R E S E A R C H R E P O R T

sEMG Analysis During Landing in Children With Autism Spectrum Disorder: A Pilot
Study
Marcelo R. Rosales, BS; Jennifer Romack, PhD; Rosa Angulo-Barroso, PhD
Department of Kinesiology, California State University, Northridge, California.

Purpose: The aim was to explore the timing and duration of muscle activation during a landing task in children with autism
spectrum disorder (ASD), and compare their responses to those of children who are developing typically (TD).
Methods: Six children (ages 3-4.5 years), half with ASD, hung from a vertical bar, landed, and reacted to a light cue that
signaled the child to run to the right or left or to stay in place. Electromyography and kinematics were recorded and
compared between groups.
Results: Children with ASD had more and longer bursts of muscle activation during preimpact. In contrast, children TD
Downloaded from http://journals.lww.com/pedpt by BhDMf5ePHKbH4TTImqenVIukvCljlO2sFdxsECC2BlPG26vrifukQFwowIWH0QOj on 07/13/2018

displayed more and longer burst of muscle activation during impact.

Conclusion: The results suggest that children with ASD have a less developed landing strategy compared with their peers
TD. Further investigation into the neuromuscular components in children with ASD will guide future interventions for this
population. (Pediatr Phys Ther 2018;30:192–194)
Key words: autism, electromyography, landing

INTRODUCTION
Autism spectrum disorder (ASD) is a heterogeneous dis-
ability affecting 1 in every 68 children in the United States.1
Children diagnosed with ASD are typically characterized by
socioemotional2 and motor behavior deficits.3,4 These motor
behavioral problems have made ASD of interest when it comes
to early interventions to help prevent and improve these motor
deficits. Although extensive behavioral data exist depicting
motor difficulties in children with ASD3,4 and motor interven-
tions are recommended,5 there are still many unanswered ques-
tions about the motor control characteristics of ASD.
One aspect that has been explored minimally in children
with ASD is the neuromuscular components of potential motor
deficits. One group of researchers has explored the neuromus-
cular differences in gross motor behaviors between children with
ASD and children who are developing typically (TD).6 Mohd
et al6 report that children with ASD had lower mean amplitude
muscle burst in the gastrocnemius during the stance phase, and
in the rectus femoris during the swing phase, of the gait cycle
compared with children TD. However, no differences between
the groups were found for the biceps femoris or tibialis anterior
(the other muscles examined).
In addition to walking, jumping and landing are also
common motor skills of children. Focusing on a fundamental
motor skill, such as landing, is important, given the established
relationship between fundamental motor skill and cognition7
and physical activty.8 The purpose of this pilot study was to
explore the timing and duration of muscle activation during a
landing task in children with ASD and compare their responses
to children TD. This study is an attempt to expand the literature
and provide guidance for future research.

0898-5669/110/3003-0192
Pediatric Physical Therapy
Copyright © 2018 Academy of Pediatric Physical Therapy of the American
Physical Therapy Association
Correspondence: Marcelo R. Rosales, BS, Department of Kinesiology, Cal-
ifornia State University, Northridge, 18111 Nordhoff St, Northridge, CA
91330 (marcelo.rosales.826@my.csun.edu).
At the time this article was written, Marcelo Rosales was an undergraduate
student at California State University in the Department of Kinesiology.
The authors declare no conflicts of interest.
DOI: 10.1097/PEP.0000000000000514
METHODS
Participants
Six children, 3 with ASD and 3 TD between the ages of 3 and
4.5 years, participated. Parents provided information regarding
ASD diagnosis. These children were part of a larger study con-
ducted at California State University, Northridge, exploring the
development of landing strategies in children TD. Children were
recruited from schools in the San Fernando Valley and service
learning programs at the university. Parents provided informed
consent to participate in the study. The Institutional Review
Board at California State University, Northridge, approved the
study.

192 Rosales et al Pediatric Physical Therapy

Copyright © 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
Procedures
The participants and their caregivers came to the motor
development laboratory where a questionnaire regarding diag-
nosis of developmental disabilities was completed.
Participants were asked to change into compression shorts
and a tank top and to remove shoes. Anthropometrics, including
leg length, and maximal vertical reach (MVR) and maximal ver-
tical jump (MVJ) were assessed and recorded.
A trained laboratory member placed 10 Trigno Delsys
surface electromyography (sEMG) devices and 15 reflective
markers on participants. sEMG was placed and recorded on
both sides of the body for gastrocnemius (G), tibialis anterior
(T), rectus femoris (Q), semitendinosus (H), and erector spinae
(E). Reflective markers were placed on the following anatomical
landmarks: center of the forehead, base of the skull, cervical ver-
tebrae 7 (C7), and on both sides for the acromion, lateral epi-
condyle of the elbow, greater trochanter of the hip, lateral side of
the knee, lateral malleolus of the ankle, and the fifth metatarsal.
Participants were asked to land from a height-adjusted bar.
Height of the bar was calculated according to the following for-
mula: Bar height = (40% leg length + 40% MVJ + MVR). For
safety reasons, the individual bar height was tested for each child
using few assisted drops before data collection.
After the bar height was tested and deemed safe, the partic-
ipants completed 15 drop trials, where they were asked to land
onto 2 force plates (Kistler, 9286BA; one foot on each force plate)
and react to a randomized light cue. The light cue instructed the
child to stay in place (stay trial), or to run to the left or right (left
or right trials). Five trials for each condition were completed per
condition.
Surface Electromyography
Delsys EMG Works software was used to rectify and smooth
the sEMG data using a root mean square technique (window
length 0.1 seconds, window overlap 0.005 seconds). For each
sEMG channel, 2 independent coders manually identified the
onset/offset of muscle bursts (≥80 ms) during preimpact (750
ms before), impact, and postimpact (750 ms after). The onset
and offset times, the duration of the burst, and percentages of
trials that contained a muscle burst were calculated.
Kinematics
A Qualisys three-dimensional motion capture system was
used to collect kinematic data. Hip and knee angles at impact
and maximum flexion were calculated. The duration from
impact to maximum hip and knee flexion was calculated.
Statistical Analysis
Descriptive statistics were calculated for anthropometric,
sEMG, and kinematic variables using SPSS v.22. Independent
t tests were used to guide our exploration of the data but not
to determine statistical significance. Results with P values lower
than .05 are reported. Mean values across all trials for the sEMG
and kinematic data are presented and used to compare groups.
Pediatric Physical Therapy
Copyright © 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
This aggregated average was conducted since results were sim-
ilar regardless of drop condition (ie, left, right, and stay). Lastly,
interrater reliability for the sEMG data was calculated, achieving
96.2% agreement for muscle bursts and 91.8% agreement for
the timing of onset/offset.
RESULTS
Participants
Three children with ASD (4.18 ± 0.25 years; 66.6% male)
and 3 children with TD (3.63 ± 0.73 years; 66.6% male) had
differences between groups for average height (ASD 110.633 ±
2.72 cm, TD 99.13 ± 4.94 cm), weight (ASD 19.23 ± 0.93
kg, TD 14.8 ± 0.85 kg), and reach height (ASD 135.17 ± 2.75
cm, TD 120.27 ± 2.75 cm). There were no group differences
in average BMI (ASD 15.75 ± 1.43 kg/m2 , TD 15.08 ± 0.71
kg/m2 ), leg length (ASD 45.97 ± 2.50 cm, TD 41.30 ± 4.48
cm), bar height used (ASD 157.52 ± 11.39 cm, TD 139.20 ±
11.39 cm), or MVJ (ASD 145.07 ± 7.28 cm, TD 126.30 ± 11.48
cm) between groups.
Differences in sEMG
Differences were found in the duration and percentage of
sEMG burst. Participants with TD had longer burst during
impact for G (left and right), T (left), and H (left and right). The
children with ASD had longer burst during preimpact for E (left
and right) and G (right) (Figure). The Figure graphs the average
onset/offset times and the average length of muscle bursts during
preimpact (A) and impact for both groups (B). Postimpact data
had fewer differences between the groups for onset/offset time
and durations and therefore were not presented.
Children TD, on average, had a greater percentage of trials
that included an EMG burst in E (left [ASD 45.0% ± 21.0%, TD
69.8 ± 27.5%] and right [ASD 50.2% ± 19.9%, TD 69.3% ±
17.5%]), G (left [ASD 60.6% ± 19.6%, TD 79.4% ± 14.2%]),
and Q (left [ASD 43.4% ± 28.6%, TD 71.3% ± 26.2%]) at
postimpact; and H (right [ASD 94.3% ± 8.6%, TD 100% ±
0%]) and E (right [ASD 94.3% ± 8.4%, TD 100% ± 0%]) at
impact. Taken together, these results indicate that children with
ASD had longer muscle bursts prior to impact, whereas children
TD had longer and more muscle activation at and after impact.
Kinematics
No differences were found between the groups for the angle
of the knee and hip at impact or maximum flexion. However,
there were differences in the duration from impact to maximum
knee (ASD 1.09 ± 0.25 seconds, TD 0.99 ± 0.15 seconds) and
hip flexion (ASD 1.14 ± 0.2 seconds, TD 1.07 ± 0.11 seconds).
Children TD had shorter duration for both maximum knee and
hip flexion.
DISCUSSION
The results from this pilot study support that children TD
had more and longer sEMG bursts during impact compared with
children with ASD. Prior literature in children TD suggests that
sEMG Analysis During Landing in Children With ASD 193

Fig. sEMG onsets and offsets for all conditions during preimpact (A) and impact
(B). Length of the vertical bars indicates duration of the muscle burst. Horizontal
black continuous line denotes impact. Horizontal black dashed line denotes max
hip flexion for ASD. Horizontal gray dashed line denotes max hip flexion for TD.
The circled muscles indicate group differences. ASD indicates autism spectrum
disorder; E, erector spinae; G, gastrocnemius; H, semitendinosus; L, left; Q, rectus
femoris; R, right; T, tibialis anterior; TD, typical development.
longer bursts of muscle activation, during what we defined as
impact, represent a more mature landing strategy.9,10 In con-
trast, children with ASD may have longer preactivations com-
pared with children TD for some muscles. These results indi-
cate a less developed landing strategy in children with ASD,
as defined by the timing of muscle activation. Compared with
previous studies,3,4 our results are in agreement with a devel-
opmental delay in ASD, although our data included a younger
cohort.
The various landing conditions (stay, left, and right) had
similar results for both kinematics and sEMG data. This could
be due to the fact that the light cue was triggered as the landing
was initiated, and therefore, children may have performed this
task as 2 separate actions (land and stay, or land and run). A
more detailed analysis of the postimpact data including reaction
time, and stepping and attention strategies, where potential con-
dition differences may exist, might provide insight into group
differences in the decision-making process for our task.
Kinematic differences were found in the duration from
impact to maximum knee and hip flexion. Children TD showed
shorter duration to achieve maximum knee and hip flexion.
Paired with the results from the sEMG data, where most
extensor muscles (E, Q, and G) had greater percentage of acti-
194 Rosales et al
Copyright © 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
vation postimpact, children TD were exhibiting a more effi-
cient landing strategy from impact to maximum flexion and
postimpact. However, further study is needed to investigate this
claim.
There were no significant differences in maximum hip and
knee flexion. Lack of group differences may be explained by a
small sample size (low power) and/or large variability in the
kinematics. However, no angular amplitude differences in the
lower extremity may exist between these 2 populations during
landing. The only differences might be in the timing variables
and duration of muscle activation. If this is accurate, observation
of movement amplitude patterns in young children with ASD
may not elucidate underlying neuromuscular deficiency. Cau-
tion should be used in interpreting our results. Further inves-
tigation is needed using a larger sample size and more robust
kinematics.
Due to the limitations of this pilot study, we can only sug-
gest that differences in sEMG may exist between children with
and without ASD. If correct, our findings might have thera-
peutic relevance, suggesting that interventions for children with
ASD should target muscle activation timing to prevent poten-
tial deficits in landing, a skill widely used for all children during
recreational and structured physical activities.
ACKNOWLEDGMENTS
Thank you to all of the participants and their families for
being part of the study. Also, a special thank you to the labora-
tory interns in the laboratory that helped with the project.
REFERENCES
1. Centers for Disease Control and Prevention. Prevalence of autism spec-
trum disorder among children aged 8 years—autism and developmental
disabilities monitoring network, 11 sites, United States, 2010. MMWR
Surveill Summ. 2014;63:1-21.
2. America Psychiatric Association. Neurodevelopmental disorders. In:
Diagnostic and Statistical Manual of Mental Disorders: DSM-5. 5th ed.
Washington, DC: American Psychiatric Association Publishing; 2013:
50-59.
3. Whyatt CP, Craig CM. Motor skills in children aged 7-10 years, diag-
nosed with autism spectrum disorder. J Autism Dev Disord. 2012;42(9):
1799-1809.
4. Staples KL, Reid G. Fundamental movement skills and autism spectrum
disorders. J Autism Dev Disord. 2010;40(2):209-217.
5. Downey R, Rapport MJ. Motor activity in children with autism: a review
of current literature. Pediat Phys Ther. 2012;24(1):2-20.
6. Mohd MN, Jailani R, Tahir NM, et al. EMG signals analysis of BF and RF
muscles in autism spectrum disorder (ASD) during walking. Int J Adv
Sci Eng Inform Technol. 2016;6(5):793-798.
7. Lopes L, Santos R, Pereira B, Lopes VP. Associations between gross
motor coordination and academic achievement in elementary school
children. Hum Mov Sci. 2013;32(1):9-20.
8. Barnett LM, Van Beurden E, Morgan PJ, Brooks LO, Beard JR. Childhood
motor skill proficiency as a predictor of adolescent physical activity. J
Adolesc Health. 2009;44(3):252-259.
9. Christoforidou A, Patikas DA, Bassa E, et al. Landing from dif-
ferent heights: biomechanical and neuromuscular strategies in trained
gymnasts and untrained prepubescent girls. J Electromyogr Kinesiol.
2017;32:1-8.
10. Lazaridis S, Bassa E, Patikas D, Giakas G, Gollhofer A, Kotzamanidis C.
Neuromuscular differences between prepubescents boys and adult men
during drop jump. Eur J Appl Physiol, 2010;110(1):67-74.
Pediatric Physical Therapy