocity of the CoP course the
nitive effort significant differences between the different
repeated measures ANOVA with “stimulus type” as fac-
conditions (v2(4) ¼ 194.252, p < .001) could be
tor and Greenhouse-Geisser correction (Greenhouse-
revealed. Post hoc pairwise comparisons between all
Geisser ¼ .774) revealed no statistically significant dif-
conditions (Dunn-Bonferroni-tests) showed significant
ference between the different stimulus types, (F(3.096,
differences between all conditions (all p < .005) except
256.928) ¼ 1.935, p ¼ .122, partial g2 ¼ .023).
for the comparison between object-based mental rotation
Regarding the maximum range of CoP course in
task with human figures and egocentric mental rotation
anterior-posterior direction the repeated measures
task (p ¼ 1.000). The physical effort also differed signifi-
ANOVA with “stimulus type” as factor and
cant between the conditions (v2(4) ¼ 100.056, p <
Greenhouse-Geisser correction revealed a statistically
.001). The post-hoc Dunn-Bonferroni-tests showed that
significant difference between the different stimulus
only the fixation cross condition differed significantly
types, (F(3.609, 299.523) ¼ 64.455, p < .001, partial g2
from all other conditions (p < .001).
¼ .437). Bonferroni adjusted post hoc analysis showed
that during the task where participants looked at the fix-
Influence of Rotation Angles during Postural
ation cross the maximum range of CoP course in anter-
Stability Task
ior or posterior direction was significantly higher than
in all other tasks (all p < .001). Additionally, the max- Since the stimuli in the mental rotation tasks were dis-
imum range in anterior-posterior direction was statistic- played on a random basis, it was possible that individual
ally significant higher in the math task than in the test subjects were not shown certain angular rotations.
egocentric mental rotation task (p < 0.05). All other To deal with these missing data, the mean values per
tasks showed no statistically significant differences stimulus type and rotation angle were input. A total of
between them. 3.8% of the data was missing.
Regarding the maximum range of CoP course in Concerning the mean amplitude, see Figure 4, the
medio-lateral direction the repeated measures ANOVA repeated measures ANOVA showed a significant main
with “stimulus type” as factor and Greenhouse-Geisser effect for the factor “stimulus type” (F(1.814, 150.566)
correction revealed a statistically significant difference ¼ 52.564, p < .001, partial g2 ¼ .388), for the factor
between the different stimulus types, (F(3.114, 258.453) “rotation angle” (F(2.737, 227.147) ¼ 19.522, p < .001,
¼ 13.843, p < .001, partial g2 ¼ .143). Bonferroni partial g2 ¼ .190) and a significant interaction between
adjusted post hoc analysis showed that during the task these two factors (F(4.488, 372.509) ¼ 7.474, p < .001,
where participants looked at the fixation the maximum partial g2 ¼ .083). Additional simple main effect analysis
range of CoP course in medio-lateral direction was sig- was conducted to detect possible differences of the rota-
nificantly higher than in all other tasks (all p < .001). tion angles dependent on the type of stimulus. For the
All other tasks did not differ significantly from object-based task with human figures the mean amplitude
each other. of the body sway at the rotation angle 180 differed

6 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

FIGURE 3. Results of the ASS questionnaire.

significantly to 0 (p ¼ .003), to 60 (p ¼ .017) and to slopes of all three tasks did not differ significantly from
120 (p ¼ .049). For the egocentric mental rotation task, each other (F(1.484, 123.190) ¼ .385, p ¼ .619, partial
no statistical significant differences could be shown. In g2 ¼ .005)
the object-based task with cube figures the mean ampli- Concerning the maximum range of motion in anterior-
tude of the body sway at the rotation angle of 0 differed posterior direction, see Figure 6, the analysis of variance
significantly from all other rotation angles (p < .001). ruled out a significant main effect for the factor
The mean amplitude at the rotation angle of 60 differed “stimulus type” (F(1.669, 138.548) ¼ 42.282, p < .001,
significant from 120 (p ¼ .024). The mean slopes of all partial g2 ¼ .337), for the factor “rotation angle”
three tasks differed significantly from each other (F(2.591, 215.056) ¼ 24.146, p < .001, partial g2 ¼
(F(2,166) ¼ 8.898, p < .001, partial g2 ¼ .097). Post .225) and a significant interaction between these two fac-
hoc pairwise comparisons with Bonferroni correction tors, (F(4.343, 360.461) ¼ 10.026, p < .001, partial g2
revealed only significant differences between the mean ¼ .108). Additional simple main effect analysis was con-
slopes of the egocentric and the object-based task with ducted to analyse a possible difference dependent on
cube figures (p < .001). For a better understanding of rotation angle and type of stimulus. It was shown that
the influence of the stimulus type another simple main for the object-based task with human figures the max-
effect analysis was calculated for each angle. In Figure 4 imum range of motion in anterior-posterior direction at
this can be read as vertical difference. For the mean the rotation angle of 180 differed significantly from 0
amplitude of the body sway at the rotation angle 0 no (p < .001), 60 (p < .001) and 120 (p ¼ .005). For the
significant differences between the tasks could be shown. egocentric mental rotation task no statistical significant
At an angular disparity of 60 the object-based task with differences could be shown. In the object-based task
cube figures differed significant from the other two tasks with cube figures the maximum range of motion in
(p < .001). For the mean amplitude at the rotation angle anterior-posterior direction at the rotation angle of 0 dif-
of 120 again the object-based task with cube figures dif- fered significantly from all other rotation angles (p <
fered significant from the two other tasks (p < .001). At .001). For the rotation angle of 60 it differed significant
a rotation angle of the stimuli of 180 the mean ampli- from 120 (p ¼ .035). The mean slopes of all three tasks
tude of all three tasks showed significant differences differed significantly from each other (F(2,166) ¼
between each other (p < .05). 12.282, p < .001, partial g2 ¼ .129). Post hoc pairwise
Concerning the sway velocity, see Figure 5, the ana- comparisons with Bonferroni correction revealed signifi-
lysis of variance showed neither a significant main effect cant differences between the mean slopes of the object-
for the factor “stimulus type” (F(1.576, 130.815) ¼ .220, based task with human figures and the object-based task
p ¼ .750, partial g2 ¼ .003) nor for the factor “rotation with cube figures (p ¼ .013) as well as between the ego-
angle” (F(1.964, 162.980) ¼ .091, p ¼ .910, partial g2 ¼ centric task and the object-based task with cube figures (p
.001). Also the interaction between these two factors < .001). For a better understanding of the influence of the
revealed no statistical significant difference (F(2.530, stimulus type another simple main effect analysis was cal-
209.987) ¼ .588, p ¼ .595, partial g2 ¼ .007). The mean culated for each angle. In Figure 6 this can be read as

7
P. Hofmann and P. Jansen

FIGURE 4. Mean amplitude of CoP course per FIGURE 5. Sway velocity of CoP course per
stimulus type. stimulus type.

vertical difference. For the maximum range of motion in

anterior-posterior direction at the rotation angle of 0 no
significant differences between the tasks could be shown.
At an angular disparity of 60 the maximum span in anter-
ior-posterior direction differed significant for all tasks (p <
.05). For 120 the maximum span in anterior-posterior dir-
ection during the object-based task with cube figures was
significantly higher than in the two other tasks (p < .001).
At an angle of the stimuli of 180 the maximum span in
anterior-posterior direction of all three tasks showed signifi-
cant differences between each other (p < .05).
Concerning the maximum range of motion in medio-
lateral direction, see Figure 7, the analysis of variance
showed a significant main effect for the factor “stimulus
type” (F(1.722, 142.886) ¼ 65.367, p < .001, partial g2
¼ .441), for the factor “rotation angle” (F(2.580,
214.151) ¼ 16.754, p < .001, partial g2 ¼ .168) and a FIGURE 6. Maximum range of CoP course in anterior-
significant interaction between these two factors posterior direction per stimulus type.
(F(4.877, 404.787) ¼ 6.583, p < .001, partial g2 ¼
.073). The simple main effect analysis for the differences
of the rotation angles per type of stimulus demonstrated revealed only significant differences between the mean
that for the object-based task with human figures the slopes of the egocentric task and the object-based task with
maximum range of motion in medio-lateral direction at cube figures (p < .001). For a better understanding of the
the rotation angle of 180 differed significantly from 0 influence of the stimulus type another simple main effect
(p ¼ .006) and from 60 (p ¼ .019). For the egocentric analysis was calculated for each angle. In Figure 7 this can
mental rotation task, no statistical significant differences be read as vertical difference. For the rotation angle of 0 a
could be shown. In the object-based task with cube fig- significant higher range of the object-based task with cube
ures the maximum range of motion in medio-lateral direc- figures compared to the object-based task with human fig-
tion at the rotation angle of 0 differed significantly from ures could be shown (p ¼ .008). At an angular disparity of
all other rotation angles (p < .005). At the rotation angle 60 the maximum span in medio-lateral direction during
of 60 it differed significantly from 120 (p ¼ .004). The the object-based task with cube figures was significantly
mean slopes of the three tasks differed significantly from higher than in the two other tasks (p < .001). Also, for
each other (F(2,166) ¼ 8.853, p < .001, partial g2 ¼ .096). 120 the maximum span in medio-lateral direction during
Post hoc pairwise comparisons with Bonferroni correction the object-based task with cube figures was significantly

8 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

leads to postural stabilization compared to the fixation of

a point on the screen, we have extended this finding by
focusing on three different kinds of mental rotation tasks
(two object-based tasks, one egocentric task). Our results
provide evidence that all three tasks differ significantly
from a neutral condition, where participants stood still
and looked at a fixation cross. All mental rotation tasks
stabilized the body sway regarding the mean amplitude,
the maximum range of motion in anterior-posterior direc-
tion and in medio-lateral direction. It is critical to note
that the results of the ASS show that the fixation cross
task was perceived as significantly more physically
demanding than the other tasks. The extent to which
more strain actually occurred or whether this subjective
assessment only arose due to a shift in attention can only
be objectively controlled with a parallel EMG measure-
FIGURE 7. Maximum range of CoP course in medio- ment, which was not performed in this study.
lateral direction per stimulus type. However, our results as well as the one from Dault et al.
(2001) are in contrast to many studies regarding dual task
designs for balance experiments, where a simultaneous con-
ducted task (but no mental rotation task) often leads to a
higher than in the two other tasks (p < .001). At an angle deterioration of postural stability (Mujdeci et al., 2016;
of the stimuli of 180 the same result as for 60 and 120 Pellecchia, 2003; Shumway-Cook et al., 1997; Simoneau
was shown (p < .001), with the maximum range of motion et al., 1999). Dault et al. (2001) proposed the explanation
in medio-lateral direction of the object-based task with that their results come from a co-contraction control strat-
cube figures significantly higher than the other two tasks. egy of postural muscles of the central nervous system,
leading to a tighter control of postural sway. Another pos-
Discussion sible explanation could be that the improved postural stabil-
ity is the result of a shift of attention to the cognitive task
The aim of the study was to investigate the influence and therefore the automaticity and efficiency of the postural
of a mental rotation task with different kinds of stimuli control processes are enhanced (Huxhold et al., 2006;
on a simultaneously executed motor task, i.e., still Shumway-Cook & Woollacott, 2007). Richer et al. (2017)
upright standing. We showed less body sway, indicating investigated whether less body sway during the conduction
a more stable position, during solving mental rotation of cognitive tasks could be due to a stiffening strategy or
tasks than during standing and just looking at a fixation due to automatic postural control processes. As they found
cross (Stabilization - Hypothesis 1). Taking into account no muscle activity around the ankle joint and concluded
the angular disparities of the mental rotation tasks, our enhanced postural stability comes from automatic postural
results ruled out a more stable position for egocentric control processes, we also believe that our finding of pos-
mental rotation tasks than for object based mental rota-
tural stabilization comes from the shift of attention to the
tion tasks with cube figures but not for object-based
mental rotation or math task. During looking at the fixation
mental rotation tasks with human figures (Perspective -
cross task, not so much attention is needed and results in a
Hypothesis 2a). Considering the angular disparities,
more internal focus of postural control which leads to a
embodied stimuli showed significantly reduced body
deterioration of postural stability (Wulf et al., 2001). Either
sway compared to the classical cube figures
way, our results show that a simultaneous conducted men-
(Embodiment - Hypothesis 2b). Additionally, it was
tal rotation task leads to increased postural stabilization
shown that within a specific mental rotation task the
compared to the fixation of a cross, resulting in a smaller
angular disparity of the stimuli has an influence on body
CoP displacement. Furthermore, Dault et al. (2001) investi-
sway only for object-based mental rotation tasks but not
gated two other working memory tasks and found no dif-
for egocentric ones (Angular disparity - Hypothesis 2c).
ferences in body sway between both tasks and the mental
rotation task. We also found no significant difference
Stabilization during Mental Rotation
between a math task and the mental rotation tasks concern-
Regarding the influence of a simultaneous executed ing the body sway. This might imply that only the conduc-
mental rotation task on body sway, our results are in line tion of a secondary task leads to postural stabilization,
with those from Dault et al. (2001). While they showed regardless of the type of the task. Taking into account the
that an egocentric mental rotation task with a linesman results of the ASS questionnaire and those from Pellecchia

9
P. Hofmann and P. Jansen

(2003), who showed that postural sway enlarges with embodiment (Wilson, 2002). The already discussed
increasing difficulty of the concurrent cognitive task, we results, with an egocentric task leading to a more stable
would have expected the highest body sway for the math position then an object-based task with cube figures, pro-
task. As there is no difference in body sway between the vide first evidence for the effect of embodied figures on
math task and the mental rotation tasks this has to be postural sway. If we take the several angular disparities
investigated further. into account, our results clearly show an impact of
embodied stimuli, resulting in less body sway measured
Differences in Postural Stability for Different Types with the CoP displacement parameters (Mean amplitude,
of Mental Rotation Tasks (Perspective Hypothesis) maximum span in anterior/posterior and medio/lateral
direction). This result is consistent with the statements of
Furthermore, our results are not in accordance with
Kawasaki and Higuchi (2013), even though their actual
hypothesis 2a. We assumed that during an egocentric
finding concerning the sway parameter "sway velocity"
task the subject has to imagine rotating its own body
could not be replicated. They described a decrease in
(Kessler & Rutherford, 2010) and because of the close-
sway velocity, in their words as “stabilization of body
ness of egocentric tasks to kinesthetic imagery tasks (see
sway”, after the egocentric mental rotation of embodied
explanation in the introduction), we expected that solving
stimuli (images of feet) compared to mental rotation of
an egocentric mental rotation task causes more body sway
car images. We did not find any change in sway velocity
than an object based mental rotation task. But for the angu-
but a reduction in the already mentioned parameters,
lar disparities of 60 , 120 and 180 the mean amplitude of
indicating a reduction of body sway. It is questionable
sway, the maximum span in anterior-posterior direction and
whether the results are comparable to Kawasaki and
the maximum span in medio-lateral direction were more
Higuchi (2013) because we checked for the differences
stable in the egocentric condition than in the object -based
during object-based mental rotation tasks while they
condition with cube figures. For zero degrees, no differen-
investigated egocentric tasks. Since the use of human
ces were detectable in these parameters which might be
figures in mental rotation tasks is a common way to
caused by the easiness of the task. Additionally, the mean
induce embodiment, there is reason to believe that the
slopes, i.e., the ratio how much the body sway increases
use of human figures as stimuli leads to a reduction of
per angle, for the mean amplitude of sway, the maximum
body sway by triggering a sensorimotor simulation
span in anterior-posterior direction and the maximum span
mechanism (Voyer & Jansen, 2016). Furthermore there
in medio-lateral direction were significant smaller in the
is evidence that it is more easy for participants to men-
egocentric condition than in the object-based condition with
tally rotate embodied stimuli than abstract shapes, like
cube figures. In line with Kessler and Rutherford (2010),
cube figures (Amorim et al., 2006; Jansen et al., 2012).
one possible explanation, why our hypothesis 2a is not
This confronts us with the issue of different task difficul-
true, is the differently perceived mental rotation task diffi-
ties which may overshadow an embodiment effect. In the
culty. For all angular disparities of the mental rotation
ASS questionnaire the object-based task with cube fig-
tasks, which are different from zero, we observed the pat-
ures was rated as significantly more difficult than the
tern that the egocentric task leads to a stabilization of the
object-based task with human figures. Therefore it is
body sway compared to the object-based task with cube
possible that the higher body sway only appears during
figures. However, the latter is also perceived as signifi-
the cube figures condition because the task was per-
cantly more difficult than the egocentric task, indicating
ceived as subjectively more difficult (Pellecchia, 2003).
that in this case the difficulty of the task has an impact on
For future experiments the task difficulty should be
body sway. This might be an explanation because the study
determined in a pre-test to guarantee for equal task diffi-
of Pellecchia (2003) showed increasing postural sway with
culty. The fact that the results suggest an influence of
increasing difficulty of the concurrent cognitive task. These
embodied stimuli, despite the different task difficulty,
observations are further confirmed by our finding that the
offers potential for further investigations in this direction.
egocentric task and the object-based task with human fig-
As seen in Kawasaki and Higuchi (2016) a stabilization
ures were classified as equally difficult. Between both
in body sway can be shown directly after the mental
tasks, there was no difference in body sway, apart from a
rotation of feet stimuli but not of hand stimuli. As the
few exceptions (mean amplitude: 180 ; maximum span in
feet and the ankles are essential parts in the postural con-
anterior-posterior direction: 60 and 180 ).
trol of human upright standing (Gage et al., 2004;
Winter et al., 1998; 2003), the authors conclude that the
The Relevance of Embodiment
mental rotation of such body parts relates to the ability
(Embodiment Hypothesis)
to stand as still as possible. Our results show that a sta-
Another assumption (Hypothesis 2b) was that there is bilization also occurs while rotating whole body figures.
a difference in body sway between mentally rotating Now the question arises whether this is due to the fact
cube figures and human figures caused by the effect of that whole body figures also include the feet or whether

10 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

perhaps body parts in general, which are involved in (Kessler & Rutherford, 2010), it still deserves to be
maintaining an upright posture, contribute to postural sta- examined more precisely in further studies.
bilization. This should be systematically investigated in
further studies. First, the stimuli used in Kawasaki (feet Limitations
and hands) should also be examined during a simultan- There were several limitations in this study. The major
eous conducted postural task. Further, it would be pos- limitation is that the difficulty in the tasks were rated as
sible to examine whole body stimuli with covered parts different. In order to make reliable statements, the mental
of the body (either covering it piece by piece starting rotation tasks must be perceived as equally difficult in
from the feet or starting from the head). This could their complexity, to eliminate the effect of cognitive
reveal which body part in a mental rotation task is essen- effort on body sway (Pellecchia, 2003). Our sample con-
tial for postural stabilization and therefore which parts of sisted of healthy sports students who were instructed to
the body are relevant to cause this effect. stand still on both legs. This task might have been sim-
ply too easy for this target group. If one wants to stick
Influence of Angular Disparity (Angular to the two-legged stand as a task, since it corresponds
Disparity Hypothesis) very closely to everyday life, then other target groups,
The more explorative approach to investigate the influ- such as older people or patients with vestibular diseases,
ence of angular disparity of the mental rotation stimuli would also be interesting. Due to the many cognitive
within the mental rotation tasks (Hypothesis 2c) revealed tasks and the fact of preventing fatigue during the pos-
that in egocentric tasks, angular disparity has no influ- tural task, the mental rotation experiments consisted of
ence on postural sway. For the object-based tasks there very few items compared to classical mental rotation
studies. Further studies should concentrate on more spe-
is a tendency that higher rotation angles lead to more
cific types of mental rotation tasks with more trials.
postural sway. Here a parallel to the common behavioral
Additionally there are directions which implicate, that
results of mental rotation tasks can be seen, as the clas-
variability in CoP course is functional, task-dependent
sical mental rotation paradigm of Shepard and Metzler
and cannot be claimed as an unstable stance, as long as
(1971) shows a linear increase in reaction time with
it doesn’t cause a loss of balance or a decreased per-
increasing angular disparity between two presented stim-
formance of standing (Haddad et al., 2013). For further
uli. As this means, physically, the mental rotation speed
studies, it could be interesting to consider more ways of
is approximately constant, this could be an explanation measuring postural stability.
for the inconsistency of the CoP velocity and displace-
ment data. In line with the constant mental rotation Summary
speed, the sway velocity shows no significant differen-
ces. On the other hand, similar to the increasing angular In summary, this study investigated the relation
disparity, the body sway data shows higher values for between mental rotation and postural stability. Our
CoP displacement parameters (i.e., spatial disparity) for results clearly show that the simultaneous conduction of
larger angles in object-based tasks. This explanatory a mental rotation task stabilizes the postural sway in a
approach is still very speculative but offers great poten- both-legged narrow stance. Furthermore, the stabilizing
tial for further investigations. The more or less similar influence of embodied stimuli compared to cube figures
increase in body sway during object-based mental rota- was observed. In addition, an effect of the angular dis-
tion task might also be a further hint for the common parity of the stimuli on body sway for object-based men-
process theory (Wohlschl€ager & Wohlschl€ager, 1998). A tal rotation tasks has been discovered. Future studies
should investigate the influence of simultaneous mental
further confirmation is that the rather classical increase
rotation tasks on body sway for different mental rotation
of response time with increasing angular disparity is usu-
stimuli and in different target groups, as this will give
ally more pronounced in object-based tasks as compared
more precise insights in the interaction between mental
to egocentric mental rotation tasks (Jola & Mast, 2005;
rotation and postural stability.
Michelon & Zacks, 2006). Also, this holds true for our
findings regarding the body sway as we see no differen-
ces in sway between the angles in the egocentric task. Conflict of Interests
This finding is partly comparable with the classical The authors declare to have no conflicts of interest
behavioural finding for egocentric mental rotation tasks
where reaction times tend to increase for angles above
60 or 90 (Keehner et al., 2006; Michelon & Zacks, REFERENCES
2006). Although the lack of increase contradicts our Amorim, M.-A., Isableu, B., & Jarraya, M. (2006). Embodied
hypothesis that egocentric tasks lead to more body sway spatial transformations: "body analogy" for the mental rota-
due to the imagination of rotating one’s own body tion of objects. Journal of Experimental Psychology.

11
P. Hofmann and P. Jansen

General, 135(3), 327–347. https://doi.org/10.1037/0096-3445. lifespan. Exercise and Sport Sciences Reviews, 41(2),
135.3.327 123–132. https://doi.org/10.1097/JES.0b013e3182877cc8
Amorim, M.-A., & Stucchi, N. (1997). Viewer- and object-cen- Hufschmidt, A., Dichgans, J., Mauritz, K. H., & Hufschmidt,
tered mental explorations of an imagined environment are M. (1980). Some methods and parameters of body sway
not equivalent. Cognitive Brain Research, 5(3), 229–239. quantification and their neurological applications. Archiv fur
https://doi.org/10.1016/S0926-6410(96)00073-0 Psychiatrie und Nervenkrankheiten, 228(2), 135–150. https://
Andersson, G., Hagman, J., Talianzadeh, R., Svedberg, A., & doi.org/10.1007/BF00365601
Larsen, H. C. (2002). Effect of cognitive load on postural Hunter, M. C., & Hoffman, M. A. (2001). Postural control:
control. Brain Research Bulletin, 58(1), 135–139. https://doi. Visual and cognitive manipulations. Gait & Posture, 13(1),
org/10.1016/S0361-9230(02)00770-0 41–48. https://doi.org/10.1016/S0966-6362(00)00089-8
Bauer, R., Jost, L., G€unther, B., & Jansen, P. (2021). Huxhold, O., Li, S.-C., Schmiedek, F., & Lindenberger, U.
Pupillometry as a measure of cognitive load in chronometric (2006). Dual-tasking postural control: Aging and the effects
mental rotation tasks with abstract and embodied figures. of cognitive demand in conjunction with focus of attention.
[Manuscript submitted for publication]. University of Brain Research Bulletin, 69(3), 294–305. https://doi.org/10.
Regensburg. 1016/j.brainresbull.2006.01.002
B€usch, D., Pabst, J., Naundorf, F., Braun, J., Marschall, F., Jansen, P., & Kaltner, S. (2014). Object-based and egocentric
Schumacher, K., Wilhelm, A., & Granacher, U. (2015). mental rotation performance in older adults: The importance
Subjektive Beanspruchung im Krafttraining. Krafttraining: of gender differences and motor ability. Neuropsychology,
Kraftvoll Durchs Leben»: Jahresta Gung Der DvsSektion Development, and Cognition. Section B, Aging,
Trainingswissenschaft Vom 28.–30. Mai 2015 in Potsdam Neuropsychology and Cognition, 21(3), 296–316. https://doi.
(Abstractband), 13. org/10.1080/13825585.2013.805725
Carpenter, M. G., Frank, J. S., Winter, D. A., & Peysar, G. W. Jansen, P., Lehmann, J., & van Doren, J. (2012). Mental rota-
(2001). Sampling duration effects on centre of pressure sum- tion performance in male soccer players. PLoS One, 7(10),
mary measures. Gait & Posture, 13(1), 35–40. https://doi. e48620. https://doi.org/10.1371/journal.pone.0048620
org/10.1016/S0966-6362(00)00093-X Jeka, J. J., & Lackner, J. R. (1994). Fingertip contact influences
Clapp, S., & Wing, A. M. (1999). Light touch contribution to bal- human postural control. Experimental Brain Research,
ance in normal bipedal stance. Experimental Brain Research, 100(3), 495–502. https://doi.org/10.1007/BF02738408
125(4), 521–524. https://doi.org/10.1007/s002210050711 Jeong, B. Y. (1991). Respiration effect on standing balance.
Conrad, B., & Sch€onle, P. (1979). Speech and respiration. Archives of Physical Medicine and Rehabilitation, 72(9),
Archiv fur Psychiatrie und Nervenkrankheiten, 226(4), 642–645.
251–268. https://doi.org/10.1007/BF00342238 Jola, C., & Mast, F. W. (2005). Mental object rotation and ego-
Dault, M. C., Frank, J. S., & Allard, F. (2001). Influence of a centric body transformation: Two dissociable processes?
visuo-spatial, verbal and central executive working memory Spatial Cognition & Computation, 5(2-3), 217–237. https://
task on postural control. Gait & Posture, 14(2), 110–116. doi.org/10.1080/13875868.2005.9683804
https://doi.org/10.1016/S0966-6362(01)00113-8 Jolicoeur, P., Regehr, S., Smith, L. B. J. P., & Smith, G. N.
Dault, M. C., Yardley, L., & Frank, J. S. (2003). Does articula- (1985). Mental rotation of representations of two-dimen-
tion contribute to modifications of postural control during sional and three-dimensional objects. Canadian Journal of
dual-task paradigms? Cognitive Brain Research, 16(3), Psychology/Revue Canadienne de Psychologie, 39(1),
434–440. https://doi.org/10.1016/S0926-6410(03)00058-2 100–129. https://doi.org/10.1037/h0080118
Di Rienzo, F., Collet, C., Hoyek, N., & Guillot, A. (2014). Jost, L., & Jansen, P. (2020). A novel approach to analyzing all
Impact of neurologic deficits on motor imagery: A system- trials in chronometric mental rotation and description of a flex-
atic review of clinical evaluations. Neuropsychology Review, ible extended library of stimuli. Spatial Cognition &
24(2), 116–147. https://doi.org/10.1007/s11065-014-9257-6 Computation, 20(3), 234–256. https://doi.org/10.1080/13875868.
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). 2020.1754833
Gpower 3: A flexible statistical power analysis program for Kaltner, S., & Jansen, P. (2018). Sex of human stimulus matters:
the social, behavioral, and biomedical sciences. Behavior Female and male stimuli are processed differently in mental
Research Methods, 39(2), 175–191. https://doi.org/10.3758/ rotation tasks. Journal of Cognitive Psychology, 30(8),
bf03193146 854–862. https://doi.org/10.1080/20445911.2018.1530677
Gage, W. H., Winter, D. A., Frank, J. S., & Adkin, A. L. Kaltner, S., Jansen, P., & Riecke, B. E. (2017). Stimulus size mat-
(2004). Kinematic and kinetic validity of the inverted pendu- ters: Do life-sized stimuli induce stronger embodiment effects
lum model in quiet standing. Gait & Posture, 19(2), in mental rotation? Journal of Cognitive Psychology, 29(6),
124–132. https://doi.org/10.1016/S0966-6362(03)00037-7 701–716. https://doi.org/10.1080/20445911.2017.1310108
Ganczarek, J., Ruggieri, V., Nardi, D., & Olivetti Belardinelli, Kawasaki, T., & Higuchi, T. (2013). Immediate beneficial
M. (2015). Intersection of reality and fiction in art percep- effects of mental rotation using foot stimuli on upright pos-
tion: Pictorial space, body sway and mental imagery. tural stability in healthy participants. Rehabilitation Research
Cognitive Processing, 16, 233–236. https://doi.org/10.1007/ and Practice, 2013, 890962. https://doi.org/10.1155/2013/
s10339-015-0702-0 890962
Haddad, J. M., Rietdyk, S., Claxton, L. J., & Huber, J. E. Kawasaki, T., & Higuchi, T. (2016). Improvement of postural
(2013). Task-dependent postural control throughout the stability during quiet standing obtained after mental rotation

12 Journal of Motor Behavior

 The Relation of Mental Rotation and Postural Stability

of foot stimuli. Journal of Motor Behavior, 48(4), 357–364. Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-
https://doi.org/10.1080/00222895.2015.1100978 dimensional objects. Science, 171(3972), 701–703. https://
Kawasaki, T., Yasuda, K., Fukuhara, K., & Higuchi, T. (2014). doi.org/10.1126/science.171.3972.701
Relationship between mental rotation of body parts and pos- Shumway-Cook, A., Woollacott, M., Kerns, K. A., & Baldwin,
tural stability during quiet stance. Journal of Imagery M. (1997). The effects of two types of cognitive tasks on
Research in Sport and Physical Activity, 9(1), 39–46. https:// postural stability in older adults with and without a history
doi.org/10.1515/jirspa-2014-0001 of falls. The Journals of Gerontology. Series A, Biological
Keehner, M., Guerin, S. A., Miller, M. B., Turk, D. J., & Sciences and Medical Sciences, 52(4), M232–M240. https://
Hegarty, M. (2006). Modulation of neural activity by angle doi.org/10.1093/gerona/52a.4.m232
of rotation during imagined spatial transformations. Shumway-Cook, A., & Woollacott, M. H. (2007). Motor con-
NeuroImage, 33(1), 391–398. https://doi.org/10.1016/j.neuro- trol: Translating research into clinical practice. Lippincott
image.2006.06.043 Williams & Wilkins.
Kessler, K., & Rutherford, H. (2010). The two forms of visuo- Simoneau, M., Teasdale, N., Bourdin, C., Bard, C., Fleury, M.,
spatial perspective taking are differently embodied and sub- & Nougier, V. (1999). Aging and postural control: Postural
serve different spatial prepositions. Frontiers in Psychology, perturbations caused by changing the visual anchor. Journal
1, 1–12. https://doi.org/10.3389/fpsyg.2010.00213 of the American Geriatrics Society, 47(2), 235–240. https://
Michelon, P., & Zacks, J. M. (2006). Two kinds of visual per- doi.org/10.1111/j.1532-5415.1999.tb04584.x
spective taking. Perception & Psychophysics, 68(2), Steggemann, Y., Engbert, K., & Weigelt, M. (2011). Selective
327–337. https://doi.org/10.3758/bf03193680 effects of motor expertise in mental body rotation tasks:
Mujdeci, B., Turkyilmaz, D., Yagcioglu, S., & Aksoy, S. Comparing object-based and perspective transformations.
(2016). The effects of concurrent cognitive tasks on postural Brain and Cognition, 76(1), 97–105. https://doi.org/10.1016/
sway in healthy subjects. Brazilian Journal of j.bandc.2011.02.013
Otorhinolaryngology, 82(1), 3–10. https://doi.org/10.1016/j. Stins, J. F., Schneider, I. K., Koole, S. L., & Beek, P. J.
bjorl.2015.10.011 (2015). The influence of motor imagery on postural sway:
Palmieri, R. M., Ingersoll, C. D., Stone, M. B., & Krause, B. A. Differential effects of type of body movement and person
(2002). Center-of-pressure parameters used in the assessment of perspective. Advances in Cognitive Psychology, 11(3),
postural control. Journal of Sport Rehabilitation, 11(1), 51–66. 77–83. https://doi.org/10.5709/acp-0173-x
https://doi.org/10.1123/jsr.11.1.51 Voyer, D., & Jansen, P. (2016). Sex differences in chronometric
Pellecchia, G. L. (2003). Postural sway increases with atten- mental rotation with human bodies. Psychological Research,
tional demands of concurrent cognitive task. Gait & Posture, 80(6), 974–984. https://doi.org/10.1007/s00426-015-0701-x
18(1), 29–34. https://doi.org/10.1016/S0966-6362(02)00138-8 Voyer, D., & Jansen, P. (2017). Motor expertise and performance
Pietsch, S., & Jansen, P. (2012). Different mental rotation per- in spatial tasks: A meta-analysis. Human Movement Science,
formance in students of music, sport and education. Learning 54, 110–124. https://doi.org/10.1016/j.humov.2017.04.004
and Individual Differences, 22(1), 159–163. https://doi.org/ Vuillerme, N., Nougier, V., & Teasdale, N. (2000). Effects of a
10.1016/j.lindif.2011.11.012 reaction time task on postural control in humans.
Potvin-Desrochers, A., Richer, N., & Lajoie, Y. (2017). Neuroscience Letters, 291(2), 77–80. https://doi.org/10.1016/
Cognitive tasks promote automatization of postural control in S0304-3940(00)01374-4
young and older adults. Gait & Posture, 57, 40–45. https:// Wexler, M., Kosslyn, S. M., & Berthoz, A. (1998). Motor proc-
doi.org/10.1016/j.gaitpost.2017.05.019 esses in mental rotation. Cognition, 68(1), 77–94. https://doi.
Rhea, C. K., Kiefer, A. W., Haran, F. J., Glass, S. M., & org/10.1016/S0010-0277(98)00032-8
Warren, W. H. (2014). A new measure of the CoP trajectory Wilson, M. (2002). Six views of embodied cognition.
in postural sway: Dynamics of heading change. Medical Psychonomic Bulletin & Review, 9(4), 625–636. https://doi.
Engineering & Physics, 36(11), 1473–1479. https://doi.org/ org/10.3758/bf03196322
10.1016/j.medengphy.2014.07.021 Winter, D. A. (1995). Human balance and posture control dur-
Rhea, C. K., Kiefer, A. W., Wright, W. G., Raisbeck, L. D., & ing standing and walking. Gait & Posture, 3(4), 193–214.
Haran, F. J. (2015). Interpretation of postural control may https://doi.org/10.1016/0966-6362(96)82849-9
change due to data processing techniques. Gait & Posture, Winter, D. A., Patla, A. E., Ishac, M., & Gage, W. H. (2003).
41(2), 731–735. Motor mechanisms of balance during quiet standing. Journal
Richer, N., & Lajoie, Y. (2019). Cognitive task modality influ- of Electromyography and Kinesiology, 13(1), 49–56. https://
ences postural control during quiet standing in healthy older doi.org/10.1016/S1050-6411(02)00085-8
adults. Aging Clinical and Experimental Research, 31(9), Winter, D. A., Patla, A. E., Prince, F., Ishac, M., & Gielo-
1265–1270. https://doi.org/10.1007/s40520-018-1068-9 Perczak, K. (1998). Stiffness control of balance in quiet
Richer, N., Saunders, D., Polskaia, N., & Lajoie, Y. (2017). The standing. Journal of Neurophysiology, 80(3), 1211–1221.
effects of attentional focus and cognitive tasks on postural https://doi.org/10.1152/jn.1998.80.3.1211
sway may be the result of automaticity. Gait & Posture, 54, Wohlschl€ager, A., & Wohlschl€ager, A. (1998). Mental and
45–49. https://doi.org/10.1016/j.gaitpost.2017.02.022 manual rotation. Journal of Experimental Psychology.
Rodrigues, E. C., Lemos, T., Gouvea, B., Volchan, E., Human Perception and Performance, 24(2), 397–412.
Imbiriba, L. A., & Vargas, C. D. (2010). Kinesthetic motor https://doi.org/10.1037/0096-1523.24.2.397
imagery modulates body sway. Neuroscience, 169(2), Woollacott, M., & Vander Velde, T. (2008). Non-visual spatial
743–750. https://doi.org/10.1016/j.neuroscience.2010.04.081 tasks reveal increased interactions with stance postural

13
P. Hofmann and P. Jansen

control. Brain Research, 1208, 95–102. https://doi.org/10. Psychology Section A, 54(4), 1143–1154. https://doi.org/10.
1016/j.brainres.2008.03.005 1080/713756012
Wraga, M., Creem, S. H., & Proffitt, D. R. (1999). The influ- Yardley, L., Gardner, M., Leadbetter, A., & Lavie, N. (1999).
ence of spatial reference frames on imagined object- and Effect of articulatory and mental tasks on postural control.
viewer rotations. Acta Psychologica, 102(2-3), 247–264. NeuroReport, 10(2), 215–219. https://doi.org/10.1097/
https://doi.org/10.1016/S0001-6918(98)00057-2 00001756-199902050-00003
Wraga, M., Shephard, J. M., Church, J. A., Inati, S., & Zacks, J. M., Mires, J., Tversky, B., & Hazeltine, E. (2000).
Kosslyn, S. M. (2005). Imagined rotations of self versus Mental spatial transformations of objects and perspective.
objects: An fMRI study. Neuropsychologia, 43(9), Spatial Cognition and Computation, 2(4), 315–332. https://
1351–1361. https://doi.org/10.1016/j.neuropsychologia. doi.org/10.1023/A:1015584100204
2004.11.028
Wulf, G., McNevin, N., & Shea, C. H. (2001). The automati- Received October 8, 2020
city of complex motor skill learning as a function of atten- Revised January 25, 2021
tional focus. The Quarterly Journal of Experimental Accepted March 1, 2021

14 Journal of Motor Behavior