Research in Developmental Disabilities 31 (2010) 1536–1542

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Visual tasks and postural sway in children with and without autism
spectrum disorders
Chih-Hui Chang a, Michael G. Wade b, Thomas A. Stoffregen b, Chin-Yu Hsu a, Chien-Yu Pan a,*
a
Department of Physical Education, National Kaohsiung Normal University, Taiwan
b
School of Kinesiology, University of Minnesota, USA

A R T I C L E I N F O A B S T R A C T

Article history: We investigated the influences of two different suprapostural visual tasks, visual searching
Received 19 May 2010 and visual inspection, on the postural sway of children with and without autism spectrum
Accepted 5 June 2010 disorder (ASD). Sixteen ASD children (age = 8.75  1.34 years; height = 130.34  11.03 cm)
were recruited from a local support group. Individuals with an intellectual disability as a co-
Keywords: occurring condition and those with severe behavior problems that required formal
Postural control
intervention were excluded. Twenty-two sex- and age-matched typically developing (TD)
Autism spectrum disorder (ASD)
children (age = 8.93  1.39 years; height = 133.47  8.21 cm) were recruited from a local
Visual performance
Children public elementary school. Postural sway was recorded using a magnetic tracking system
(Flock of Birds, Ascension Technologies, Inc., Burlington, VT). Results indicated that the ASD
children exhibited greater sway than the TD children. Despite this difference, both TD and ASD
children showed reduced sway during the search task, relative to sway during the inspection
task. These findings replicate those of Stoffregen et al. (2000), Stoffregen, Giveans, et al. (2009),
Stoffregen, Villard, et al. (2009) and Prado et al. (2007) and extend them to TD children as well
as ASD children. Both TD and ASD children were able to functionally modulate postural sway
to facilitate the performance of a task that required higher perceptual effort.
ß 2010 Elsevier Ltd. All rights reserved.

Posture control is a fundamental requirement for behavior success. Many studies have examined relations between the
control of standing posture and the simultaneous performance of non-postural tasks, such as listening to speech, or reading
(for a review, see Woollacott & Shumway-Cook, 2002). However, the role the postural control system plays in accomplishing
certain goals is still controversial (Riccio & Stoffregen, 1988; Woollacott & Shumway-Cook, 2002). Several studies have found
that the control of stance is not an end itself, but is influenced by the goals of behavior (e.g., Stoffregen, Hove, Bardy, Riley, &
Bonnet, 2007; Stoffregen, Pagulayan, Bardy, & Hettinger, 2000; Stoffregen, Smart, Bardy, & Pagulayan, 1999). Riccio and
Stoffregen argued that the relationship between postural control and optical flow is not autonomous but task-dependent.
The essential features of children with autism spectrum disorder (ASD) are qualitative impairments in social interaction
and communication, as well as restricted repetitive and stereotyped patterns of behavior, interests, and activities. Motor
impairments have been categorized as ‘‘associated symptoms’’ of individuals with ASD (Ming, Brimacomb, & Wagner, 2007).
Abnormalities of posture, such as walking on tiptoe, and odd body postures, may be present in ASD children (DSM-IV-TR,
American Psychiatric Association, 2000). In the present study, we examined relations between the performance of visual
tasks and the control of standing posture in children with and without ASD.

* Corresponding author at: Department of Physical Education, National Kaohsiung Normal University, P.O. Box 30-24, 80299 Kaohsiung City, Taiwan.
Tel.: +886 7 717 2930x3531; fax: +886 7 711 4633.
E-mail address: chpan@nknucc.nknu.edu.tw (C.-Y. Pan).

0891-4222/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ridd.2010.06.003
 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542 1537

1. Postural control and suprapostural activity

Suprapostural tasks are tasks or behavioral goals that are superordinate to the control of posture, and for which
performance metrics differ from those related to the control of stance (Stoffregen et al., 1999). For example, reading is
evaluated in terms of comprehension or reading rate; we do not assess the success of reading in terms of the position or
motion of the body’s center of mass. Several studies have yielded results consistent with the hypothesis that control of stance
can be modulated in ways that tend to facilitate the performance of suprapostural visual tasks. Researchers have varied the
distance of visual targets (Stoffregen et al., 1999), the degree of perceptual versus cognitive demand of visual tasks
(Stoffregen et al., 2007), and the motion of visual targets (Stoffregen, Bardy, Bonnet, & Pagulayan, 2006). In each case,
postural control has been influenced by the degree of visual stabilization required in performance of suprapostural visual
tasks. Such effects persist even when standing on a ship at sea, where postural control must continuously be adjusted to
accommodate unpredictable motion of the surface of support (Yu et al., 2010).
Stoffregen et al. (2000) evaluated the sway of healthy young adults during performance of two visual tasks. In the search
task (ST) participants searched a block of text for designated target letters. In the inspection task (IT) participants were
required only to maintain their gaze within the borders of a blank target. The positional variability of the body was reduced
during performance of the search task, relative to body motion during performance of the inspection task. Similar effects,
using the search task and the inspection task, have been observed in healthy elderly adults (Prado, Stoffregen, & Duarte,
2007) and in experienced crewmembers on a ship at sea (Stoffregen, Villard, & Yu, 2009).
Overall, the above studies have found that the positional variability of postural activity tends to be reduced during
performance of suprapostural visual tasks that require stabilization of the visual system relative to targets in the
environment. Posture can be modulated to facilitate the performance of suprapostural tasks, and postural control can be
functionally integrated with the simultaneous performance of suprapostural visual tasks. The studies discussed above have
included only adults as participants. There has been little empirical evaluation of the influence of suprapostural visual tasks
on the postural control of children, either typically developing (TD) or ASD.

2. Motor control and postural control of ASD children

ASD children have been found to have deficits in various motor control areas, including hypotonia and motor apraxia
(Ming et al., 2007), overall gross motor development, locomotor and object control skills (Berkeley, Zittel, Ptiney, & Nichols,
2001; Pan, Tsai, & Chu, 2009), manual dexterity, ball skills and balance (Green et al., 2002, 2009), and a general deficit in
manual responses to visual stimuli (Todd, Mills, Wilson, Plumb, & Mon-Williams, 2009).
The postural activity of ASD children differs from that of TD children. Molly, Dietrich, and Bhattacharya (2003) found that
ASD children (n = 8, aged 84–148 months) swayed in a larger area than did TD children, either with the eyes open or closed.
Kohen-Raz, Volkmar, and Cohen (1992) found that an ASD group (n = 91, aged 6–20 years) showed greater variability in the
overall postural performance than the control group (n = 166, aged 4–11 years). Minshew, Sung, Jones, and Furman (2004)
found reduced postural stability in high-functioning autistic individuals (n = 79, aged 5–52 years) and a delay in the
development of postural stability in this population. Weimer, Schatz, Lincoln, Ballantyne, and Trauner (2001) measured the
duration of stance on one-leg with the eyes closed. They found that TD participants were able to maintain this stance for a
longer period than participants with Asperger’s Syndrome, one subtype of ASD (n = 10.9–19.9 years).
The above studies showed that ASD children were more posturally unstable than their TD counterparts. However, none of
these studies included within-participants variations in the nature or difficulty of visual tasks during stance.

3. Suprapostural tasks and postural control in ASD children

We were interested in the influences of different behavior goals on postural control in two different groups of individuals,
the ASD and TD children. There has been no research about possible influences of suprapostural visual tasks on postural
control of ASD children. In general, there has been little research regarding characteristics of postural control in ASD children.
In the absence of existing empirical research on ASD children, we can consider effects of suprapostural visual tasks
on the postural control of other clinical and quasi-clinical populations. Prado et al. (2007) investigated the influences
of suprapostural visual tasks on the standing posture of healthy elderly adults. Using the search task and the inspection
task from Stoffregen et al. (2000), they evaluated postural control in young adults and healthy elderly adults.
Overall sway magnitude was greater among elderly adults than among the younger adults. Despite this overall
difference, the healthy elderly adults reduced their sway during performance of the search task, relative to sway during
performance of the inspection tasks. The magnitude of the task-specific reduction in sway was equivalent for the two
age groups.
We examined the influence of suprapostural visual tasks on the standing posture of TD and ASD children. Following
Stoffregen et al. (2000), we contrasted a visual search task with a task that required inspection of a featureless target. We
assumed that visual search would require more precise control of the visual system (i.e., fixation and eye movements) than
inspection of a blank target. For this reason, we predicted that postural sway would be reduced during performance of the
search task relative to sway during performance of the inspection task. Following previous research, we also expected that
ASD children would tend to sway more overall than TD children.
1538 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542

4. Method

4.1. Participants

Sixteen ASD children (all male, age = 8.75  1.34 years; height = 130.34  11.03 cm; weight = 29.04  8.06 kg) were recruited
from a local support group. Individuals with an intellectual disability as a co-occurring condition and those with severe behavior
problems that required formal intervention were excluded. Diagnoses included mild autistic disorder (n = 8) and Asperger’s
syndrome (n = 8). Twenty-two sex and age-matched TD children (all male, age = 8.93  1.39 years; height = 133.47  8.21 cm;
weight = 32.20  10.72 kg) were recruited from a local public elementary school. All children had normal to corrected normal
vision, and parents reported no history of disease or malfunction of the vestibular apparatus, recurrent dizziness, or falls. Consent
was obtained from the children’s guardians prior to the study. The experimental procedure was approved by the Institute of
Review Board of National Kaohsiung Normal University.

4.2. Apparatus and materials

We followed the basic method of Stoffregen et al. (2000). Two suprapostural tasks, a search task (ST) and an inspection
task (IT) with different perceptual difficulties were manipulated. Targets consisted of a sheet of 13.5 cm  17 cm white paper
and subtended a visual angle of 198  248. For the IT, the white paper was plain, while for the ST, lines of Arabic numerals
were printed on the white paper. The numerals were single-digit and printed using a 12 point Arial Black Bold fond. There
were three different pages for use on different trials, each consisting of 560 (14 lines  40 numerals) Arabic numerals.
We recorded postural activity using a magnetic tracking system (Flock of Birds, Ascension Technologies, Inc., Burlington,
VT). Children were required to stand with their heels on a mark on the floor. They were then 0.4 m from the target and 0.4 m
from the Extended Range Transmitter (ERT) of the Flock of Birds system. One receiver was attached to a bicycle helmet worn
by the children, while a second receiver was attached to the skin between the shoulders (between the children’s shoulder
blades at approximately the level of the seventh cervical vertebra) using double-coated tape. Each receiver was sampled at
50 Hz. Data from the tracking system were stored on disk for later analysis. The experimental set-up is as Fig. 1.

4.3. Procedure

After an experimenter explained the experimental procedure to the children, they were attached to the Flock of Birds
receivers. They were instructed to stand with their feet together, heels on a mark on the floor, and hands in their pockets. The
height of the center of the target was placed at eye-level for each child. They were then required to stand comfortably, face
forward without moving their feet and hands and ‘‘stare intently’’ or search for targeted Arabic numerals on the designated
target.
There were 6 trials for each child, 3 for IT and 3 for ST (2 conditions  3 trials). Each trial lasted for 70 s and the latter 60 s of
each trial were retained for analysis. There were 1 min rests between every three trials. The order of the two conditions was
counterbalanced. For the IT, children were to keep their gaze within the blank white target and to minimize head
[(Fig._1)TD$IG]

Fig. 1. Experimental set-up. ERT = Extended Range Transmitter of the Flock of Birds System.
 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542 1539

movements. For the ST, children were instructed to count the number of instances that a designated numeral occurred
among the numerals printed on the target. On separate trials, three different target numerals were used: 2, 3, and 4.
Participants were instructed that, if they finished a page before the end of a trial, they were to start back at the top of the page,
and continue the total frequency count. At the end of each trial, participants reported where they were in the text when the
trial ended and the number of targets they had detected. The number of target numerals in individual pages ranged from 52
to 62.

4.4. Data analysis

4.4.1. Visual performance

For the IT group, visual performance was not measured. Following previous research (Stoffregen et al., 1999, 2000;
Stoffregen, Villard, et al., 2009), we assumed that the participants maintained their gaze on the target. For the ST group, the
percentage of absolute error was used to represent visual accuracy and was calculated using the following formula, {(jtotal
target number  counted target numberj)/total target number}  100. An independent-samples t-test was performed to
compare the differences between the two groups.

4.4.2. Postural motion

We evaluated the positional variability of the head and torso separately in the anterior–posterior (AP) and mediolateral
(ML) axes. We conducted separate analyses of variance (ANOVA) on movement of head and torso in the AP and ML axes. Each
ANOVA was 2 (Group)  2 (Task), with the last factor as the repeated measure.

5. Results

5.1. Visual performance

For the ST, while counting the number 2, six ASD children and two TD children reported that they forgot either how many
numbers they had counted, or where they were in the text. While counting numbers 3 and 4, five ASD children reported that
they forgot either how many numbers they had counted or where they were in the text, whereas all TD children reported
their counts and final position. An independent-sample t-test on the mean percentage of absolute error of the two groups
was not significant, t(94) = 0.16, p > 0.05. The mean percentage of absolute error of the TD children (M = 17.86%, SD = 18.57%)
did not differ from that of the ASD children (M = 24.14%, SD = 23.40%).

5.2. Postural sway

Data on head movement are summarized in Fig. 2. In the AP axis, the main effect of group was significant, F(1, 36) = 8.48,
p < 0.05, h2 = 0.19, with greater positional variability among the ASD children than among the TD children. The main effect of
task was also significant, F(1, 36) = 21.96, p < 0.05, h2 = 0.38, with positional variability being greater during performance of
the inspection task, relative to sway during performance of the search task. The Task  Group interaction was not significant,
p > 0.05.
[(Fig._2)TD$IG]

Fig. 2. Positional variability of the head in the AP and ML axes. IT = inspection task, ST = search task, TD = typically developing children, and ASD = children
with autism spectrum disorder. Bar indicates standard deviation.
[(Fig._3)TD$IG]
1540 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542

Fig. 3. Positional variability of the torso in the AP and ML axes. IT = inspection task, ST = search task, TD = typically developing children, and ASD = children
with autism spectrum disorder. Bar indicates standard deviation.

For the movement of the head in the ML axis, the main effect of group was significant, F(1, 36) = 22.68, p < 0.05, h2 = 0.39,
with greater positional variability among the ASD children than among the TD children. The main effect of Task and the
Task  Group interaction were not significant, each p > 0.05.
Data on torso movement are summarized in Fig. 3. In the AP axis, the main effect of group was again significant, F(1,
36) = 25.94, p < 0.05, h2 = 0.42, with greater positional variability among the ASD children than among the TD children. The
main effect of task was also significant, F(1, 36) = 16.08, p < 0.05, h2 = 0.31, with positional variability being greater during
performance of the inspection task, relative to sway during performance of the search task. The Task  Group interaction was
not significant, p > 0.05.
Finally, for the movement of the torso in the ML axis, the main effect of group was significant, F(1, 36) = 22.73, p < 0.05,
h2 = 0.39, with greater positional variability among the ASD children than among the TD children. The main effect of Task and
the Task  Group interaction were not significant, each p > 0.05.
Overall, in the AP axes of head and torso, the positional variability of TD children was smaller than that of the ASD
children, and the children swayed with less variability while doing ST than while doing IT. In the ML axes of head and torso,
the positional variability of TD children was smaller than that of the ASD children.

6. Discussion

We evaluated the influence of suprapostural visual tasks on the postural control in children with and without ASD. We
measured postural activity while children searched for designated target numbers and inspected a plain target. We
replicated a common finding that the overall magnitude of postural activity was greater among ASD children than among TD
children. However, we also found that both ASD and TD children reduced the positional variability of their head and torso
activity during performance of the search task, relative to sway during performance of the inspection task. We discuss these
results in turn.
Consistent with previous studies (Kohen-Raz et al., 1992; Minshew et al., 2004; Molloy, Dietrich, & Bhattacharya, 2003;
Weimer et al., 2001), we found that the ASD children tended to sway more than TD children. This effect has been reported in
studies using a wide variety of methods. Molloy et al. evaluated sway in terms of displacements of the center of pressure.
Minshew et al. used dynamic posturography, while Weimar et al. evaluated the time that participants could stand on one-leg
with their eyes closed. In the present study, we directly measured the kinematics of the head and torso. Given the diversity of
methods and the similarity of outcomes, it seems safe to conclude that ASD children are characterized by an increase in the
overall magnitude of sway during upright stance.
Despite differences between TD and ASD children, both groups showed reduced sway variability in the AP direction
during the search task, relative to sway during the inspection task. These findings replicate those of the previous studies with
young adults and elderly adults (Prado et al., 2007; Stoffregen et al., 2000; Stoffregen, Villard, et al., 2009) and extend them to
children with and without ASD.
We found that the organization and execution of postural control were influenced by concurrent suprapostural activity.
The organization of postural sway is acutely sensitive to and robustly influenced by variations in suprapostural tasks that
have no mechanical effects on body motion. The postural control systems of TD and ASD children appeared to have been
governed by the same rules. Both TD and ASD children modulated their postural sway in response to variations in the
suprapostural task.
 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542 1541

There are many differences between children with and without ASD (e.g., DSM-IV-TR, American Psychiatric
Association, 2000). The results of the present study suggest that these groups do not differ in their tendency to modulate
the magnitude of standing body sway in response to variations in suprapostural visual tasks. It may be that ASD does not
impact a child’s ability to integrate the control of stance with the performance of simultaneously suprapostural
activities. Another possibility is that ASD could impact this integration only for certain types of suprapostural activities.
When two healthy adults engage in conversation, the postural activity of their bodies becomes coupled (e.g., Shockley,
Santana, & Fowler, 2003; Stoffregen, Giveans, Villard, Yank, & Shockley, 2009). This coupling does not occur when each
person converses with someone else; thus, the coupling appears to be related to the social interaction. ASD is
characterized by deficits in socialization and social interactions (DSM-IV-TR, American Psychiatric Association, 2000).
Given this, it may be that children with ASD exhibit less coupling of postural activity with conversational partners, or
coupling that is simply different from what has been observed in healthy adults. This possibility will be an interesting
topic for future research.

7. Conclusion

Researchers with adults have found that standing body sway is modulated in response to variations in the perceptual
demand of suprapostural visual tasks (e.g., Stoffregen et al., 2000, 2007; Stoffregen, Giveans, et al., 2009). In the present
study, we found that this effect occurs also in TD children. The existence of this relationship in 8-year-old children raises
questions about the origin and developmental trajectory of functional relations between postural and suprapostural activity.
At any age, visual performance can be negatively impacted by uncontrolled movement of the head or body. Some studies
have suggested that the onset of self-controlled locomotion (e.g., crawling) gives rise to functional changes in visual
attention and in the use of visual information for the control of posture (e.g., Uchiyama et al., 2008). When do children gain
control of relations between postural activity and visual performance?
Prado et al. (2007) confirmed that healthy elderly adults tend to sway more than young adults but that despite this
difference both age groups modulated their sway in response to characteristics of suprapostural visual tasks. In the present
study, we have found essentially the same pattern of results in the contrast between children with and without ASD. The
reduction in sway during performance of more demanding visual tasks is robust even on ships at sea (e.g., Stoffregen,
Giveans, et al., 2009; Yu et al., 2010). Taken together, these findings suggest that functional integration of postural activity
with the performance demands of suprapostural visual tasks may be a general and very robust feature of perception and
action in humans (cf. Woollacott & Shumway-Cook, 2002).

Acknowledgements

Portions of the data were presented at the Fifteenth International Conference on Perception and Action, Minneapolis, MN,
July 2009. We would like to express our gratitude to all of the children and families who participated in this study.

References

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: Text revision (4th TR ed.). Washington, DC: American Psychiatric
Association.
Berkeley, S. L., Zittel, L. L., Pitney, L. V., & Nichols, S. E. (2001). Locomotor and object control skills of children diagnosed with Autism. Adapted Physical Activity
Quarterly, 18, 405–416.
Green, D., Baird, G., Barnett, A. L., Henderson, L., Huber, J., & Henderson, S. E. (2002). The severity and nature of motor impairment in Asperger’s syndrome: a
comparison with Specific Developmental Disorder of Motor Function. Journal of Child Psychology and Psychiatry, 43, 655–668.
Green, D., Charman, T., Pickles, A., Chandler, S., Loucas, T., Simonoff, E., et al. (2009). Impairment in movement skills of children with autistic spectrum disorders.
Developmental Medicine & Child Neurology, 51, 311–316.
Kohen-Raz, R., Volkmar, F. R., & Cohen, D. J. (1992). Postural control in children with autism. Journal of Autism and Developmental Disorders, 22, 419–432.
Ming, X., Brimacombe, M., & Wagner, G. C. (2007). Prevalence of motor impairment in autism spectrum disorders. Brain and Development, 29, 565–570.
Minshew, N. J., Sung, K., Jones, B. L., & Furman, J. M. (2004). Underdevelopment of the postural control system in autism. Neurology, 63, 2056–2061.
Molloy, C. A., Dietrich, K. N., & Bhattacharya, A. (2003). Postural stability in children with autism spectrum disorder. Journal of Autism & Developmental Disorders,
33, 643–652.
Pan, C.-Y., Tsai, C.-L., & Chu, C.-H. (2009). Fundamental movement skills in children diagnosed with autism spectrum disorders and attention deficit hyperactivity
disorder. Journal of Autism and Developmental Disorders, 39, 1694–1705.
Prado, J. M., Stoffregen, T. A., & Duarte, M. (2007). Postural sway during dual tasks in young and elderly adults. Gerontology, 53, 274–281.
Riccio, G. E., & Stoffregen, T. A. (1988). Affordances as constraints on the control of stance. Human Movement Science, 7, 265–300.
Shockley, K., Santana, M.-V., & Fowler, C. A. (2003). Mutual interpersonal postural constraints are involved in cooperative conversation. Journal of Experimental
Psychology: Human Perception and Performance, 29, 326–332.
Stoffregen, T. A., Bardy, B. G., Bonnet, C. T., & Pagulayan, R. J. (2006). Postural stabilization of visually guided eye movements. Ecological Psychology, 18,
191–222.
Stoffregen, T. A., Giveans, M. R., Villard, S., Yank, J. R., & Shockley, K. (2009). Interpersonal postural coordination on rigid and non-rigid surfaces. Motor Control, 13,
471–483.
Stoffregen, T. A., Hove, P., Bardy, B. G., Riley, M., & Bonnet, C. T. (2007). Postural stabilization of perceptual but not cognitive performance. Journal of Motor Behavior,
39, 126–138.
Stoffregen, T. A., Pagulayan, R. J., Bardy, B. G., & Hettinger, L. J. (2000). Modulating postural control to facilitate visual performance. Human Movement Science 19.
Stoffregen, T. A., Smart, L. J., Bardy, B. G., & Pagulayan, R. J. (1999). Postural stabilization of looking. Journal of Experimental Psychology: Human Perception and
Performance, 25, 1641–1658.
Stoffregen, T. A., Villard, S., & Yu, Y. (2009). Body sway at sea for two visual tasks and three stance widths. Aviation, Space, and Environmental Medicine, 80, 1039–
1043.
1542 C.-H. Chang et al. / Research in Developmental Disabilities 31 (2010) 1536–1542

Todd, J., Mills, C., Wilson, A. D., Plumb, M. S., & Mon-Williams, M. A. (2009). Slow motor responses to visual stimuli of low salience in autism. Journal of Motor
Behavior, 41, 419–426.
Uchiyama, I., Anderson, D. I., Campos, J. J., Witherington, D., Frankel, C. B., Lejeune, L., et al. (2008). Locomotor experience affects self and emotion. Developmental
Psychology, 44, 1225–1231.
Weimer, A. K., Schatz, A. M., Lincoln, A., Ballantyne, A. O., & Trauner, D. A. (2001). ‘‘Motor’’ impairment in Asperger syndrome: Evidence for a deficit in
proprioception. Journal of Developmental and Behavioral Pediatrics, 22, 92–101.
Woollacott, M., & Shumway-Cook, A. (2002). Attention and the control of posture and gait: A review of an emerging area of research. Gait and Posture, 16, 1–14.
Yu, Y., Yank, J. R., Katsumata, Y., Villard, S., Kennedy, R. S., & Stoffregen, T. A. (2010). Visual vigilance performance and standing posture at sea. Aviation, Space, and
Environmental Medicine, 81, 375–382.