Social Learning Through Structured Exercise for Students with Autism Spectrum Disorders

Abstract

Increased participation in structured exercise (SE) routines has positive effects on physiological, cognitive, and social development (SD) for students of all ages and abilities. SE offers unique opportunities for social learning in a non-academic context. SE allows students to practice vital social skills such as observation, imitation, and self-regulation. Unfortunately, SD during SE is often overshadowed by more commonly known physiological benefits. Researchers of autism spectrum disorders (ASD) are encouraged by SE findings which offer alternative methods for learning social skills the students struggle to develop. Examining the social impact of increased SE for these students bolsters the value of SE findings beyond physiological effects, illuminating the complex, often overlooked positive relationship between SD and exercise for students with ASD.

Full text

REVIEW PAPER
Social Learning Through Structured Exercise for Students with Autism
Spectrum Disorders
Kian Habib
1,2
&Tina Montreuil
1,3
&Armando Bertone
1,3
Received: 20 June 2017 / Accepted: 25 April 2018
#Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
Increased participation in structured exercise (SE) routines has positive effects on physiological, cognitive, and social develop-
ment (SD) for students of all ages and abilities. SE offers unique opportunities for social learning in a non-academic context. SE
allows students to practice vital social skills such as observation, imitation, and self-regulation. Unfortunately, SD during SE is
often overshadowed by more commonly known physiological benefits. Researchers of autism spectrum disorders (ASD) are
encouraged by SE findings which offer alternative methods for learning social skills the students struggle to develop. Examining
the social impact of increased SE for these students bolsters the value of SE findings beyond physiological effects, illuminating
the complex, often overlooked positive relationship between SD and exercise for students with ASD.
Keywords Autism spectrum disorders .Physical activity .Exercise .Social development .Human development .Adolescents .
Students .Special needs
Introduction
Autism spectrum disorders (ASD) significantly affect an indi-
vidual’s development of social interaction and communication
skills (American Psychiatric Association 2013). Individuals
may display narrow interests, resistance to change, and diffi-
culty following rules. ASD research over the past 20 years has
predominantly focused on how to change these restrictive and
repetitive behaviors without a clear consensus (Lord 2010).
Social development (SD) has become a priority as such symp-
toms negatively impact learning, interfering with communica-
tion and attention during academic activities (Foti et al. 2014;
Lord 2010). Students with ASD have an impaired ability to
attend to tasks and control movement in academic settings.
Current research findings support exercise and physical
education (PE) programs as socially beneficial for all types
of students.
While the physiological benefits of exercise for all individ-
uals have been well established for many years, recently there
has been evidence of the potential for significant SD benefits
for students with ASD and other neurodevelopmental disor-
ders. Researchers, teachers, and caretakers have observed in-
creased behavior regulation, attention, and communication
skills when students with ASD exercise and participate in in-
clusive PE classes (Pan 2010; Richler et al. 2010; Srinivasan
et al. 2014). Increased physical activity (PA) levels, especially
during a structured exercise (SE) routine, have been found to
improve social interaction and increase self-regulation (Foti
et al. 2014). Findings have indicated that students with ASD
improve their ability to control their off-task behaviors in the
classroom directly following participation in SE (Lang et al.
2010). Additionally, attending inclusive PE classes has been
found to enhance SD for students with ASD (Pan et al. 2011).
However, the question remains, why do students with ASD
gain social and communication skills through participation in
SEs when they often struggle to develop these skills in tradi-
tional classroom settings? Are exercises and physical move-
ment alone involved in the stimulation of SD or are there
perhaps other influences at play? Participation in both UE (un-
structured exercise) like recess and SE such as PE class in-
volves physiological exertion as well as social interaction
which make both activities beneficial. Yet students with ASD
*Kian Habib
kian.habib@mail.mcgill.ca
1
Department of Educational and Counselling Psychology, McGill
University, Room 518, 3700 McTavish, Montreal, QC H3A 1Y2,
Canada
2
Perceptual Neuroscience Laboratory (PNLab) for Autism and
Development, Montreal, Canada
3
Faculty of Medicine, Department of Psychiatry, McGill University,
Montreal, Canada
Review Journal of Autism and Developmental Disorders
https://doi.org/10.1007/s40489-018-0139-3

often have fewer opportunities to participate in SE when com-
pared to typically developing students, further limiting their
SD (Hilton et al. 2008; Holifield et al. 2010; Pan and Frey
2006).
Components of social learning are utilized throughout most
PA studies, yet their potential influence on findings is often
inadequately addressed (Hilton et al. 2008; Hsu et al. 2004).
For example, studies will examine academic scores following
exercises or may focus on how to address the low fitness
levels of students. Also, studies rarely use social scales to
measure the impact of SE, opting to record physiological data
such as heart rates. However, examining the influence of so-
cial learning in a context of SE greatly enhances our under-
standing of how and why students at vastly different ends of
the ASD spectrum improve social skills, providing valuable
insight beyond physiological findings.
Defining Social Learning Components
Reliance on social learning theory has been particularly effective
for teaching students with ASD (Bushwick 2001;Langetal.
2010;Lord2010; Rutherford and Rogers 2003). Social learning
theory emphasizes the roles of observation, modeling, retention,
and motivation for the development of behavior throughout an
individual’s lifespan (Bandura 1977;Bushwick2001;Fotietal.
2014). Each of these four components has been found to have a
positive influence on the SD of students with ASD at different
ages and ability levels (Foti et al. 2014; Ingersoll 2008; Pan et al.
2011). Direct observation of human behavior often helps students
understand social norms and appropriate behavior. Essential so-
cial skills, such as understanding facial expressions, eye contact,
and implicit communication, are said to be learned through direct
observation (Bandura 1991; Bushwick 2001;Fotietal.2014;
Lord 2010; Rizzolatti and Sinigaglia 2010).
Modeling of appropriate behavior and inclusion in classrooms
with typically functioning (TF) peers helps students with ASD
develop social skills (Downey and Rapport 2012; Pan et al.
2011). Physically imitating actions of other individuals, as op-
posed to listening or reading instructions, often simplifies com-
munication and minimizes confusion (Hamilton et al. 2007).
This can be especially important for students with severe forms
of ASD (Gallese et al. 2012; Mostofsky et al. 2004). For exam-
ple, the simple act of throwing a ball would be more easily
described by physically performing the action as opposed to
describing it using words. Students with severe ASD may be
non-verbal, yet modeling offers an alternative form of commu-
nication that relies on body language and physical expression.
Retention of social skills is often increased by consistently
reinforcing the modeling of appropriate behaviors (Bushwick
2001;Ingersoll2008). Reinforcement involves use of praise
and rewards to increase motivation. Motivation to behave ap-
propriately, complete tasks correctly, and participate in
classroom activities is vital to prevent the rigid routines and
limited interests, characteristic of ASD, which interfere with
learning in classrooms (Bass et al. 2009; Pan et al. 2011).
Motivation can also be affected by levels within the spectrum
of ASD (Hilton et al. 2008). For example, higher functioning
students may be motivated by praise and encouragement
whereas lower functioning students may need more salient
rewards such as toys or food (Srinivasan et al. 2014).
These components of social learning theory are often uti-
lized directly and indirectly in SE studies for students with
ASD, yet research literature often inadequately addresses
these influences. While exercise research findings support in-
creased SE and extended PE classes may help increase SD for
students with ASD, often they are limited by a focus on form,
length, and intensity level of activity (Hsu et al. 2004;Pan
et al. 2011). This often overshadows the positive impact on
SD (observation, modeling, retention, and motivation) during
exercise routines (Bahrami et al. 2012; Spitzer and Hollmann
2013;ToddandReid2006).
Utilization of Social Learning
During Structured Physical Activities
Many SE studies incorporate components of social learning
for students with ASD even if they are not the focus of the
study. A review of SE by Lang et al. (2010) is an excellent
example of how each of the four components of social learn-
ing are often incorporated in a SE context. Lang et al. (2010)
found that exercise had an overall positive effect on behavior
for students regardless of age or place on the ASD spectrum.
Specifically, exercise was found to lower the incidence of
aggression and off-task behaviors. SE activities were imple-
mented for 64 individuals with ASD with an average age of
12.5 years (M= 12.5 years, SD = 3.27). Lang et al.’s(2010)
intervention allowed for many opportunities for positive rein-
forcement and modeling, vital social learning cues for stu-
dents with ASD (Bushwick 2001;Ingersoll2008).
Participants were verbally given instructions, feedback, and
praise throughout the activity.
PAwas defined as jogging, weight training, and bike riding.
Non-verbal communication was also consistently utilized
through eye contact and observation of correct jogging form
by the instructor before and during the activity. Exercises were
modeled, and physical guidance was used to teach activities.
Most participants had limited experiences with any type of
SE, so an instructor would jog alongside students to help
guide them through implicit and explicit communication.
Lang et al. (2010) described the instructors’interactions with
students as being necessary to ensure participants exercised
for a defined length of time and correctly performed the ac-
tivity. Therefore, while exercising, participants were concur-
rently socially interacting with others by observing
Rev J Autism Dev Disord

appropriate behavior with their instructor prompting and pro-
viding praise as well as technical instruction for each student
when necessary. Despite a range within the spectrum of ASD
for the various participants, exposure to a variety of commu-
nication methods likely increased their ability to correctly par-
ticipate. Immediately following these exercises, participants
experienced decreases in aggression and off-task behaviors
in the classroom (Lang et al. 2010).
While Lang et al.’s(2010) findings indicate that negative
behaviors were decreased immediately following exercises,
the long-term effects on behavior were not addressed. While
findings were positive, the retention of exercise skills learned
may have differed for students with different severities of
ASD. For example, lower functioning students may rely more
heavily on modeling movements and may lose jogging skills
over time if not continuously practiced and reinforced.
However, students deemed as high functioning on the ASD
spectrum may have an easier time retaining their jogging skills
which may decrease social learning opportunities. The use of
observation, praise, and modeling was likely to decrease if
these students already had learned how to jog.
In contrast to Lang et al. (2010), Movahedi et al. (2013)
examined that social behavior improvements for students with
ASD were retained for a longer period of 30 days following
participation in a Kata (karate) program lasting 3 months. A
total of 30 students with an ASD diagnosis from ages 5 to
16 years (M= 9.13 years, SD = 3.27) participated. Movahedi
et al. (2013) employed several strategies to reduce group dif-
ferences and confounding variables in their study. They were
selected from the same educational institution to ensure they
had experienced the similar treatment strategies. They were
matched into pairs based on severity of autism, chronological
age, and gender pre-intervention. These pairs were then ran-
domly assigned into an exercise (n= 15) or control (n=15)
group. Participants in an exercise intervention group learned
karate techniques by viewing a video before each class. They
worked with trainers learning techniques 30 to 90 min a day,
4 days per week, for 3 months. Various components of social
learning took place throughout these sessions. A karate video
was used which allowed students to observe and imitate tech-
niques. Students also practiced skills as a group where they
could observe peers and model appropriate movement.
Additionally, exercise took place in what is described as a
highly motivational climate. Verbal praise, prompting, and
encouragement were used as often as possible during each
exercise session. The use of these social components during
the karate activities likely contributed to behavior
improvements.
Social dysfunction was measured using the Gilliam Autism
Rating Scale-Second Edition (GARS-2). This social interac-
tion subscale has been widely used in both educational set-
tings and research studies. The GARS-2 consists of a 14-item
questionnaire for caretakers and teachers that describe specific
observable as well as measurable social behaviors. Students
are then rated on the frequency of the occurrence of eachof the
14 social behaviors under typical circumstances over a 6-h
period. This subscale has excellent psychometric properties
and is both valid and reliable (Worley and Matson 2011).
Differences in social dysfunction severity baselines (pre-
intervention) were examined by performing an independent t
test for the GARS-2 and indicated no significant differences
t(24) = −.31, p=.76.
Findings for the Kata intervention group revealed signifi-
cant improvement in social interactions post-intervention.
Dysfunction baseline levels were decreased by 40.32% across
participants. Movahedi et al. (2013) findings indicate that so-
cial improvements were retained for at least 30 days following
the program for the participants in the exercise group, with no
change in the control group. These findings indicate the pos-
sibility that social gains are not always ephemeral and may
have a positive impact on SD over longer periods of time.
Further examination of how long SD skills are sustained for
students with ASD may aid in the development of targeted
intervention programs which address social deficits while
maximizing retention of skills over longer periods of time
beyond 30 days.
These findings are promising; however, Movahedi et al.
(2013) may have overestimated the impact of this specific
karate activity. While the form of karate is discussed as an
important component which influenced findings, the social
components utilized could be incorporated into a wide variety
of SE types such as team sports like basketball and soccer.
Videos providing observation opportunities and practicing
with a teacher led group of peers also occurs in many other
sports contexts to help improve skill through visual observa-
tion (Anderson-Hanley et al. 2011; Vazou et al. 2005).
Increased social interaction would appear more likely a result
of implementation of social learning components (integration
with peers, teacher modeling) as opposed to the specific type
of exercise.
Physical Exertion and Fatigue
The influence of physical exertion and fatigue on behavior is
often overstated or at times misunderstood, in SE studies. For
instance, Lang et al.’s(2010) findings support the use of in-
creased exercise, yet exertion during exercise was unlikely the
sole factor that improved behavior for students. Despite varied
types of exercises, components of social learning were consis-
tently present throughout the activities. Participants did not
exercise alone; an instructor accompanied them, consistently
modeling socially appropriate behavior. If exercises had taken
place in an isolated setting with no prompting or social inter-
action, social behavior changes may have differed significant-
ly. Despite variations in the physical intensity of exercise, all
Rev J Autism Dev Disord

participants displayed improvements in behavior regardless of
baseline fitness levels.
This suggests that social interaction, praise, and modeling
appropriate behavior during a SE activity may be an effective
tool for development of social skills. Positive changes in
behavior were unlikely attributed to simply increasing SE
intensity levels to make participants too tired to be
aggressive or engage in off task behavior. Lang et al. (2010)
discussed how it is unlikely that student fatigue from SE’s
influenced behavior changes because students increased their
attentiveness in the classroom and performed better on their
academic tasks immediately following their participation.
These findings support that tools integrated from social learn-
ing theory during SE routines likely influenced behavior
changes, not solely the increase of energy expenditure.
However, the effects of physical exertion cannot be
discounted completely. Participants had different functioning
and fitness levels. Often students with severe ASD struggle
with inappropriate movement and have numerous sensory
problems which may have been alleviated through exertion
during SE (Morrison et al. 2011; Todd et al. 2010).
Behaviors may have improved for these students on the sur-
face, but it would be beneficial to examine ifthere are different
social outcomes related to the different levels on the spectrum
of ASD.
A study by Petrus et al. (2008) provides an example of how
increased exertion in SEs may lead to increased self-regulation
of stereotypic behaviors for students on the low end of the
spectrum of ASD. The effect of increased SE exertion on de-
creasing harmful stereotypic behaviors during school was ex-
amined for students age 4 through 15 years. This systemic
review of previous research of SE concluded more rigorous
levels of exercise correlated with short-term reductions of ste-
reotypic behavior. Like Lang et al.’s(2010) research, regardless
of variances in intensity levels and type of exercises, stereotypic
behavior (abnormal, repetitive movement) was decreased im-
mediately following SE (Petrus et al. 2008).
However, Petrus et al. (2008) found that although reduc-
tions in behaviors were greater following increased levels of
SE, reductions in movement were likely a result of exhaustion
or fatigue rather than skill development. Exercise interven-
tions in this review included intense aerobic exercise and jog-
ging for intervals lasting 20 min. The exercise interventions
focused solely on the reduction of stereotypic behavior
through energy expenditure. Opportunity for social learning
was reduced as exercises were performed solo and often at a
vigorous pace. Interaction or inclusion with TF students was
non-existent.
These findings exemplify that only increasing exertion dur-
ing SE is not an effective way to maximize benefits for stu-
dents academically or socially. Additionally, while decreasing
stereotypic behavior for students with ASD is a concern, it has
become less of a priority compared to developing social and
communication skills (Richler et al. 2010). Findings by
Nicholson et al. (2011) suggested that intensity level of SEs
was far less important than routine, modeling, and consistent
incorporation within classrooms. It was concluded that con-
cise low-intensity SEs were found to be easier to incorporate
into daily school activities and allow students of all fitness
levels to participate, providing much needed opportunities to
bolster SD through practice and observation (Mahar et al.
2006; Nicholson et al. 2011; Pan et al. 2011).
Fitness Level
Students with ASD often have lower levels of fitness and
fewer opportunities to participate in SEs, such as PE classes
and sports teams, than TF peers (Memari et al. 2012; Pan
2008). These deficits in SE participation may further com-
pound the social and cognitive development issues for
students with ASD, a result of the failure to provide the
same experiences as TF students. Pan and Frey (2006)exam-
ined how these differences in activity may affect behavior.
Pan and Frey (2006) compared activity levels of students
with ASD with TF students. Activity levels of 30 students
with ASD aged 10–19 years were measured. To ensure precise
measurement, students were required to wear an accelerome-
ter (an electronic device which records activity level) for
1 week. SE levels for students with ASD were found to be
lower than standards for TF students. Pan and Frey’s(2006)
findings indicated that participation in extracurricular activi-
ties was also significantly limited for students with ASD.
However, the lower levels of exercise were unlikely to be a
physiological explanation for behavior problems. Opportunity
for social interaction was theorized to be more influential on
development than a low fitness level from a lack of exercise
(Pan and Frey 2006). Furthermore, there is evidence that fit-
ness level and physical ability may contribute to a lack of
social interaction, especially for higher functioning students
with ASD, by preventing them from developing skills and
joining teams. This is especially damaging for higher func-
tioning students because of their higher baseline social skills
which increase their ability to interact appropriately in inclu-
sive settings with TF peers (Hilton et al. 2008). Participation
in all types of social activities outside of school was found to
be much lower for higher functioning students with ASD
compared with TF students. While motor impairments are
more common for students with ASD, Hilton et al.’s(2008)
findings indicated that social factors may also limit
participation.
The influence of these social factors for participation was
examined by Frey and Sandt (2005). Participants in the study
were 15 students with ASD and 13 TF students ages 5 to
12 years (M= 9.2 years, SD = 1.96). Both groups of students
wore an accelerometer (a device that keeps track of PA level)
Rev J Autism Dev Disord

for a total of 5 days for the entire day. While Frey and Sandt
(2005) hypothesized that PA levels would differ significantly
between the two groups, their data did not reveal any differences
in activity level between the two groups. Despite both groups
having similar activity levels, the two groups had significant
differences in the structure of exercise activities during both SE
and UE. Students with ASD were observed having less recess
time (UE) than TF students because of their need for extra aca-
demic tutoring and behavioral support. PA levels remained sim-
ilar because of negative repetitive behaviors rather than partici-
pation in an appropriate exercise activity. Students in the ASD
group had very little social interaction during UE in contrast to
TF students who were far more likely to participate in organized
group orientated PA. SE differences were found to be even more
significant for students with ASD (Frey and Sandt 2005).
Outside of school, the study found that students with ASD were
not on sports teams and often were physically active because of
negative behaviors not because of a constructive activity.
Additionally, during PE classes, students required extra verbal
instructions, and physical prompts to complete activities. This
reduced the actual time that was spent exercising properly, limit-
ing the opportunity for SD during SE in PE classes. Researchers
concluded that similar PA levels do not necessarily indicate equal
opportunity for SD. The need for proper supervision, staff sup-
port, and additional time was recommended to allow for addi-
tional social opportunities for students with ASD which are ex-
tremely scarce outside of school.
Findings by Hilton et al. (2008) provide further evidence
about how differing levels of support impact SD for students
with ASD across the spectrum. Out-of-school activity partic-
ipation levels between a control group of TF students (N=53)
and high functioning students with ASD (N=52) were
assessed with the use of the Children’s Assessment of
Participation and Enjoyment (CAPE) and the Social
Responsiveness Scale (SRS). Significant differences were
found in PA participation patterns between the two groups of
students. High functioning students with ASD had significant-
ly lower number of activities in which they participated in, less
individuals with whom they spend time with, and significant
lack of exposure to different types of exercise environments.
Socially, higher functioning students with ASD were
foundtobelesslikelytochooseactivitiesinwhichthey
have less competence. CAPE testing indicated that en-
joyment during exercise was increased in direct relation
to ability to correctly perform an activity. In contrast, no
significant differences were found between student par-
ticipation in skill-based activities, such as swimming,
horseback riding, and singing. Skill-based activities are
planned, have specific rules, and have adult supervision
(Hilton et al. 2008). When performing activities in
which they are competent, high functioning students
with ASD participated no differently than TF students
(Hilton et al. 2008). These findings indicate that social
impairment may result from a lack of opportunity for
participation in SE.
Limited Social Development Opportunities
Students with ASD already have deficits in social skills which
would likely be exacerbated by isolation from physical activ-
ities with TF peers (Hilton et al. 2008; Pan and Frey 2006;Pan
et al. 2011). Often low fitness levels and poorly developed
motor skills also limit the amount of activities in which stu-
dents with ASD can participate. Students are often placed in
separate classrooms where their classmates struggle with sim-
ilar social and behavioral problems. In contrast, TF students
have a vast array of activities available for socializing, observ-
ing appropriate behavior, and developing communication
skills in non-academic settings. Sports teams for example fol-
low rules, communicate during the activity, and practice as a
group. Communication may involve praise for correctly
performing an activity as well as reinforcement of skills
through consistent practice. Behavior problems for students
with ASD, therefore, cannot only be attributed to a lack of
activity, but fewer opportunities for the social interactions
which take place in SE context (Hsu et al. 2004;Toddand
Reid 2006).
Findings support that children who are exposed to more
social interaction are more likely to improve and develop so-
cial skills (Nieman 2002). Nieman (2002) found that when 5
to 6-year-old school-aged children participated in aerobic ex-
ercise activity for 5–8 min daily, they reduced their self-
stimulatory behavior immediately following the activity.
Opportunities to interact socially through SE are far greater
for TF children from the beginning of their schooling. In con-
trast, students with ASD are far less likely to be on sports
teams and participate in exercise activities with peers (Sandt
and Frey 2005;Smith2003; Srinivasan et al. 2014).
Hillman et al. (2009)findingsalsosupporthowSDmaybe
improved in a SE context through participation. The effect of
exercise on attention and social behavior was assessed follow-
ing a moderate 20-min jogging routine for 20 TF children
(N= 20) with an average age of 9.5 years. The study had half
the participants rest and half jog on different days and would
then test both groups using the Wide Range Achievement Test
(a validated cognition test). Findings indicated that cognitive
test scores were higher for participants following the 20-min
aerobic exercise than scores following a 20-min resting ses-
sion. (Hillman et al. 2009). These findings suggest that SE
may positively impact SD immediately following PA.
Srinivasan et al. (2014) findings also have indicated social-
cognitive benefits of inclusive participation for students dur-
ing exercise. Deficits in social skills for students with ASD
would likely be further exacerbated by isolation from physical
activities with TF peers (Pan et al. 2011; Todd and Reid 2006).
Rev J Autism Dev Disord

SD problems therefore may possibly be attributed to limited
opportunities for participation, rather than the physiological
effect of SE. This is important because lack of exercise pro-
grams for students with ASD limits opportunities to observe
appropriate behavior and practice communication skills.
These findings are similar to Lang et al. (2010) which indicat-
ed that physical exertion itself was unlikely to influence social
behavior for students with ASD. For example, students with
ASD were found to be typically placed in special education
programs separate from other students. Parents and teachers
reported that there were few, if any, inclusive or even segre-
gated SE programs for students with ASD (Srinivasan et al.
2014). This severely limits opportunities for social interaction
and observation of appropriate behavior of TF peers.
Srinivasan et al. (2014) also found that students with ASD
attended schools significantly farther from their residences
than peers without special needs and typically have unique
transportation requirements (often traveling to school in a sep-
arate bus) which may interfere with joining sports teams and
other social activities after school. Sedentary time was also
found to be greater for students with ASD and again was
exacerbated by longer commutes to school. To address these
deficits, Srinivasan et al. (2014) suggested efforts be made by
teachers and administrators to discover ways to increase ap-
propriate in-school and leisure time SE. Increased participa-
tion in these activities would provide students with ASD more
opportunities to develop socially through observation, imita-
tion, and reinforcement.
AstudybyPan(2008) also provides evidence that social
skills may develop in part from participation in SE. The fitness
levels of students with and without ASD from 14 elementary
schools were compared during inclusive PE classes and re-
cess. This study compared the percent children with and with-
out ASD spent in moderate-to-vigorous PA (MVPA) during
inclusive recess settings. Forty-eight children, matched by age
and gender (ASD, 23 boys and 1 girl; non-ASD, 23 boys and
1 girl) aged 7–12 years from 14 schools, had their PA during
recess recorded using an accelerometer device for 5 days dur-
ing school time. Findings indicated that children with ASD
were less active overall throughout the day, during recess, PE
classes, and academic classes when compared to the control
group without any disabilities (p<.01).
Students with ASD performed SEs at the same intensity
level during PE classes as students without ASD. These stu-
dents were therefore participating in exercise activities no dif-
ferently than TF students. The SE context offered social learn-
ing opportunities such as observation of appropriate behavior
and imitation. However, during recess times with less struc-
ture, students with ASD were found to have significantly low-
er levels of participation in all forms of PA (Pan 2008). These
low levels of participation during UE further supports the
importance of incorporating SEs for students with ASD to
provide the same opportunities for social growth during SE
as TF students. The unstructured nature of recess may not
provide the same social stimulation for students with ASD
as when students participate with instructors guiding, model-
ing, and encouraging.
Structured vs Unstructured Exercise
Students with ASD have limited opportunities for both SE as
well as UE. Each exercise context often has different implica-
tions for social learning for students. Mastrangelo (2009)ex-
amined the potential for developing social and cognitive skills
through UE which is described as Bplay.^The standard struc-
tured, behavioral-based unstructured activities were theorized
to be too goal oriented, adult controlled, and involuntary to be
truly beneficial (Mastrangelo 2009). This contrasts with find-
ings which support the use of structure, guidance, modeling of
appropriate behavior, and social interaction to improve devel-
opment for students with ASD (Richler et al. 2010). Having a
student with ASD voluntarily play in an appropriate manner
appears quite challenging. While Mastrangelo (2009)isin
favor of increasing unstructured activities, this might be espe-
cially challenging for individuals with ASD. UE is theorized
as a method to address social and cognitive issues for students
with ASD, yet there is a lack of empirical data to definitively
determine if it is more effective than SE. This makes it difficult
to determine the significance of its influence on SD and iden-
tify which methods are most effective to increase low levels of
UE for students with different severities of ASD.
These possible deficits in UE for pre-school-aged students
with ASD were examined by Rutherford and Rogers (2003).
Participants included 28 children with ASD and a control
group of 27 children without ASD. Frequency and length of
play with novel objects were measured during a structured
interview for 30 to 45 min. An interviewer would present a
student with a set of toys on a table. The interviewer then
instructed the students to play with the toys. If the child did
not play spontaneously, a prompt was given. This prompt was
in the form of either modeling the play behavior or giving
instructions. When data from children with ASD was com-
pared to the control group, significant deficits in attention,
imitation, and communication were revealed. Even when
prompted, they often did not play appropriately or play at
all. They also had less eye contact and were less likely to
model appropriate play compared with the control group.
Findings by Rutherford and Rogers (2003) provide evi-
dence that social learning skills are not properly developed
in very young children regardless of the level of ASD. These
findings suggested that an increase in physical play at a young
age may address deficits in expressive and receptive commu-
nication which involves important aspects of social learning
such as imitation, observation, and emotion (Bushwick 2001).
Rutherford and Roger’s(2003) findings also address
Rev J Autism Dev Disord

developmental concerns by utilizing components of social
learning during activity, unlike the more UEs described by
Mastrangelo (2009). However, findings may have been
strengthened if more frequent observations of the participants
had been conducted. Data measured indicated deficits in play
but was collected from only two observations with each par-
ticipant. While the data samples were small, they indicated
that incorporation of social learning through structured phys-
ical play may provide SD benefits for children with ASD.
SE programs have often been found to decrease negative
social behaviors for students with ASD (Pan et al. 2011;
Srinivasan et al. 2014; Todd and Reid 2006). Pan et al.
(2011) examined increased self-regulation for students with
ASD who were included in PE classes with TF students.
Researchers believed that the inclusion would allow the op-
portunity for students with ASD observe and mirror appropri-
ate in class behavior. The study concluded that students sig-
nificantly increased their attendance to task, motivation, and
appropriate participation in PE classes.
PE class settings provide many opportunities for peer-to-
peer interaction, instruction, praise, and modeling. However,
the opportunity for students with ASD to interact and observe
TF students would likely be diminished during UE activities
such as recess. The social deficits for students with ASD
would make it particularly challenging to interact and
participate with TF peers during UE. Pan et al. (2011)found
a positive effect on regulation of behavior when including
students with ASD in PA classes with TF students. Pan et al.
(2011) observed increased appropriate behavior during SE in
PE classes. Students with ASD increased their ability to con-
centrate and follow directions through physical participation
in activities with TF peers. Participants in Pan et al.’s(2011)
study were 25 students with ASD and 75 TF peers. Level of
activity was measured by an accelerometer worn by the stu-
dents and assessed during 1 week of PE classes. Students with
ASD participated in these classes concurrently with TF stu-
dents and were treated no differently by the instructor. Data
was then recorded and used to examine differences in activity
levels between the two groups. Findings from this study indi-
cated increased external regulation of behaviors in students
with ASD correlated positively with higher levels of active
participation. SE during these classes, therefore, increased
the ability for students with ASD to regulate their behaviors.
Students without ASD did not show any correlation (Pan et al.
2011).
Pan et al. (2011) also concluded that motivation for stu-
dents with ASD to regulate behavior during physical activities
increased through observation and imitation of TF students.
The findings related to motivation levels revealed the potential
use of SE not only to regulate behavior but to increase social
skills. Behavior regulation allowed students with ASD to in-
crease their ability to follow directions and concentrate
on their activity. These SE were not academic, yet
self-regulation and social skills are consistently a focus for
teachers of students with ASD, even in non-academic settings
(Lord 2010).
Future Directions
Overall, a range of SE such as basketball, soccer, biking, jog-
ging, and weight training have been shown to increase posi-
tive social behaviors, attendance to task, and communication
skills for students with ASD (Todd and Reid 2006; Srinivasan
et al. 2014). The type of exercise, intensity, and duration do
not appear to affect student behavior as much as opportunities
for SD. Examination of the influence of SD during exercise is
imperative because of its enormous impact on findings which
typically focus on physiological benefits (Srinivasan et al.
2014). Developing communication skills and increasing ap-
propriate classroom behavior are extremely challenging, yet
exercise provides an alternative learning tool for students with
ASD. Social benefits also appear to be enhanced when PA
programs incorporate social learning theory. Structured phys-
ical activities improve behavior for students with ASD
through observation, imitation, and inclusion with TF stu-
dents. Exercise affords students with ASD a unique opportu-
nity to increase vital social skills outside of traditional class-
room settings. However, SE programs need to be developed
carefully and utilize exercise not simply as a novel classroom
activity, but as an effective social learningtool (Todd and Reid
2006).
Research regarding the impact of all types of exercise on
attention and self-regulation is often limited by small sample
sizes, limited empirical attention measures, and variable types
of PA. There is a general lack of literature about how students
with ASD gain social and academic benefits from exercise.
Studies involving SE for students with ASD are scarce as
often this population does not participate together, rather they
are often integrated with TF students or work separately with
an instructor to exercise. ASD is also a spectrum disorder
making it challenging to find a large group with matching
IQ levels and social abilities. Students with ASD also often
need more guidance and supervision than TF participants be-
cause of social, behavioral, and communication difficulties.
However, the possible gains in social and cognitive skills
for students with ASD are too valuable to be ignored despite
challenges faced by researchers. Further examination of how
social learning can be utilized to provide physiological bene-
fits as well as targeted social skill improvements using exer-
cise can lead to more effective PE programs for students with
ASD in schools.
In the future, teachersmay incorporate physicalactivitiesin
the classroom to develop social skills and increase self-
regulation as part of their curriculum. While utilizing social
learning theory has been particularly effective for teaching
Rev J Autism Dev Disord

students with ASD, there has been limited direct application
related to SE to specifically increase SD. However, findings in
current exercise literature have consistently found that in-
creased exercise has enormous potential as a positive influ-
ence on SD. These findings should encourage researchers in
the fields of both education, as well as psychology, to continue
to identify the most effective methods to utilize exercise to
increase positive SD for students with ASD.
Compliance with Ethical Standards
Ethical Approval All procedures performed in studies involving human
participants were in accordance with the ethical standards of the institu-
tional and/or national research committee and with the 1964 Helsinki
declaration and its later amendments or comparable ethical standards.
Conflict of Interest The authors declare that they have no conflict of
interest.
References
American Psychiatric Association. (2013). Diagnostic and statistical
manual of mental disorders (5th ed.). Arlington: American
Psychiatric Publishing.
Anderson-Hanley, C., Tureck, K., & Schneiderman, R. L. (2011). Autism
and exergaming: effects on repetitive behaviors and cognition.
Psychology Research and Behavior Management, 4,129–137.
Bahrami, F., Movahedi, A., Marandi, S. M., & Abedi, A. (2012). Kata
techniques training consistently decreased stereotypy in children
with autism spectrum disorder. Research in Developmental
Disabilities, 33, 1183–1193.
Bandura, A. (1977). Social learning theory. Englewood Cliffs: Prentice
Hall.
Bandura, A. (1991). Social cognitive theory of self-regulation.
Organizational Behavior and Human Decision Processes, 50(2),
248–287.
Bass, M. M., Duchowny, A., & Llabre, M. M. (2009). The effect of
therapeutic horseback riding on social functioning in children with
autism. Journal of Developmental Disorders, 39,1261–1267.
Bushwick, N. (2001). Social learning and the etiology of autism. New
Ideas in Psychology, 19(1), 49–75.
Downey, R., & Rapport, M. K. (2012). Motor activity in children with
autism: a review of current literature. Pediatric Physical Therapy,
24,2–20.
Foti, F., Mazzone, L., Menghini, D., De-Peppo, L., Federico, F.,
Postorino, V., et al. (2014). Learning by observation in children with
autism spectrum disorder. Psychological Medicine, 44,2437–2447.
Gallese, V., Rochat, J. R., & Berchio, C. (2012). The mirror mechanism
and its potential role in autism spectrum disorder. Developmental
Medicine & Child Neurology, 55,15–22.
Hamilton, A. F. D. C., Brindley, R. M., & Frith, U. (2007). Imitation and
action understanding in autistic spectrum disorders: how valid is the
hypothesis of a deficit in the mirror neuron system?
Neuropsychologia, 45(8), 1859–1868.
Hillman, C. H., Pontifex, M. B., Raine, L. B., Castelli, D. M., Hall, E. E.,
& Kramer, A. F. (2009). The effect of acute treadmill walking on
cognitive control and academic achievement in preadolescent chil-
dren. Neuroscience, 159(3), 1044–1054.
Hilton, L. H., Crouch, M. C., & Israel, H. (2008). Out of school partici-
pation patterns in children with high-functioning autism spectrum
disorders. The American Journal ofOccupational Therapy, 62,554–
563.
Holifield, C., Goodman, J., Hazelkorn, M., & Heflin, L. J. (2010). Using
self-monitoring to increase attending to task and academic accuracy
in children with autism. Focus on Autism and Other Developmental
Disabilities, 25(4), 230–238.
Hsu, H. C., Chen, C. L., Cheng, P. T., Chen, C. H., Chong, C. Y., & Lin,
Y. Y. (2004). The relationship of social function with motor and
speech functions in children with autism. Chang Gung Medical
Journal, 27(10), 750–757.
Ingersoll, B. (2008). The social role of imitation in autism: implications
for the treatment of imitation deficits. Infants & Young Children,
21(2), 107–119.
Lang, R., Koegel, L. K., Ashbaugh, K., Regester, A., Ence, W., & Smith,
W. (2010). Physical exercise and individuals with autism spectrum
disorders: a systematic review. Research in Autism Spectrum
Disorders, 4(4), 565–576.
Lord, C. (2010). Autism: from research to practice. American
Psychologist, 65,815–826.
Mahar, M. T., Murphy, S. K., Rowe, D. A., Golden, J., Shields, A. T., &
Raedeke, T. D. (2006). Effects of a classroom-based program on
physical activity and on-task behavior. Official Journal of the
American College of Sports Medicine, 38,2086–2094.
Mastrangelo, S. (2009). Play and the child with autism spectrum disorder:
from possibilities to practice. International Journal of Play Therapy,
18(1), 13–30.
Memari, A. H., Ghaheri, B., Ziaee, V., Kordi, R., Hafizi, S., & Moshayedi, P.
(2012). Physical activity in children and adolescents with autism
assessed by triaxialaccelerometry. Pediatric Obesity, 7,47–63.
Morrison, H., Roscoe, E. M., & Atwell, A. (2011). An evaluation of
antecedent exercise on behavior maintained by automatic reinforce-
ment using a three-component multiple schedule. Journal of Applied
Behavior Analysis, 44,523–541.
Mostofsky, S. H., Bunoski, R., Morton, S. M., Goldberg, M. C., &
Bastian, A. J. (2004). Children with autism adapt normally during
a catching task requiring the cerebellum. Neurocase, 10(1), 60–64.
Movahedi, A., Bahrami, F., Marandi, S. M., & Abedi, A. (2013).
Improvement in social dysfunction of children with autism spectrum
disorders following long term Kata techniques training. Research in
Autism Spectrum Disorders, 7,1054–1061.
Nicholson, H., Kehle, T. J., Bray, M. A., & Heest, J. V. (2011). The effects
of antecedent physical activity on the academic engagement of chil-
dren with autism spectrum disorder. Psychology in the Schools,
48(2), 198–213.
Nieman, P. (2002). Psychosocial aspects of physical activity. Paediatrics
& Child Health, 7(5), 309–312.
Pan, C. Y. (2008). Objectively measured physical activity between chil-
dren with autism spectrum disorders and children without disabil-
ities during inclusive recess settings in Taiwan. Journal of Autism
and Developmental Disorders, 38(7), 1292–1301.
Pan, C. Y. (2010). Effects of water exercise swimming program on aquat-
ic skills and social behaviors in children with autism spectrum dis-
orders. Autism, 14(1), 9–28.
Pan, C. Y., & Frey, G. C. (2006). Physical activity patterns in youth with
autism spectrum disorders. Journal of Autism and Developmental
Disorders, 36(5), 597.
Pan, C. Y., Tsai, C. L., Chu, C. H., & Hsieh, K. W. (2011). Physical
activity and self-determined motivation of adolescents with and
without autism spectrum disorders in inclusive physical education.
Research in Autism Spectrum Disorders, 5(2), 733–741.
Petrus, C., Adamson, S. R., Block, L., Einarson, S. J., Sharifnejad, M., &
Harris, S. R. (2008). Effects of exercise interventions on stereotypic
behaviours in children with autism spectrum disorder.
Physiotherapy Canada, 60(2), 134–145.
Richler, J., Huerta, M., Bishop, S. L., & Lord, C. (2010). Developmental
trajectories of restricted and repetitive behaviors and interests in
Rev J Autism Dev Disord

children with autism spectrum disorders. Development and
Psychopathology, 22,55–69.
Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the parieto-
frontal mirror circuit: interpretations and misinterpretations. Nature
Reviews Neuroscience, 11(4), 264–274.
Rutherford, M. D., & Rogers, S. J. (2003). Cognitive underpinnings of
pretendplayinautism.Journal of Autism and Developmental
Disorders, 33,289–302.
Sandt, D. D. R., & Frey, G. C. (2005). Comparison of physical activity
levels between children with and without autistic spectrum disor-
ders. Adapted Physical Activity Quarterly, 22(2), 146–159.
Smith, A. L. (2003). Peer relationships in physical activity contexts: a
road less traveled in youth sport and exercise psychology research.
Psychology of Sport and Exercise, 4(1), 25–39.
Spitzer, S. S., & Hollmann, W. (2013). Experimental observations of the
effects of physical exercise on attention, academic and prosocial
performance in school settings. Trends in Neuroscience and
Education, 2(1), 1–6.
Srinivasan, S. M., Pescatello, L. S., & Bhat, A. N. (2014). Current per-
spectives on physical activity and exercise recommendations for
children and adolescents with autism spectrum disorders. Physical
Therapy, 94(6), 875–889.
Todd, T., & Reid, G. (2006). Increasing physical activity in individuals
with autism. Focus on Autism and Other Developmental
Disabilities, 21(3), 167–176.
Todd, T., Reid, G., & Butler-Kisber, L. (2010). Cycling for students with
ASD: self-regulation promotes sustained physical activity. Adapted
Physical Activity Quarterly, 27,226–241.
Vazou, S., Ntoumanis, N., & Duda, J. L. (2005). Peer motivational cli-
mate in youth sport: a qualitative inquiry. Psychology of Sport and
Exercise, 6(5), 497–516.
Worley,J.A.,&Matson,J.L.(2011).Diagnostic instruments for
thecorefeaturesofASD. In International handbook of au-
tism and pervasive developmental disorders (pp. 215–231).
New York: Springer
Rev J Autism Dev Disord