Advances in Neurodevelopmental Disorders

https://doi.org/10.1007/s41252-023-00314-9

ORIGINAL PAPER

Fluctuations of the Center of Pressure in Autism Spectrum Disorder

Naomi Tsugita1,2 · Shino Ogawa1 · Nao Maki1 · Zu Soh3 · Toshio Tsuji3 · Yasuko Funabiki1

Accepted: 7 January 2023

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023

Abstract
Objectives  The static standing position of autism spectrum disorder (ASD) is unstable. However, the cause has not been
clarified. We will investigate the fluctuation of center of pressure (COP) by detrended fluctuation analysis (DFA) and con-
tribute to the elucidation of the cause in the future.
Method  We investigated the characteristics of fluctuations in the COP in 16 individuals with ASD and 13 individuals with
typical development (TD). All participants stood on a Wii Balance Board for 70 s during which time we obtained COP data
at 100 Hz. The eyes-open and eyes-closed conditions were performed once each. We obtained the locus length, total locus
length, outer peripheral area, and the mean value and standard deviation of the coordinate position, and also calculated the
mean value, standard deviation, coefficient variability, and alpha index of velocity from the acquired data, which we used
as evaluation indices.
Results  The locus lengths in the mediolateral and anteroposterior directions and the total length, as well as the outer periph-
eral area of the COP, found no significant differences between the groups. The alpha index showing the strength of long-term
correlations of the standing position by DFA of moving distance per 100 Hz in the ASD group was significantly larger than
that in the TD group (p = .011) in the anteroposterior direction under the eyes-closed condition.
Conclusions  Postural sway in the ASD group did not differ from TD but was maintained from a long-term perspective.

Keywords  ASD · Posture · Sway · Fractal · Stability

Autism spectrum disorder (ASD) is a complex develop-
mental condition involving persistent challenges with social
communication, restricted interests, and repetitive behavior
(American Psychiatric Association: APA, 2022). It is char-
acterized by impairments in communication and social inter-
action, as well as stereotypic behaviors, sensory problems
such as hypersensitivity and bluntness, and motion problems
such as clumsy and awkward movements (Baranek et al.,
2006; Dziuk et al., 2007; Ornitz et al., 1977).
Postural sway has been evaluated not only for the magni-
tude of the sway itself, but also for its fluctuation (Doumas
* Yasuko Funabiki
funabiki.yasuko.8a@kyoto-u.ac.jp
1
Graduate School of Human and Environmental Studies,
Kyoto University, Yoshida‑Nihonmatsu‑Cho, Sakyo‑Ku,
Kyoto, Japan
2
Research Fellow of the Japan Society for the Promotion
of Science, Tokyo, Japan
3
Graduate School of Advanced Science and Engineering,
Hiroshima University, Hiroshima, Japan
et al., 2016). Previous studies on postural fluctuations have
focused on fractal fluctuations in the center of pressure
(COP) to evaluate postural control strategies. Fractal fluctua-
tion in postural sway is a long-term correlation that indicates
the extent to which the current sway is affected by past sway.
The COP in the standing position comprises a fine sway
including fractal characteristics. Several studies have been
conducted on individuals with typical development (TD).
A previous study showed that the fractal characteristics of
the COP in healthy elderly individuals were weaker than
those in young individuals (Duarte & Sternad, 2008). In
contrast, Minamisawa et al. (2009) showed that the fractal
fluctuations of the COP did not differ significantly between
individuals with Parkinson’s disease with postural instabili-
ties similar to those of ASD and healthy individuals, and
that fractal fluctuation can be evaluated on the basis of the
alpha index calculated by detrended fluctuation analysis. The
normal alpha index has a positive value between 0.5 and 1.5
indicates completely random (white noise), while a value
of 1.0 indicates 1/f fractal (pink noise), and a value of 1.5
indicates random (brown) noise. The closer the alpha index

13
Vol.:(0123456789)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.


is to 1.0, the stronger the long-term correlation (Hausdorff
et al., 1995; Peng et al., 1995). In this case, it is easier to
predict future sway from past sway and to control postural
sway consistently. Conversely, as the alpha index approaches
0.5 or 1.5, postural sway has a weaker long-term correlation,
and there is no controlling law. It is difficult to reflect past
sway in the current sway, resulting in poor postural control.
In a study on children, Dziuk et al. (2007) suggested that
movement disorders are widely associated with social and
communicative behavioral disorders. Kroeger et al. (2007)
also reported that children with ASD improved their prosocial
behavior by playing ball, pretend play, and gameplay. These
findings suggest that the motor function in ASD is related to
and affects quality of life (QOL) and well-being. However, the
reason for this low motor function remains unclarified.
In the present study, we aimed to clarify the characteris-
tics of the posture maintenance function of ASD by compar-
ing it with that of individuals with TD. In addition, on the
basis of previous research results, we hypothesized that pos-
tural sway is greater in individuals with ASD than in those
with TD, and that fluctuations are constant and unstable.
Method
Participants
This study included 29 adult participants, of whom 16 had ASD
(male: 8; female: 8; age: 32.7 ± 7.5 years) and 13 had TD (male:
7; female: 6; age: 31.5 ± 9.6 years). We recruited the individuals
with ASD from outpatient clinics at in our hospital and the
community and the individuals with TD from the community.
We calculated the Full-Scale intelligence quotient (IQ), Verbal
IQ, and Performance IQ using the Wechsler Adult Intelligence
Scale-Third Edition (WAIS-III; Wechsler, 1997). To assess ASD
characteristics, we used the Autism Diagnostic Observation
Schedule, Second Edition (ADOS-2; Lord et al., 2012), which
includes communication, mutual interpersonal relationships,
communication and interpersonal relationships, and imagination
and creativity. To further assess neurodevelopmental disorders,
including ASD, attention-deficit/hyperactivity disorder
(ADHD), developmental coordination disorder (DCD), and
learning disabilities (LD), we used the Multi-dimensional
Scale for pervasive developmental disorder (PDD) and ADHD
(MSPA) (Funabiki et al., 2011). The MSPA comprises 14
domains of clinical and behavioral features, including five core
features of PDD, three ADHD, two developmental coordination
disorders, and four problem areas (Funabiki et  al., 2011;
Funabiki & Shiwa, 2018).
A psychiatric doctor confirmed the diagnosis of ASD
using the Diagnostic and Statistical Manual of Mental Dis-
orders, Fourth Edition Text Revision (APA, 2013). In addi-
tion, all participants completed questionnaires regarding
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Advances in Neurodevelopmental Disorders
medications, dizziness, history of falling, and visual acuity.
We conducted the experiments after confirming that all par-
ticipants, except those taking anti-vertigo medication, did
not have abnormal vision or risk of falling.
Procedure
The participants stood on a Wii Balance Board (Nintendo,
Japan) in a resting position for 70 s. They were instructed
to stand as still as possible with their legs aligned and to
keep their arms relaxed and naturally on the sides of their
body. The measurement room was quiet, had 380 lx illu-
minance, and had a flat floor. The measurement conditions
were as follows: (1) eyes open (EO): with eyes open, the
participants focused on a point light source set 2 m in front
of the participant adjusted to the height of each partici-
pant’s eyes; (2) eyes closed (EC). The measurements were
taken once for each condition with a 30-s pause between
each condition. The Wii Balance Board, which has verified
accuracy and precision (Clark et al., 2010), was connected
to a desktop computer via Bluetooth in the 2.4 GHz band,
and the barycentric coordinates and weights were obtained
at a sampling rate of 100 Hz using a library provided by the
open-source project Wii Brew. To remove the initial effect
when participants first stood on the Wii Balance Board, we
deleted data from the first 20 s and outliers. We defined
outliers as obviously large movements or movements in the
opposite direction. We obtained two time-series data for x in
the mediolateral (ML) direction and y in the anteroposterior
(AP) direction and defined the time length as N.
Measures
Index of the Center of Pressure
We calculated each locus length (LL) from time-series data
x and y values (Ushiyama et al., 2008), total locus length
(TLL), and the outer peripheral area (OPA) (Imaoka et al.,
1997).
LL: the total distance moved by the COP in each direction
Δx(i) = x(i+1) − x(i)
∑N−1 (i = 1, 2, 3, … , N)
LL (mm) = i=1 �Δx(i) �
y was also evaluated in the AP direction.
TLL: the total distance moved by the COP
∑N−1 √
TLL(mm) = Δx(i) 2 + Δy(i) 2
i=1
OPA:
We defined the origin of the OPA as the calculated mean
of the coordinates in each direction. The mean value was
Advances in Neurodevelopmental Disorders

then subtracted from the original time-series data to create cumulative sum of the deviations from the first point of x to
new time-series data. that data point.
1 ∑N ∑k
X = x(i) − x(i) S(k) = i=1 [x(i) − x] (k = 1, 2, 3, … , N)
N
1 ∑Ni=1
Y = y(i) − y
i=1 (i)
N Subsequently, we divided S(k) into equal-sized segments
First, we divided the circumference of the origin into 120 with a time length of m, where m ranged from ten to half
equal parts at 3°. Next, we selected the coordinate point the length of S(k) (Delignières & Marmelat, 2014). Each
(point P(n)) with the longest distance from the origin for segment is expressed as S(m; t), where t is the number of
each area. The distance between point P(n) and the origin was segments. We then cut off the surplus data by dividing S(k)
represented by r(n). Furthermore, we calculated the area of a by m. Next, we calculated the least-squares approximate
triangle (T(n)) connecting the points P(n), P(n+1), and the ori- straight line Strend(m; t) for each segment and subtracted it
gin and integrated all areas of the triangles. Because sin 3° is from S(m; t) (detrend). By recombining the detrended seg-
close to the true value when calculating the OPA (Asai et al., ments, we created a new time-series data SS with a time
2018; Imaoka et al., 1997), we adopted sin 3° in this study. length C composed of each time point j. Then, we calculated
a new parameter F(m) by applying the root mean square to
T(n) =
r(n) r(n+1) sin3◦
(n = 1 … 120) SS(j). F(m) normally increases with increasing m because it
2 ∑120 follows the power law.
OPA = n=1 T(n)
√
Index of Fluctuation in the Center of Pressure 1 ∑C 2
F(m) = [SS(j) ] (j = 1, 2, 3, … , C)
C j=1

To assess the COP fluctuations, we calculated the standard Finally, we calculated the slope of the regression line in
deviation (SD) of the COP coordinate positions per 100 Hz a diffusion plot with ­log10(m) on the horizontal axis and
in each ML and AP direction, as well as the alpha index of ­log10F(m) on the vertical axis using a least-square method
the strength of the long-term correlations in each direction. (Fig. 1). We defined it as the alpha index (Hausdorff et al.,
We also evaluated the mean value, SD, alpha index, and 1995; Peng et al., 1995).
coefficient of variation (CV) of the distance moved by the Coefficient of Variation: Ratio of Variation to Mean
COP per 100 Hz in each direction. The calculation of the Value
alpha index and CV is shown below, using the x-axis as an The CV was calculated using the following formulas. CV
example. is an abstract number unaffected by the measurement unit
Fractal Index Alpha: the Strength of Long-Term because the SD and mean have the same units.
Correlations
∑N
We calculated the fractal index alpha using detrended Meanvalue = N1 i=1 x(i)
fluctuation analysis, which indicates the strength of long- �
∑N 2
term correlations in time-series data using the root mean SD = N1 i=1 (x(i) − x)
SD
square (Hausdorff et al., 1995; Peng et al., 1995). First, we CV = Mean
created the new time-series data, S(k), each of which was the

Fig. 1  Log–log diffusion plots

obtained from DFA on the COP.
This is a representative example
of the log–log diffusion plots of
participants in the mediolateral
direction, collected during the
standing position with eyes
open and eyes closed

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Advances in Neurodevelopmental Disorders

Data Analyses

WAIS-III, Wechsler Adult Intelligence Scale-Third Edition; ASD, autism spectrum disorder; TD, typical development; FSIQ, full-scale intelligent quotient (IQ); VIQ, verbal IQ; PIQ, perfor-
First, we applied Shapiro–Wilk tests to assess the nor-
mality of the data. We used the unpaired t test for inter-

Effect size
group comparisons of all items of the WAIS-III and the

0.23
0.15
0.69
domains of restricted interests/behaviors and fine motor
function of the MSPA with normal distributions (p > 0.05)
using the Shapiro–Wilk test. The Mann–Whitney U test
was used for all items of the ADOS-2, and the remaining

TD vs ASD
domains of the MSPA showed non-normal distributions

P-value
(p < 0.05). We examine the differences between the groups

0.54
0.69
0.08
in the WAIS-III, ADOS-2, and MSPA. All significance
levels were set at p < 0.05. Next, we conducted a 2 × 2 × 2
repeated-measures analysis of variance (ANOVA) on
the LL, the SD and alpha index of the coordinate posi-
tions, and the mean value, SD, CV, and alpha index of the

93−129
83−130
90−128
Range
velocity based on the factors of the group, condition, and
direction. We also performed a 2 × 2 repeated-measures
ANOVA on TLL and OPA with the factors of group and
condition. We conducted Bonferroni’s tests for ANOVA
results, showing interactions.
Data processing was performed using MATLAB R2017b
(The MathWorks, Japan), and statistical analysis was per-

11.8
12.4
13.5
SD
formed using IBM SPSS Statistics version 26.0 (IBM,
Japan).

TD group (n=13)
Results Mean

112.3
111.2
111.3
The results for the WAIS-III without group differences are
summarized in Table 1. The ADOS-2 and MSPA scores are
shown in Table 2, with group-differences across all domains.

Locus Length, Total Locus Length, and Outer

81−124
81−140
84−123
Range

Peripheral Area of the COP

-First, we show the analysis results for LL in Fig.  2a

Table 1  Group comparison of intellectual level by WAIS-III

and Table 3. We observed the main effects for condition

and direction (p < 0.001 for each) but not for the group
(p = 0.155). In addition, we found an interaction for condi-
10.6
14.3
10.9

tion × direction (p = 0.001), and the following Bonferroni’s

SD

tests showed that the LL was significantly longer in the AP

ASD group (n=16)

direction than in the ML direction under both conditions

and significantly longer in the EC condition than in the EO
mance IQ; SD, standard deviation

condition in both directions (p < 0.001 for each). No other

main effects or interactions were found.
Mean

109.8
113.2
102.9

Next, we calculated TLL and OPA for each condition

(Fig. 2b,c and Table 3). The TLL and OPA in the EC condi-
tion were significantly longer or larger than those in the EO
condition (TLL: p < 0.001; OPA: p = 0.003). However, no
significant differences were observed between the groups
(TLL: p = 0.149; OPA: p = 0.206) or interactions (TLL:
Items

FSIQ
VIQ
PIQ

p = 0.325; OPA: p = 0.613).

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Advances in Neurodevelopmental Disorders

Table 2  Group comparison of autistic traits by ADOS-2 and MSPA

Evaluation index Items ASD group (n=16) TD group (n=13) TD vs ASD Effect size
P-value
Mean SD Mean SD

ADOS-2 Communication 2.3 1.8 0.6 0.7 0.02 * 0.44

Mutual interpersonal relationship 4.8 3.6 1.2 1.1 0.00 ** 0.64
Communication and interpersonal 7.1 5.0 1.9 1.6 0.00 ** 0.61
relationship
MSPA Communication 3.5 0.6 1.5 0.4 0.00 ** 0.76
Group adaptability 3.6 0.5 1.4 0.4 0.00 ** 0.86
Empathy 3.3 0.5 1.3 0.3 0.00 ** 0.86
Restricted interests/behaviors 3.6 0.6 1.7 0.4 0.00 ** 3.11
Sensory 2.6 0.7 1.6 0.5 0.00 ** 0.61
Stereotyped/repetitive motion 1.6 0.7 1.0 0.0 0.00 ** 0.58
Gross motor 2.5 0.9 1.5 0.4 0.00 ** 0.53
Fine motor 2.0 0.8 1.5 0.5 0.04 * 0.82
Inattention 3.6 0.6 2.1 0.6 0.00 ** 0.74
Hyperactivity 2.1 0.9 1.3 0.3 0.01 ** b 0.48
Impulsivity 2.7 0.9 1.4 0.4 0.00 ** 0.64
Sleep cycle 2.6 0.6 1.5 0.6 0.00 ** 0.68
Learning 1.7 0.7 1.2 0.4 0.03 * 0.40
Language development a 1.8 0.8 1.2 0.6 0.03 * 0.41

ADOS-2 Autism Diagnostic Observation Schedule-Second Edition; MSPA Multi-dimensional Scale for PDD and ADHD; ASD autism spectrum
disorder; TD typical development; SD standard deviation
a
 There were fewer answers in language development (ASD group = 15, TD group = 12) because neither the participants nor their parents
recalled language development in infancy
b
 p = 0.009
*p < 0.05; **p < 0.01

Standard Deviation and Alpha Indexes
of the Coordinate Positions
The SDs of the time-series data for the coordinate positions
are shown in Fig. 3a and Table 3. We observed a main effect
of the direction (p = 0.026). We also observed an interac-
tion between condition × direction (p = 0.002). The results
of Bonferroni’s tests showed that the SD of the coordinate
position in the AP direction was significantly larger than that
in the ML direction under the EO condition (EO: p = 0.001;
EC: p = 0.757) and that the SD under the EC condition was
significantly larger than that in the EO condition in the ML
direction (ML: p = 0.001; AP: p = 0.163). No other main
effects or interactions were found.
The results of the alpha index in the time-series data of
the coordinate positions are presented in Fig. 3b and Table 3.
We observed main effects of condition and direction (condi-
tion: p < 0.001; direction: p = 0.002). We also observed an
interaction between condition × direction (p < 0.001). The
results of the subsequent simple main effect analysis showed
that the alpha index of the coordinate position was signifi-
cantly larger in the AP direction than in the ML direction
under the EO condition (EO: p < 0.001; EC: p = 0.870) and
significantly larger in the EO condition than in the EC condi-
tion in the AP direction (ML: p = 0.053; AP: p < 0.001). No
other main effects or interactions were found.
Mean Values, Standard Deviations, Coefficients
of Variation, and Alpha Indexes of Velocity
The test results for the mean velocity are shown in Fig. 4a
and Table 3.The main effects of the condition (EC > EO)
and direction (AP > ML) (p < 0.001 for each) were observed.
We also observed an interaction between condition × direc-
tion (p = 0.001). Bonferroni’s tests showed that the mean
velocity was significantly longer in the AP direction than in
the ML direction under both conditions (p < 0.001 for each)
and significantly higher in the EC condition than in the EO
condition in both directions (p < 0.001 for each). No other
main effects or interactions were found.
Next, we calculated the SD of the velocity (Fig.  4b
and Table 3). The main effects for condition and direction
(p < 0.001 for each) and interaction for condition × direc-
tion (p < 0.001) were observed. Subsequent Bonferroni’s
tests showed that the SD of velocity was significantly larger
in the AP direction than in the ML direction under both

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.


Fig. 2  Locus length a, total locus length b, and outer peripheral area
c. A significant interaction was observed in the condition × direction
only for the locus length a. No significant differences were observed
between the groups. **p < 0.01. ASD, autism spectrum disorder;
conditions (EO, p < 0.001; EC, p = 0.037) and significantly
larger in the EC condition than in the EO condition in both
directions (ML: p < 0.001; AP: p = 0.001). No other main
effects or interactions were found.
The CV of velocity is presented in Fig. 4c and Table 3.
There were main effects of the condition and direction (con-
dition: p = 0.037; direction: p < 0.001). Although the CV of
the ASD group was larger than that of the TD group, the
difference was insignificant (p = 0.066). An interaction was
observed between condition × direction (p = 0.042). The
results of Bonferroni’s tests showed that the CV of the veloc-
ity was significantly larger in the ML direction than in the
AP direction under both conditions (p < 0.001 for each) and
significantly larger in the EC condition than in the EO condi-
tion in the ML direction (ML: p = 0.014; AP: p = 0.375). No
other main effects or interactions were found.
Finally, the alpha indices of the velocity are shown
in Fig. 4d and Table 3. There was a main effect of the
direction (p < 0.001). The alpha index did not differ
between the groups (p = 0.061). We observed an
interaction only for direction × group (p = 0.011). Simple
main effect analyses showed a significantly larger alpha
index in the ML direction than in the AP direction only
in the TD group (TD group: p < 0.001; ASD group:
p = 0.231). The alpha index was also significantly larger
in the ASD group than in the TD group in the AP direction
(ML: p = 0.569; AP: p = 0.013). No other main effects or
interactions were found.f
Discussion
The results of the present study showed no significant
differences in LL, TLL, or OPA (indicating the magnitude
of sway) between the ASD and TD groups. Molloy et al.
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Advances in Neurodevelopmental Disorders
TD, typical development; EO, eyes-open condition; EC, eyes-closed
condition; ML, mediolateral direction; AP, anteroposterior direction;
Error bar, standard deviation
(2003) reported a significantly larger OPA in children with
ASD than in TD children. In addition, in a study of college
students, Mache et al. (2021) revealed that the ASD group
swayed more than the TD group in both eyes open and
closed conditions. In contrast, Doumas et al. (2016) found
no difference in postural instability between the ASD and
TD groups in the postural instability of adults. Minshew
et  al. (2004) reported that postural instability in ASD
improves at approximately 12 years of age. Considering
this previous research, the lack of significant differences
between the ASD and TD groups in our study might be
because that the participants were adults. Our study, in
which the average age of participants was over 30 years,
supports the study by Minshew et al. (2004) that postural
stability in ASD is equivalent to TD by that time. Although
the differences were not significant, LL, TLL, and OPA in
the EC condition in the ASD group were larger than those
in the TD group (Fig. 1 and Table 3). Some weaknesses
present in childhood may have persisted as a characteristic
of individuals with ASD. We also showed larger COP, LL,
and OPA in the AP direction than in the ML direction and in
the EC condition than in the EO condition, regardless of the
presence of ASD. These results are consistent with previous
findings in individuals without ASD (Stins et al., 2015). This
could be due to the larger range of motion in the skeleton of the
human lower leg in the AP direction compared to that in the
ML direction and the blocking of visual information by closing
the eyes. Individuals with ASD might have further weakness
in postural control, where even individuals without ASD have
difficulties. This is especially true when the eyes are closed, as
individuals with ASD have a visual advantage and an effective
visual field compared to TD children (Dakin & Frith, 2005).
In addition to the magnitude of the overall sway
described above, such as COP, LL, and OPA, we also
explored sway stability by analyzing the coordinate
Advances in Neurodevelopmental Disorders
position and velocity dispersion. We observed that
the standard deviation and strength of the long-term
correlation in the COP coordinate position, velocity,
and its standard deviations, and the degree of variation
correlation did not differ between the ASD and TD
groups. These results suggest similar functions of
somatosensory and vestibular sensations to acquire
information on postural position and strategies for
postural control in both groups.
In our results, the difference was not statistically signifi-
cant in the CV. However, our results showed a greater vari-
ation in the COP in the ML direction for ASD than for TD.
This is because the stable postural control might be more
difficult and unstable in the ML direction than in the AP
direction because the muscle groups in the ML direction are
thinner and smaller than those in the AP direction.
Children with ASD show abnormalities or less adept
ability in the use of hip strategy (Chen & Tsai, 2015) and
COP with fast movement velocity (Memari et al., 2013 and
Lidstone et al., 2020), as well as an inability to constrain
weight-shifting (Fournier et al., 2010). Given the lack of
significance of the group differences in our adult data, sway
instability in ASD may improve and become less significant
with growth, similar to the magnitude of sway described
above. Regarding the brain functions that control human
movement, the cerebellums of individuals with ASD and
ADHD differ in both structure and function compared to
those of individuals with TD (Nayate et al., 2005). The cer-
ebellum controls anticipatory postural adjustments so that
posture does not collapse before motion, which is part of
a feedforward control system and plays an important role
in the postural control system (Pisotta & Molinari, 2014;
Seidler et al., 2004; Mosconi et al., 2015). The nearly sig-
nificant difference in sway variability between the ASD
and TD groups in our study might be due to the number of
participants.
In contrast, the velocity in the ASD group showed a
stronger long-term correlation than in the TD group only in
the AP direction. Thus, individuals with ASD maintained
stable postural sway long-term despite the seemingly unsta-
ble posture described above. To the best of our knowledge,
this is a novel finding of this study. In both groups, the
strengths of the long-term correlations were not affected
by visual sense, while the magnitudes and postural control
variations were larger with closed eyes. Although the lack
of visual cues may seem disadvantageous in ASD, less short-
term interference may strengthen individuals with ASD in
long-term correlations. In addition, Delignières et al. (2011)
revealed that the COP in a static standing position was more
affected by velocity information than by position informa-
tion in the TD individual. Similar results have been obtained
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
in patients with ASD. Furthermore, our results showed a
significant group difference only in the velocity alpha index,
not the coordinate position’s alpha index. This suggests that
velocity control is more stable in ASD individuals than in
TD individuals. We did not consider crossovers, such as Del-
ignières et al. (2011), but ASD may be less susceptible to
time delays in COP stability than TD. There are open-loops
that do not correct the change in posture even when receiv-
ing external stimuli and closed-loops that correct the tilt of
the posture to return to the upright position in motor control
of humans (Collins & De Luca, 1993, 1994; Bottaro et al.,
2008). In our study, the postural control mechanism by the
closed-loop in patients with ASD was weaker than that in
patients with TD, and the generation of active control torque
may have been weak.
Limitations and Future Research
Our study had several limitations. First, in addition to the
small number of participants, the age range of the partici-
pants was wide. A sufficient number of participants in each
age group would have clarified the results. The second factor
is the number of executions for each condition. To eliminate
the effect of practice, measurements were performed in only
one trial for each condition. Multiple measurements on dif-
ferent days are desirable.
Our findings suggest that the posture of individuals
with ASD may be stable from a long-term perspective,
although the posture of individuals with ASD is
unstable at first glance. This result predicts that the
difference in motor impairment between ASD and
TD is attenuated with age, but the instability has not
completely disappeared. In addition, it is conceivable that
the characteristic visual processing ability and control
strategies of the hip joints and lower legs in individuals
with ASD might differ from those with TD. Additional
investigations are needed to determine the mechanism and
factors required to maintain the long-term correlation of
the COP observed in ASD, including musculoskeletal and
brain function assessments. These studies in individuals
with ASD will lead to the maintenance and promotion of
motor function and QOL and well-being.
Acknowledgements  We thank all participants who contributed to this
study.
Author Contribution  NT collaborated with conceptualization, data
curation, methodology, creation of software, data analysis, writing—
original draft and editing, and funding acquisition. SO collaborated
with data collection, investigation, data curation, validation, and writ-
ing—review and editing. NM collaborated with data collection, investi-
gation, and data curation. ZS collaborated with creation of software. TT
collaborated with supervision. YF collaborated with conceptualization,
13


Table 3  Descriptive statistics values of the evaluation indices of COP

Index of the COP ASD group (n=16) TD group (n=13) Main effect Interaction Simple main
effect

13
Mean SD Range Mean SD Range F-value (1, 27) partialη2 F-value (1, 27) F-value (1, 27)

LL EO_ML (mm) 589.64 101.19 436.40−808.90 565.64 82.71 435.80−722.30 condition: 24.32 0.47 13.12 EO: 122.34
direction: 56.34 0.68 EC: 22.07
EO_AP (mm) 768.36 116.53 597.50−923.10 740.30 124.11 553.50−992.20 ML: 24.21
EC_ML (mm) 775.41 210.24 527.20−1311.40 701.79 151.39 450.50−1046.20 AP: 21.75
EC_AP (mm) 900.46 136.11 658.00−1145.60 815.90 75.68 696.40−920.40
TLL EO (mm) 1073.59 156.88 836.40−1296.10 1032.15 155.17 844.00−1358.20 condition: 23.61 0.47 - -
EC (mm) 1320.54 255.10 936.70−1831.90 1194.66 152.53 909.90−1469.30
OPA EO ­(mm2) 129.01 58.96 43.60−295.10 98.36 45.29 19.40−171.50 condition: 10.73 0.28 - -
EC ­(mm2) 188.34 120.27 53.00−486.80 141.65 110.58 20.70−386.10
Coordinate posi- SD EO_ML (mm) 5.21 1.69 2.80−9.60 4.54 1.33 2.30−7.00 direction: 5.55 0.17 12.07 EO: 12.55
tion ML: 15.59
EO_AP (mm) 6.69 2.15 2.90−10.10 6.59 2.71 2.20−10.50
EC_ML (mm) 6.48 2.23 3.40−11.60 5.70 2.01 2.20−8.80
EC_AP (mm) 6.17 1.64 3.90−9.20 5.78 3.16 2.10−12.40
Alpha EO_ML 1.13 0.12 0.96−1.37 1.11 0.15 0.89−1.33 condition: 43.88 0.62 20.82 EO: 34.39
direction: 12.21 AP: 60.51
EO_AP 1.29 0.16 1.02−1.56 1.34 0.13 1.08−1.54
EC_ML 1.05 0.15 0.81−1.35 1.06 0.11 0.92−1.32
EC_AP 1.04 0.13 0.81−1.23 1.08 0.15 0.89−1.47

Velocity Mean EO_ML (mm) 0.12 0.02 0.09−0.16 0.11 0.02 0.09−0.15 condition: 24.54 0.48 14.13 EO: 131.22
direction: 58.9 0.69 EC: 22.03
EO_AP (mm) 0.15 0.02 0.12−0.19 0.15 0.02 0.11−0.20 ML: 25.05
AP: 20.80
EC_ML (mm) 0.16 0.04 0.10−0.26 0.14 0.03 0.09−0.21

EC_AP (mm) 0.18 0.03 0.13−0.23 0.16 0.02 0.14−0.18

SD EO_ML (mm) 0.13 0.03 0.09−0.19 0.12 0.02 0.09−0.15 condition: 20.67 0.43 16.04 EO: 58.10
direction: 22.04 0.45 EC: 4.80
EO_AP (mm) 0.16 0.03 0.12−0.23 0.15 0.02 0.12−0.20 ML: 22.47
AP: 15.55
EC_ML (mm) 0.18 0.05 0.11−0.31 0.15 0.04 0.09−0.25

EC_AP (mm) 0.19 0.03 0.14−0.26 0.17 0.02 0.14−0.19

CV EO_ML 1.08 0.05 1.01−1.17 1.06 0.05 1.03−1.20 condition: 4.79 0.15 4.57 EO: 22.68
direction: 71.42 0.73 EC: 63.88
EO_AP 1.04 0.06 1.00−1.26 1.01 0.01 1.00−1.04 ML: 6.97

EC_ML 1.12 0.06 1.01−1.22 1.09 0.06 1.02−1.21

EC_AP 1.05 0.05 0.97−1.16 1.02 0.03 0.99−1.07

Alpha EO_ML 0.65 0.11 0.44−0.93 0.62 0.13 0.40−0.85 direction: 19.19 0.42 7.52 TD: 22.99 AP:

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

7.04
EO_AP 0.62 0.12 0.44−0.97 0.55 0.09 0.34−0.68

EC_ML 0.67 0.11 0.50−0.90 0.67 0.10 0.49−0.78

EC_AP 0.66 0.14 0.40−0.93 0.56 0.11 0.39−0.69

COP center of pressure; ASD autism spectrum disorder; TD typical development; SD standard deviation; LL locus length; TLL total locus length; OPA outer peripheral area; CV coefficient of
variation; EO eyes-open condition; EC eyes-closed condition. ML mediolateral direction; AP anteroposterior direction
Advances in Neurodevelopmental Disorders
Advances in Neurodevelopmental Disorders

Fig. 3  SD a and alpha index b

in the time-series data of the
coordinate positions. Interac-
tions were observed only for the
condition × direction for both
indices. No significant differ-
ences were observed between
the groups. **p < 0.01. ASD,
autism spectrum disorder;
TD, typical development; SD,
standard deviation; EO, eyes-
open condition; EC, eyes-closed
condition; ML, mediolateral
direction; AP, anteroposterior
direction; pos, coordinate
position; Error bar, standard
deviation

Fig. 4  Mean value a, SD b,
CV c, and alpha index d in the
time-series data of velocity. An
interaction was observed only
for the alpha index for direc-
tion × group d. The mean value,
standard deviation (SD), and
coefficient variation (CV) of the
velocity showed interactions
in the condition × direction.
*p < 0.05; **p < 0.01. ASD,
autism spectrum disorder; TD,
typical development; SD, stand-
ard deviation; CV, coefficient of
variation; EO, eyes-open condi-
tion; EC, eyes-closed condition;
ML, mediolateral direction; AP,
anteroposterior direction; dis,
distance moved per 100 Hz;
Error bar, standard deviation

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.


resources, funding acquisition, supervision, project administration, and
writing—review and editing. All authors have read and approved the
final manuscript.
Funding  This research was supported by Grants-in-Aid for Scientific
Research (A) (17H00883) and on Innovative areas (24119004), Grant-
in-Aid for Challenging Exploratory Research (20K20649), and Grant-
in-Aid for JSPS Research Fellow (18J14536).
Data Availability  The data presented in this article is part of a larger
study and thus not available as Supplementary Material at the present
time.
Declarations  1016/j.​physa.​2013.​09.​037
Ethics Approval  We conducted the present study with the approval
of the Institutional Ethics Committee of our university hospital, and
also in accordance with the Declaration of Helsinki, and the Ethical
Guidelines for Medical and Health Research Involving Human Subjects
by the Japanese Ministry of Health, Labour, and Welfare.
Informed Consent  All participants read the explanatory material about
this study. Then, participants cooperated with the questionnaire and the
experiment only if they consented to the study.
Conflict of Interest  The authors declare no competing interests.
References
American Psychiatric Association [APA]. (2013). Diagnostic and sta-
tistical manual on mental disorders-fourth edition-text revision
(DSM-IV-TR). American Psychiatric Press.
American Psychiatric Association (2022) https://​www.​psych​iatry.​org/​
patie​nts-​famil​ies/​autis​m#:​~:​text=​Autism%​20Spe​ctrum%​20Dis​
order%​20%​28ASD%​29%​20is%​20a%​20com ​plex%​20dev​elopm​
ental​,of%​20the​se%​20cha​lleng​es%​20var​ies%​20bet​ween%​20ind​
ividu​als%​20with%​20aut​ism
Asai, M., Ueda, N., & Shojaku, H. (2018). Development of application
software for analyzing signals from stabilometry using Microsoft
Excel. Equilibrium Research, 77(2), 88–98. https://​doi.​org/​10.​
3757/​jser.​77.​88
Baranek, G. T., David, F. J., Poe, M. D., Stone, W. L., & Watson,
L. R. (2006). Sensory experiences questionnaire: Discriminating
sensory features in young children with autism, developmental
delays, and typical development. Journal of Child Psychology
and Psychiatry, 47(6), 591–601. https://​doi.​org/​10.​1111/j.​1469-​
7610.​2005.​01546.x
Bottaro, A., Yasutake, Y., Nomura, T., Casadio, M., & Morasso, P.
(2008). Bounded stability of the quiet standing posture: An inter-
mittent control model. Human Movement Science, 27(3), 473–
495. https://​doi.​org/​10.​1016/j.​humov.​2007.​11.​005
Clark, R. A., Bryant, A. L., Pua, Y., McCrory, P., Bennell, K., & Hunt,
M. (2010). Validity and reliability of the nintendo wii balance
board for assessment of standing balance. Gait & Posture, 31,
307–310. https://​doi.​org/​10.​1016/j.​gaitp​ost.​2009.​11.​012
Chen, F., & Tsai, C. (2015). A light fingertip touch reduces postural
sway in children with autism spectrum disorders. Gait & Posture,
43(1), 137–140. https://​doi.​org/​10.​1016/j.​gaitp​ost.​2015.​09.​012
Collins, J. J., & De Luca, C. J. (1993). Open-loop and closed-loop con-
trol of posture: A random-walk analysis of center-of-pressure tra-
jectories. Experimental Brain Research, 95(2), 308–318. https://​
doi.​org/​10.​1007/​BF002​29788
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Advances in Neurodevelopmental Disorders
Collins, J. J., & De Luca, C. J. (1994). Random walking during quiet
standing. Physical Review Letters, 73(5), 764–767. https://d​ oi.o​ rg/​
10.​1103/​PhysR​evLett.​73.​764
Dakin, S., & Frith, U. (2005). Vagaries of visual perception in autism.
Neuron, 48, 497–507. https://​doi.​org/​10.​1016/j.​neuron.​2005.​10.​018
Delignières, D., Torre, K., & Bernard, P. L. (2011). Transition from
persistent to anti-persistent correlations in postural sway indi-
cates velocity-based control. Plos Computational Biology, 7(2),
e1001089. https://​doi.​org/​10.​1371/​journ​al.​pcbi.​10010​89
Delignières, D., & Marmelat, V. (2014). Strong anticipation and long-
range cross-correlation: Application of detrended cross-corre-
lation analysis to human behavioral data. Physica a: Statistical
Mechanics and Its Applications, 394, 47–60. https://​doi.​org/​10.​
Doumas, M., McKenna, R., & Murphy, B. (2016). Postural control defi-
cits in autism spectrum disorder: The role of sensory integration.
Journal of Autism and Developmental Disorders, 46, 853–861.
https://​doi.​org/​10.​1007/​s10803-​015-​2621-4
Duarte, M., & Sternad, D. (2008). Complexity of human postural con-
trol in young and older adults during prolonged standing. Experi-
mental Brain Research, 191, 265–276. https://​doi.​org/​10.​1007/​
s00221-​008-​1521-7
Dziuk, M. A., Larson, J. C. G., Apostu, A., Mahone, E. M., Denckla,
M. B., & Mostofsky, S. H. (2007). Dyspraxia in autism: Asso-
ciation with motor, social, and communicative deficits. Devel-
opmental Medicine & Child Neurology, 49, 734–739. https://​
doi.​org/​10.​1111/j.​1469-​8749.​2007.​00734.x
Fournier, K. A., Kimberg, C. I., Radonovich, K. J., Tillman, M.
D., Chowd, J. W., Lewis, M. H., Bodfish, J. W., & Hass, C. J.
(2010). Decreased static and dynamic postural control in chil-
dren with autism spectrum disorders. Gait & Posture, 32(1),
6–9. https://​doi.​org/​10.​1016/j.​gaitp​ost.​2010.​02.​007
Funabiki, Y., Kawagishi, H., Uwatoko, T., Yoshimura, S., & Murai,
T. (2011). Development of a multi-dimensional scale for PDD
and ADHD. Research in Developmental Disabilities, 32(3),
995–1003. https://​doi.​org/​10.​1016/j.​r idd.​2011.​01.​052
Funabiki, Y., & Shiwa, T. (2018). Weakness of visual working mem-
ory in autism. Autism Research, 11, 1245–1252. https://​doi.​org/​
10.​1002/​aur.​1981
Hausdorff, J. M., Peng, C. K., Ladin, Z., Wei, J. Y., & Goldberger,
A. L. (1995). Is walking a random walk? evidence for long-
range correlations in stride interval of human gait. Journal of
Applied Physiology, 78, 349–358. https://​doi.​org/​10.​1152/​jappl.​
1995.​78.1.​349
Imaoka, K., Murase, H., & Fukuhara, M. (1997). Collection of data
for healthy subjects in stabilometry. Equilibrium research Supple-
ment, 12, 1–84. https://​doi.​org/​10.​3757/​jser.​56.​12Sup​pleme​nt_1
Kroeger, K. A., Schultz, J. R., & Newsom, C. (2007). Comparison
of two group-delivered social skills programs for young children
with autism. Journal of Autism and Developmental Disorders, 37,
808–817. https://​doi.​org/​10.​1007/​s10803-​006-​0207-x
Lidstone, D. E., Miahd, F. Z., Postonc, B., Beasleyd, J. F., & Dufekc,
J. S. (2020). Examining the specificity of postural control deficits
in children with autism spectrum disorder using a cross-syndrome
approach. Research in Autism Spectrum Disorders, 72, 101514.
https://​doi.​org/​10.​1016/j.​rasd.​2020.​101514
Lord, C., Rutter, M., DiLavore, P. C., Risi, S., Gotham, K., & Bishop,
S. (2012). (ADOS™-2) Autism diagnostic observation schedule™
second edition manual, Western psychological services.
Mache, M., Todd, T., Jarvis, D., & Geary, K. (2021). Comparison of
postural control among college students with and without autism
spectrum disorder. Advances in Neurodevelopmental Disorders,
5(6). https://​doi.​org/​10.​1007/​s41252-​021-​00204-y
Memari, A. H., Ghanouni, P., Gharibzadeh, S., Eghlidi, J., Ziaee, V.,
& Moshayedi, P. (2013). Postural sway patterns in children with
autism spectrum disorder compared with typically developing
Advances in Neurodevelopmental Disorders
children. Research in Autism Spectrum Disorders, 7, 325–332.
https://​doi.​org/​10.​1016/j.​rasd.​2012.​09.​010
Minamisawa, T., Takakura, K., & Yamaguchi, T. (2009). Detrended
fluctuation analysis of temporal variation of the center of pressure
(COP) during quiet standing in Parkinsonian patients. Journal
of Physical Therapy Science, 21(3), 287–292. https://​doi.​org/​10.​
1589/​jpts.​21.​287
Minshew, N. J., Sung, K. B., Jones, B. L., & Furman, J. M. (2004).
Underdevelopment of the postural control system in autism. Neu-
rology, 63, 2056–2061. https://​doi.​org/​10.​1212/​01.​WNL.​00001​
45771.​98657.​62
Molloy, C. A., Dietrich, K. N., & Bhattacharya, A. (2003). Postural
stability in children with autism spectrum disorder. Journal of
Autism and Developmental Disorders, 33(6), 643–652. https://​
doi.​org/​10.​1023/B:​JADD.​00000​06001.​00667.​4c
Mosconi, M. W., Mohanty, S., Greene, R. K., Cook, E. H., Vaillan-
court, D. E., & Sweeney, J. A. (2015). Feedforward and feedback
motor control abnormalities implicate cerebellar dysfunctions in
autism spectrum disorder. The Journal of Neuroscience, 35(5),
2015–2025. https://​doi.​org/​10.​1523/​JNEUR​OSCI.​2731-​14.​2015
Nayate, A., Bradshaw, J. L., & Rinehart, N. J. (2005). Autism and
Asperger’s disorder: Are they movement disorders involving the
cerebellum and/or basal ganglia? Brain Research Bulletin, 67,
327–334. https://​doi.​org/​10.​1016/j.​brain​resbu​ll.​2005.​07.​011
Ornitz, E. M., Guthrie, D., & Farley, A. H. (1977). The early develop-
ment of autistic children. Journal of Autism and Developmental
Disorders, 7, 207–229. https://​doi.​org/​10.​1007/​BF015​38999
Peng, C. K., Havlin, S., Stanley, H. E., & Goldberger, A. L. (1995).
Quantification of scaling exponents and crossover phenomena in
nonstationary heartbeat time series. Chaos, 5, 82–87. https://​doi.​
org/​10.​1063/1.​166141
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Pisotta, I., & Molinari, M. (2014). Cerebellar contribution to feedfor-
ward control of locomotion. Frontiers in Human Neuroscience,
475(8). https://​doi.​org/​10.​3389/​fnhum.​2014.​00475
Seidler, R. D., Noll, D. C., & Thiers, G. (2004). Feedforward and feed-
back processes in motor control. NeuroImage, 22(4), 1775–1783.
https://​doi.​org/​10.​1016/j.​neuro​image.​2004.​05.​003
Stins, J. F., Emck, C., Vries, E. M., Doop, S., & Beek, P. J. (2015).
Attentional and sensory contributions to postural sway in chil-
dren with autism spectrum disorder. Gait & Posture, 42, 199–203.
https://​doi.​org/​10.​1016/j.​gaitp​ost.​2015.​05.​010
Ushiyama, J., Masan, K., Murama, T., Okuyama, S., Murayama, M.,
& Sasaki, R. (2008). Evaluations of postural stability in Keio uni-
versity students using various measures of static posturography.
Bulletin of the Institute of Physical Education, Keio university,
47(1), 13–24. https://​koara.​lib.​keio.​ac.​jp/​xooni​ps/​modul​es/​xooni​
ps/​detail.​php?​koara_​id=​AN001​35710-​00470​001-​0013.
Wechsler, D. (1997). Manual for the Wechsler Adult Intelligence Scale-
III. New York: Psychological Corporation (Fujita, K., Maekawa,
H., Dairoku, K., & Yamanaka, K. (2006). Japanese Wechsler
Adult Intelligence Scale- III. Tokyo, Nihon Bunka Kagakusha.).
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds
exclusive rights to this article under a publishing agreement with the
author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of
such publishing agreement and applicable law.
13
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

onlineservice@springernature.com