Examining the impact of physical activity on sleep quality and executive functions in children with autism spectrum disorder: A randomized controlled trial

Abstract

Sleep disturbance and executive dysfunction have been widely reported in children with autism spectrum disorder. While the positive impacts of physical activity on sleep quality and cognition are documented in children with typical development, similar studies in children with autism spectrum disorder are scarce. The objective of this study was to examine the impact of physical activity on sleep quality and cognition in children with autism spectrum disorder. A total of 40 children diagnosed with autism spectrum disorder (mean age = 9.95 years) were randomly assigned into two groups: physical activity intervention and control. Four sleep parameters (sleep efficiency, sleep onset latency, sleep duration, and wake after sleep onset) and two executive functions (inhibition control and working memory) were assessed. Results revealed a significant improvement in sleep efficiency, sleep onset latency, and sleep duration in the intervention group but not in the control group during weekdays. Moreover, a significant improvement in inhibitory control was shown in the intervention group but not in the control group. No significant improvement in working memory capacity was documented in either group (ps > 0.05). Our findings highlight the value of physical activity in improving sleep quality and cognition among children with autism spectrum disorder, but specific physical activity may be required to benefit individual executive functions.

Keywords

autism spectrum disorder children inhibition control physical activity sleep working memory

Introduction

Children with autism spectrum disorder (ASD) are characterized by deficits in social ability, abnormalities in communication, and repetitive stereotypic behavior (World Health Organization, 2016). In addition to these core symptoms, sleep disturbance is also commonly observed among children with ASD (Cortesi, Giannotti, Ivanenko, & Johnson, 2010; Liu, Hubbard, Fabes, & Adam, 2006). Several studies have reported that the prevalence of sleep disturbance is higher in children with ASD (40% to 80%) than in children with typical development (TD; 25% to 40%) (Calhoun, Fernandez-Mendoza, Vgontzas, Liao, & Bixler, 2014; Goldman, Richdale, Clemons, & Malow, 2012). The most frequently reported types of sleep disturbance include delayed sleep onset, difficulty in sleep maintenance, and insufficient sleep duration (SD; Cortesi et al., 2010; Richdale & Schreck, 2009; Souders et al., 2009). Research shows that poor sleep can exacerbate various symptoms of autism, resulting in increased stereotypic behavior (Mazurek & Sohl, 2016), more severe communication impairment (Taylor, Schreck, & Mulick, 2012), and intensified emotional control problems (Gregory & Sadeh, 2012). Meanwhile, given the importance of sleep on cognitive development, memory processes, and school performance in TD children (Curcio, Ferrara, & De Gennaro, 2006), it is reasonable to believe that consequences of sleep disturbance on cognitive functioning in children with ASD are also taxing. In a recent case-control study, Maski and colleagues (2016) indicated poor sleep could aggravate the impaired memory consolidation in children with ASD.

In fact, cognitive impairments are commonly found among children with ASD (Russo et al., 2007; Sachse et al., 2012). Executive functions generally refer to several interacting but potentially dissociable mental operations, including planning, working memory, inhibition, and flexibility (Ozonoff, 1995). Children with ASD are often reported to have poor working memory capacity (Barendse et al., 2013; Russell, Jarrold, & Henry, 1996), difficulties in switching attention between tasks (Reed, Watts, & Truzoli, 2013; Wallace et al., 2016), and inhibition response problems (Hopkins, Yuill, & Branigan, 2017), which can seriously affect their school performance and everyday functioning (Gilotty, Kenworthy, Sirian, Black, & Wagner, 2002). Meanwhile, it is well-documented that sleep disturbance and poor cognitive functioning are closely related to each other (see Dewald, Meijer, Oort, Kerkhof, & Bogels, 2010, for review). With the long-term negative interaction effects posed by sleep disturbance and executive dysfunction, developing an effective intervention that can directly address both cognitive deficits and sleep disturbance is critical. Physical activity is one type of intervention that may be useful in this context.

Over decades, a number of studies have reported compelling evidence for the benefits of physical activity on sleep quality (see Kredlow, Capozzoli, Hearon, Calkins, & Otto, 2015 for review) and executive functions (Álvarez-Bueno et al., 2017). For example, Kredlow and colleagues (2015) examined the effects of acute and regular exercise on sleep. The results revealed that both acute and regular exercise could increase total sleep time, shorten sleep onset latency (SOL), reduce rapid eye movement, and promote slow wave sleep (Kredlow et al., 2015). Álvarez-Bueno and colleagues recently conducted a meta-analysis of 36 interventional studies examining the effect of physical activity on executive functions and metacognition. The design of these studies included randomized controlled trials (RCTs), non-RCTs, and controlled ore-post studies. The physical activity intervention aimed to increase the physical activity participation time or to increase the cognitive demands of physical activity tasks for healthy children aged 4 to 18 years. The outcomes included cognitive performance assessments of nonexecutive cognitive functions, core executive functions, and/or metacognition (Álvarez-Bueno et al., 2017). Overall, the findings suggested that physical activity benefits multiple facets of executive function (e.g. inhibition control and working memory capacity; Álvarez-Bueno et al., 2017). Based on these findings in children with TD, the use of physical activity as an alternative intervention to improve sleep quality and executive functions in children with ASD appears to be promising.

In the present study, we conducted an RCT to investigate the impact of physical activity intervention (12-week basketball skill learning) on sleep quality and two executive functions (inhibition and working memory) in children with ASD. Basketball skill learning intervention was chosen because it was a natural cognitive-engaging physical activity requiring learner’s attention and working memory on motor coordination, which is shown to be beneficial to both physical and cognitive developments in children with TD (Diamond & Lee, 2011). We hypothesized that the intervention would positively impact both sleep quality and executive functions in children with ASD.

Method

Participants

In previous related studies (Anderson-Hanley, Tureck, & Schneiderman, 2011; Brand, Jossen, Holsboer-Trachsler, Pühse, & Gerber, 2015; Pan et al., 2017; Wachob & Lorenzi, 2015), effect sizes on sleep quality ranged from d = 0.13 (Brand et al., 2015) to d = 1.69 (Anderson-Hanley et al., 2011). The mean value of the effect size (d = 0.95) was used to calculate the sample size required in the present study. With a 5% level of significance, a sample size of 36 (18 per group) was needed to achieve a power of 90% for interaction effect using G*Power 3.1.9.2 software (Faul, Erdfelder, Lang, & Buchner, 2007). The study was approved by the ethics committee of the appropriate university. The inclusion criteria were as follows: (1) aged 8–12 years; (2) ASD diagnosis from a physician based on the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5, American Psychiatric Association, 2016); (3) non-verbal IQ above 40 as assessed by the Wechsler Intelligence Scale for Children (Chinese revised; C-WISC; see Gong & Cai, 1993 for more information); (4) ability to follow instructions; (5) ability to perform the requested physical intervention and executive function measures; (6) no prior experience in formal basketball skill training (no formal knowledge about how to perform basketball shooting); and (7) no history of reading disabilities according to parents. The exclusion criteria were as follows: (1) having other medical conditions that limited their physical activity capacities (e.g. asthma, seizure, cardiac disease) and (2) having a complex neurologic disorder (e.g. epilepsy, phenylketonuria, fragile X syndrome, tuberous sclerosis). All the screenings were conducted by trained research assistants at different schools before the first class. After screening, a total of 50 participants (40 boys and 10 girls) from four local special schools were successfully recruited. Participants were randomly assigned to one of two groups: intervention (n = 25) and control (n = 25). To ensure equal allocation ratios for both groups, block randomization (Efird, 2011) was used. A block size of five was used in the study (i.e. five participants in one group and five in the other group for every 10 consecutively entered participants). The block randomization process was performed by a trained research assistant. Of the 50 participants, 10 were excluded (six from the intervention group and four from the control group) because of incomplete information in the go/no-go (GNG) task (see the Measures section), resulting in 19 participants in the intervention group (14 boys and five girls; mean age = 10.11 ± 1.20) and 21 participants in the control group (18 boys and three girls; mean age = 9.81 ± 1.17). We collected information about participants’ autism symptoms by asking their parents to complete the Social Responsiveness Scale (Second Edition; SRS-2; Constantino & Gruber, 2012). We also collected information about current treatments (e.g. medications, speech therapy, occupational therapy) from the participants’ parents. Six participants were reported to take regular medication treatment during the screening. The medications were hydrocortisone (n = 2), fusidic acid cream (n = 3), and risperidone (n = 1). The side effects of these medications were non-significant according to the parents; therefore, the participants were allowed to participate in the present study. Written consent was obtained from participants’ parents/guardians. The study was approved by the university ethics committee. Demographic data for the two groups are shown in Table 1.

Table 1.

Demographic statistics of participants of each group.

	Intervention group (n = 19)	Control group (n = 21)	p
Gender	14 boys and 5 girls	18 boys and 3 girls	0.36
Age (years)	10.11 ± 1.20	9.81 ± 1.17	0.43
Weight (kg)	36.48 ± 4.02	35.14 ± 4.52	0.34
Height (m)	1.35 ± 0.08	1.37 ± 0.08	0.67
BMI (kg/m²)	19.86 ± 2.71	18.76 ± 2.27	0.17
Non-verbal IQ	64.72 ± 19.00	67.51 ± 15.01	0.09
SRS-2 Row-scores	75.94 ± 10.92	77.19 ± 11.05	0.72
Medication (n)
Yes	2	4
No	17	17

BMI: body mass index; SRS-2: Social Responsiveness Scale (Second Edition).

Intervention group

A 12-week basketball skill learning intervention consisting of 24 sessions (two sessions per week; 45 min per session) was implemented in a hall/gymnasium of each participating school in the morning. The staff-to-participant ratio for both groups ranged from 1:3 to 1:2 depending on the attendance. The overall attendance rate was 97.79%. Each intervention session was conducted in an identical format, comprising three activities: warm-up (10 min), basketball (30 min), and cool-down (5 min). In the basketball activity, participants were partnered with a staff member to learn different basketball skills (see Supplemental material for more information). To motivate participants, they were positively reinforced verbally with compliments for their efforts in learning and practicing the skill.

Control group

Participants in the control group received no basketball intervention and were asked to follow their normal daily routine without participating in any additional physical activity/exercise program between the two measurements (T1–T2).

Procedure

Each participant attended two 1-week assessments, in which their habitual sleeping patterns and executive functions (i.e. inhibition control and working memory capacities) were assessed. The two 1-week assessments were conducted before the intervention (T1: baseline) and immediately after the 12 weeks of physical activity intervention or regular treatment (T2: post-intervention). Presentations of the two cognitive assessments were counterbalanced across participants. The procedure of the present study is shown in Figure 1.

Figure 1.

Flowchart.

Measures

Sleep quality

Four sleep parameters including sleep efficiency (SE, actual sleep time divided by time in bed, expressed as a percentage), SOL (length of time to fall asleep, expressed in minutes), SD (expressed in hours), and wake after sleep onset (WASO, length of wake time after falling to sleep, expressed in minutes) were measured with an actigraphy accelerometer (Model no: ActiGraph GT3X; Wachob & Lorenzi, 2015) and a sleep log (Morgan, David, & Gasoigne, 2007). These parameters were chosen because they were previously found to be the most representative measures of sleep behaviors and sleep quality in children with ASD (Malow et al., 2006; Souders et al., 2017). For the actigraphy assessment, participants were asked to wear the device on the non-dominant wrist for seven consecutive days (Monday to Sunday) and were instructed to only take it off when they were taking baths. Non-wear time was defined as 60 min of consecutive 0 scores with a 2 min spike tolerance (Wachob & Lorenzi, 2015). The night (2200–0700) was considered invalid if the wear time was less than 8 h and was excluded from data analysis. The Sadeh algorithm (Sadeh, Sharkey, & Carskadon, 1994) was implemented to identify sleep onset and sleep offset. For the sleep log assessment, participants were assisted by their parents to recall specific sleep patterns (e.g. bedtime, sleep start, sleep end, wake-up time, and sleep length) and reported them in a sleep log for the whole assessment duration (T1 and T2).

Executive functions

Two executive functions, inhibition control and working memory, were examined. These two functions were selected because they are widely considered to constitute the core components of executive functions (Barendse et al., 2013; Verté, Geurts, Roeyers, Oosterlaan, & Sergeant, 2006). Importantly, impairments of these two executive functions are relatively common in children with ASD (Andersen, Hovik, Skogli, Egeland, & Øie, 2013; Xiao et al., 2012).

For inhibition control, participants’ ability to inhibit unwanted responses to changing stimuli were measured using a computer-based version of the GNG task. The GNG task was identical to that used by Uzefovsky, Allison, Smith, and Baron-Cohen (2016). Participants were seated in front of a computer monitor with a keyboard. They were asked to press left or right arrow key as quickly as possible when the corresponding arrow appeared on the center of the screen (Go response), and to not press any key whenever the up arrow appeared on the screen (No-go response). Following 20 practice trials, participants completed 300 trials: 220 trials requiring a Go response (110 left and 110 right) and 80 trials (26.7%) requiring a No-go response (not pressing any key) (Uzefovsky et al., 2016). The stimuli were randomly presented, one at a time, for 500 ms followed by 1000 ms of blank interval using E-Prime 3.0 software (Psychology Software Tools, Inc, 2012). At an interval of 60 trials, children were offered a break of 2 min. No feedback was given upon response and the response time was recorded but not analyzed because of the unreliability of the recording procedure (Uzefovsky et al., 2016). Moreover, participants were excluded if they failed to respond correctly to at least 50% of the Go trials (10 children were excluded, as described above). As in Uzefovsky et al.’s (2016) study, a Go response in a No-go trial was coded as a false alarm (FA). FA errors are considered an indicator of inhibition control (Wright, Lipszyc, Dupuis, Thayapararajah, & Schachar, 2014).

For working memory, participants were asked to perform the Corsi block tapping task (CBTT, Corsi, 1972), forward digit span test (FDS), and backward digit span test (BDS) (Thorndike, Hagen, & Sattler, 1986) respectively to measure the capacities of visual-spatial working memory and auditory working memory. In the CBTT, participants were required to observe the sequence of blocks being “tapped,” then repeat the sequence back in order. The initial sequence length was three blocks, and after every four trials the sequence length increased by one. The task ended after four incorrectly repeated sequences of the same length, and the longest correctly repeated sequence length was recorded (de Vries, Prins, Schmand, & Geurts, 2015). For the digit span tests, digits were presented at a rate of one digit per second, and participants were required to repeat the sequence verbally. The initial sequence length was two digits, and the sequence length increased by one after every two trials. The test ended after two incorrectly repeated sequences of the same length and the maximum digit span was established. All the cognitive assessments were conducted in the morning before the school commencement by two research assistants who were blinded to the grouping.

Data analysis

All statistical analyses were conducted using SPSS (version 23.0) for Windows (SPSS Inc., Chicago, IL, USA). All data were entered into SPSS by a research assistant, who was blinded to the grouping. To assess the effects of the intervention on sleep, linear mixed model (LMM) was performed for each sleep parameter measured by sleep log and actigraph accelerometer. Actigraphy-assessing parameters were averaged across weekday nights (Monday to Friday) and weekend nights (Saturday and Sunday) for each participant. Spearman’s correlations were used to determine the relationship between the actigraphy assessment and sleep log on two sleep parameters (SE and SD) to check the validity of the sleep log. SOL and WASO were not used for validity check because these two sleep parameters could not be measured simultaneously by actigraphy assessment and sleep log. Meanwhile, to assess the effects of the intervention on executive functions, a series of 2 (block: pre vs post) × 2 (group: intervention vs control) analyses of covariance (ANCOVAs) with repeated measures was performed for each executive function outcome. Considering the potential confounding effects of developmental factors, age was controlled as a covariate for the raw scores of the executive function outcomes (i.e. CBTT score, FDS and BDS test scores, and GNG task accuracy) to compare the changes between and within groups over different time points. Finally, Spearman’s correlations were used to determine the relationship between changes in sleep parameters and changes in executive functions. A p value < 0.05 was considered statistically significant. Results are presented as mean (standard deviation). Preliminary tests of the assumptions of the statistical tests, including data normality using Shapiro–Wilk tests (all p values > 0.05), homogeneity of variance (Levene’s tests: all p values > 0.05), and homogeneity of regression slopes (all p values > 0.05) for the ANCOVAs were met.

Results

Sleep parameters

At T1, all the sleep parameters measured by actigraphy assessment and sleep logs were comparable between groups during weekday and weekend (see Tables 2 and 3).

Table 2.

Sleep parameters during weekday and weekend sleeps measured by actigraphy assessment.

Sleep parameters	Intervention group (SE)	Control group (SE)	p value (group effect)	p value (interaction effect)
Weekday
Sleep efficiency (%)				<0.001
T1	87.17 (1.69)	90.73 (1.67)	0.14
T2	96.33 (0.71)	83.90 (2.38)	<0.001
p value (time effect)	<0.001	<0.001
Cohen’s d effect size (95% CI)	0.77 (0.46, 1.08)	0.59 (0.28, 0.90)
Wake after sleep onset (min)				<0.001
T1	57.97 (8.60)	43.22 (8.19)	0.21
T2	17.00 (3.52)	81.62 (13.54)	<0.001
p value (time effect)	<0.001	<0.001
Cohen’s d effect size (95% CI)	0.66 (0.35, 0.97)	0.68 (0.37, 0.99)
Sleep duration (hour)				0.008
T1	6.79 (0.78)	6.82 (0.22)	0.94
T2	7.21 (0.16)	6.54 (0.19)	0.008
p value (time effect)	0.04	0.03
Cohen’s d effect size (95% CI)	0.29 (–0.02, 0.60)	0.16 (–0.15, 0.48)
Weekend
Sleep efficiency (%)				<0.001
T1	80.90 (1.93)	86.09 (2.27)	0.05
T2	94.46 (0.95)	87.20 (1.80)	<0.001
p value (time effect)	0.002	0.64
Cohen’s d effect size (95% CI)	0.16 (–0.16, 0.47)	0.07 (–0.23, 0.38)
Wake after sleep onset (min)				0.002
T1	99.84 (12.51)	99.41 (5.41)	0.06
T2	64.23 (13.48)	63.95 (8.17)	0.001
p value (time effect)	<0.001	0.99
Cohen’s d effect size (95% CI)	1.12 (0.81, 1.43)	0.84 (0.53, 1.15)
Sleep duration (hour)				0.18
T1	7.38 (0.27)	7.71 (0.33)	0.43
T2	8.59 (0.25)	7.48 (0.28)	0.03
p value (time effect)	0.002	0.63
Cohen’s d effect size (95% CI)	0.003 (–0.31, 0.31)	0.008 (–0.30, 0.31)

CI: confidence interval.

Table 3.

Sleep parameters of weekday and weekend sleep patterns measured by sleep log.

Sleep parameters	Intervention group (SE)	Control group (SE)	p value (group effect)	p value (interaction effect)
Weekday
Sleep efficiency (%)				0.001
T1	93.34 (0.60)	92.78 (0.54)	0.48
T2	95.77 (0.42)	92.79 (0.54)	<0.001
p value (time effect)	<0.001	0.97
Cohen’s d effect size (95% CI)	0.61 (0.30, 0.92)	0.002 (–0.3, 0.31)
Sleep onset latency (min)				0.001
T1	33.42 (3.14)	36.24 (2.61)	0.50
T2	21.57 (2.23)	37.19 (2.3)	<0.001
p value (time effect)	<0.001	0.57
Cohen’s d effect size (95% CI)	0.62 (0.31, 0.93)	0.04 (–0.27, 0.35)
Sleep duration (hour)				0.30
T1	7.78 (0.19)	7.81 (0.18)	0.93
T2	8.19 (0.22)	8.00 (0.18)	0.54
p value (time effect)	0.02	0.04
Cohen’s d effect size (95% CI)	0.38 (0.07, 0.69)	0.11 (–0.20,0.42)
Weekend
Sleep efficiency (%)				0.001
T1	94.63 (0.56)	94.00 (0.43)	0.37
T2	96.51 (0.37)	92.91 (0.46)	<0.001
p value (time effect)	<0.001	0.003
Cohen’s d effect size (95% CI)	0.53 (0.22, 0.84)	0.30 (–0.01, 0.61)
Sleep onset latency (min)				0.001
T1	33.63 (3.40)	35.62 (2.91)	0.50
T2	20.55 (2.34)	38.82 (2.27)	<0.001
p value (time effect)	0.001	0.1
Cohen’s d effect size (95% CI)	0.55 (0.24, 0.86)	0.14 (–0.17, 0.45)
Sleep duration (hour)				0.18
T1	9.67 (0.20)	9.46 (0.24)	0.17
T2	9.27 (0.22)	8.06 (0.22)	0.008
p value (time effect)	0.39	0.004
Cohen’s d effect size (95% CI)	0.006 (–0.30, 0.32)	0.005 (–0.31, 0.31)

CI: confidence interval.

Actigraphy assessments

As shown in Table 2, the LMM showed that there was significant group by time interaction effects for both SE and WASO (all ps < 0.001) during weekday and weekend. The intervention group showed significant improvements for both sleep parameters (all ps < 0.001) with moderate effect sizes (d ranged from 0.66 to 0.77) from T1 to T2. In contrast, the control group showed significant drop for SE and increase in WASO (all ps < 0.001) with moderate effect sizes (d ranged from 0.59 to 0.68) from T1 to T2. Meanwhile, the LMM indicated that there was a significant interaction effect for SD (p < 0.01) during weekday but not during weekend (p > 0.05). Both intervention group and control group showed significant improvements for SD (all ps < 0.001) with small effect sizes (d ranged 0.16 to 0.29) from T1 to T2 during weekday. However, only the intervention group showed a significant improvement for SD (p < 0.01) with small effect size (d = 0.003) from T1 to T2 while there was not significant change in control group (p > 0.05).

Sleep log

As shown in Table 3, the LMM showed that there was significant group by time interaction effects for both SE and SOL (all ps < 0.001) during weekday and weekend. The intervention group showed significant improvements for both sleep parameters (all ps < 0.001) with moderate effect sizes (d ranged from 0.61 to 0.62) from T1 to T2. In contrast, the control group showed no significant change for both parameters (all p > 0.05) with small effect sizes (d ranged from 0.002 to 0.04) from T1 to T2. Meanwhile, the LMM indicated no significant interaction effects for SD (ps > 0.05) during weekday and weekend. Both intervention group and control group showed significant improvements for SD (all ps < 0.05) with small to moderate effect sizes (d ranged 0.11 to 0.38) from T1 to T2 during weekday. No significant change of SD was shown in the intervention group (p > 0.05) from T1 to T2 during weekend. The control group, interestingly, showed a significant decrease in SD (p < 0.01) with small effect size (d = 0.005) from T1 to T2 during weekend. Comparisons of sleep parameters between groups and within groups at different timeslots by different measurements are shown in Tables 2 and 3.

Inhibition control and working memory capacity

At T1, inhibition control and working memory capacities were comparable between groups, as reflected by the GNG computer task and the three working memory tasks (i.e. the CBTT, FDS test, and BDS tests; see Table 4). Repeated-measures ANCOVAs were performed separately for each neuropsychological test to examine the effects of the physical activity intervention on executive functions after controlling for age. After adjusting for the effect of age, we found a significant interaction effect (F(1, 37) = 6.58, p = 0.02) and a significant group effect (F(1, 37) = 4.15, p = 0.04) in FA error. However, no significant interaction effects were revealed in CBTT score (F(1, 37) = 1.00, p = 0.32), FDS score (F(1, 37) = 2.21, p = 0.15), or BDS score (F(1, 37) = 0.77, p = 0.39). Subsequent tests revealed that the FA errors of the intervention group (M = 10.89, SD = 5.17) were significantly smaller than those of the control group (M = 18.38, SD = 8.15) at T2 (t(38) = −3.43, p = 0.001), and there was a significant reduction of FA errors between T1 and T2 in the intervention group (t(18) = 2.55, p = 0.02) but not in the control group (t(20) = −0.37, p = 0.72). Comparisons of neuropsychological measures between groups and within groups at different timeslots are shown in Table 4.

Table 4.

Comparisons of neuropsychological measures.

Neuropsychological assessment	Intervention group (SD)	Control group (SD)	p value (group effect)	p value (interaction effect)
Inhibition control (false alarm (FA) error)				0.02
T1	16.16 (8.31)	18.05 (7.80)	0.46
T2	10.89 (5.17)	18.38 (8.15)	<0.001
p value (time effect)	0.02	0.72
Corsi block tapping task (CBTT)				0.80
T1	4.32 (1.38)	3.86 (1.15)	0.26
T2	3.95 (1.27)	3.57 (0.93)	0.29
p value (time effect)	0.09	0.25
Forward digit span (FDS)				0.54
T1	6.05 (1.47)	5.33 (1.32)	0.11
T2	5.58 (1.30)	5.14 (1.39)	0.31
p value (time effect)	0.17	0.56
Backward digit span (BDS)				0.57
T1	4.32 (1.25)	3.67 (0.86)	0.06
T2	3.79 (0.98)	3.95 (0.92)	0.59
p value (time effect)	0.06	0.38

Correlations between actigraphy assessment and sleep log

To check the validity of the sleep log, the average values of the sleep parameters (i.e. SE and SD) across the whole week were taken. At T1, the correlation between the two assessments on SE and SD were 0.18 (p = 0.04) and 0.33 (p = 0.03), respectively. At T2, the correlations between the two assessments on SE and SD were 0.27 (p = 0.03) and 0.22 (p = 0.04), respectively.

Correlations between changes in sleep parameters and changes in executive functions

Changes in SE during weekend as measured by both assessments (i.e. actigraph and sleep log) were significantly correlated with changes in FA error. Moderate correlations were revealed between the changes in these two parameters (i.e. SE and FA error) during weekday by actigraphy assessment (r = –0.43, p = 0.006) and sleep log (r = –0.34, p = 0.03). Moreover, moderate correlations were found between changes in SOL and FA error during weekday (r = –0.33, p = 0.04) and during weekend (r = –0.47, p = 0.002). No significant correlation was found among changes in other sleep parameters and changes in other executive functions during weekday and weekend (all ps > 0.05).

Discussion

The present study examined the impact of physical activity intervention (a 12-week basketball skill learning) on sleep quality and executive functions in children with ASD. We hypothesized that the intervention would benefit both sleep quality and executive functions among the children. In accordance with this hypothesis, the intervention was found to be effective for improving SE, shortening SOL, increasing SD, reducing WASO, and enhancing inhibition control. However, it did not have any effect on working memory, and participants in the intervention and control groups did not differ in their performance on working memory assessments.

The sleep-related findings of our study are consistent with a recent small-scale study conducted by Brand and colleagues (2015), in which a physical activity intervention (30 min sessions of bicycle workout plus 30 min balance and coordination training) was shown to be effective for increasing SE and shortening SOL in 10 children with ASD (Brand et al., 2015). In the present study, participants in the intervention group exhibited greater SE, shorter SOL, longer SD, and reduced WASO than those in the control group during both weekdays and weekend (see Tables 2 and 3 for more information). One possible explanation for these findings is related to the ability of physical activity to alter melatonin levels (Lee, Kim, & Kim, 2014). Melatonin is a natural hormone serving as a key regulator of the circadian rhythm and promotes sleep onset and sleep maintenance (Wirojanan et al., 2009). Secretion of melatonin normally increases shortly after dark, peaks in the middle of the night, and falls slowly during early morning hours (Rossignol & Frye, 2011). This hormonal response allows for maintaining a normal circadian rhythm and sleeping through the night. Compared with children with TD, melatonin levels were found to be lower in some children with ASD (Krakowiak, Goodlin-Jones, Hertz-Picciotto, Croen, & Hansen, 2008; Liu et al., 2006). To counter this melatonin deficit, supplemental melatonin is commonly used to treat insomnia in children with ASD (Arendt, 2003). Previously, researchers suggested that melatonin levels could also be moderated by physical activity (Marrin et al., 2011). In one experiment, Marrin et al. (2011) asked seven healthy participants to complete a moderately intense cycling exercise in the morning, measuring their salivary melatonin concentration at baseline, during exercise, after exercise, and upon exercise recovery. The results revealed that participants’ melatonin levels were significantly increased during and after exercise compared with those at baseline and recovery. In the current study, the intervention might have increased participants’ melatonin levels, resulting in better sleep quality among them. Apart from melatonin-mediated factor, other factors such as reduced anxiety (DiLorenzo et al., 1999), improved mood (Hartescu, Morgan, & Stevinson, 2015) and increased brain-derived neurotrophic factor level (Szuhany, Bugatti, & Otto, 2015) may also mediate the mechanism of exercise–sleep relation. As a result, future studies examining the underlying mechanism of how physical activity affects sleep quality in children with ASD are therefore warranted. Meanwhile, it should be noted that we found moderate correlations for most of the sleep parameters measured by actigraphy assessment and sleep logs, suggesting that both assessment methods could be used interchangeably to measure the habitual sleep patterns of children with ASD.

For cognitive measurements, the improvement in inhibition control shown by the intervention group further strengthens the existing evidence for the cognitive benefits of physical activity reported in previous studies (e.g. Anderson-Hanley et al., 2011; Pan et al., 2017). For example, Pan and colleagues (2017) recently used a 12-week table tennis intervention combined with executive function training for 22 children with ASD to examine the impact of such an intervention on motor skill proficiency and executive functions. The results revealed significant improvements on both motor skill proficiency and executive functions in the participants after the intervention (Pan et al., 2017). A possible explanation for the cognitive benefit is related to the neurotrophic hypothesis (Khan & Hillman, 2014). According to this hypothesis, physical activity increases metabolic demands and triggers a cascade of biochemical changes, such as enhancing cerebral blood flow and increasing the availability of brain-derived neurotrophic factor (Khan & Hillman, 2014), which in turn strengthens brain plasticity for higher level cognitive activities such as those involved in executive functions (Gomez-Pinilla & Hillman, 2013). However, this hypothesis cannot explain the present findings regarding working memory, which was unaffected by the physical activity intervention. Therefore, rather than focusing purely on physiological factors, a psychological mediating mechanism should also be considered. In this context, a self-regulation theory has been proposed (Audiffren & André, 2015).

Self-regulation or self-control refers to any effort by a human to alter one’s own affective or behavioral responses according to environmental demands (Audiffren & André, 2015; McEwan, Ginis, & Bray, 2013). Under this theoretical framework, physical activity can enhance self-regulation if it is mentally effortful and attention-demanding (Lakes & Hoyt, 2004; Newell, 1991; Pesce et al., 2016; Zimmerman & Kitsantas, 1997). As such, the process of motor skill acquisition is thought to enhance self-regulation. According to the three stages of the learning model proposed by Fitts and Posner (1967), learning any new motor skill typically progresses through cognitive, associative, and autonomous stages (Fitts and Posner, 1967). The cognitive stage is characterized by the peak demand on cognitive resources, where high levels of consciousness and inhibition control are required to produce a specific sequence of actions to achieve the desired goal (see Anderson et al., 2013, for a review). In the present study, the basketball skill learning intervention might have enhanced participants’ self-regulation by requiring them to be highly attentive to regulating their body movements during exercise. Throughout this process, participants’ inhibition control was highly engaged as they had to inhibit any impulses to motor movements other than the movements they were asked to perform. As a result, the inhibition control ability of the participants in the intervention group may have been enhanced. In contrast, participants’ working memory may not have been strongly engaged because of the simplicity of the instructions and the physical guidance from the staff (for practical reasons), resulting in a lack of improvement in participants’ working memory capacity. Since the present study did not incorporate any of the measurements of self-regulation or self-control applied in other studies (Lakes, 2013; Lakes & Hoyt, 2004), we are unable to verify the explanation of our cognition-related findings by self-regulatory mechanisms. Future studies should incorporate such measurements to further elucidate this issue.

The sleep-cognition relation was further analyzed with specific correlations between changes in sleep parameters and changes in executive functions from T1 to T2. Our data revealed significant correlations of the changes in two sleep parameters: SE and SOL with the changes in inhibition control, whereas no significant correlation was found between changes in any sleep parameters and changes in any working memory measurements. This may suggest that the improvement of SE and sleep onset may play a role on improving the inhibition control. However, due to the limitation of our data (that we only have measured at two time points), the causal pathway between exercise, sleep, and executive function could not be examined. Future studies should incorporate the third time point (e.g. mid-progress) to further examine this causal relationship.

Despite the strengths of this study and its findings, there is an important limitation that requires attention The researchers did not examine the baseline physical activity level (e.g. number of hours of daily physical exercise within a 1-week period) and the basketball skill level from each participant prior to randomization and/or as exclusion/inclusion criteria to ensure that children in both groups maintained similar physical activity and basketball skill level per day. Without this control, it is difficult for the researchers to determine treatment effects. For example, children in the treatment condition may naturally engage in physical activity at a higher daily frequency as compared with children in the waitlist condition prior to the basketball intervention, or vice versa. Future study should include such assessment in baseline to address this question. Finally, this RCT protocol was not published in advance, future similar study should consider publishing the protocol first in order to strengthen the validity and reliability of the study.

Conclusion

The positive influences of physical activity on sleep quality and inhibition control revealed in our study are in accordance with the accumulating evidence of the cognitive benefits of physical activity in children with ASD. Practically, the current findings shed light on ways to improve sleep and cognitive outcomes among children. Physicians and educators may consider prescribing physical activity to children with ASD to alleviate their symptoms of sleep disturbance and executive dysfunction. Theoretically, our findings are important for guiding future researchers to further examine the exercise–sleep and exercise–cognition relationships in children with ASD through melatonin-mediated mechanisms and self-regulatory mechanisms.

Supplemental Material

AUT823910_Lay_Abstract – Supplemental material for Examining the impact of physical activity on sleep quality and executive functions in children with autism spectrum disorder: A randomized controlled trial

Supplemental material, AUT823910_Lay_Abstract for Examining the impact of physical activity on sleep quality and executive functions in children with autism spectrum disorder: A randomized controlled trial by Choi Yeung Andy Tse, Hong Paul Lee, Ka Shing Kevin Chan, Veronica Bordes Edgar, Alison Wilkinson-Smith and Wing Him Elvis Lai in Autism

Supplemental Material

AUT823910_Supplemental_material – Supplemental material for Examining the impact of physical activity on sleep quality and executive functions in children with autism spectrum disorder: A randomized controlled trial

Supplemental material, AUT823910_Supplemental_material for Examining the impact of physical activity on sleep quality and executive functions in children with autism spectrum disorder: A randomized controlled trial by Choi Yeung Andy Tse, Hong Paul Lee, Ka Shing Kevin Chan, Boades Veronica Edgar, Alison Wilkinson-Smith and Wing Him Elvis Lai in Autism

Footnotes

Acknowledgements

The authors would like to thank staff members Mr. Daniel Ling and Ms. Joyce Chan for their assistance in data collection, and student helpers Mr. Lau Man Kit, Mr. Lee Chin Hung, Miss Liu Hok Ling, Mr. Ng Cheuk Sam, Mr. Ng Ping Fun, and Mr. So Fung Yeuk for carrying out the physical activity intervention. The authors would also like to express their gratitude to all the children who participated in this study, and the participants’ teachers and parents for their support. The authors would also like to thank Benjamin Knight, MSc, from Edanz Group () for editing a draft of this manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work described in this article was partially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. EdUHK 28602517).

Supplemental material

Supplemental material for this article is available online.

ORCID iDs

Choi Yeung Andy Tse

Hong Paul Lee

References

Álvarez-Bueno

Pesce

Cavero-Redondo

Sánchez-López

Martínez-Hortelano

J. A.

Martínez-Vizcaíno

(2017). The effect of physical activity interventions on children’s cognition and metacognition: A systematic review and meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry, 56, 729–738.

American Psychiatric Association. (2016). Diagnostic and statistical manual of mental disorders: DSM-5. Washington, DC: Author.

Andersen

P. N.

Hovik

K. T.

Skogli

E. W.

Egeland

Øie

(2013). Symptoms of ADHD in children with high-functioning autism are related to impaired verbal working memory and verbal delayed recall. PLoS ONE, 8, e64842.

Anderson

D. I.

Campos

J. J.

Witherington

D. C.

Dahl

Rivera

. . . Barbu-Roth

(2013). The role of locomotion in psychological development. Frontiers in Psychology, 4, 440.

Anderson-Hanley

Tureck

Schneiderman

R. L.

(2011). Autism and exergaming: Effects on repetitive behaviors and cognition. Psychology Research and Behavior Management, 4, 129–137.

Arendt

(2003). Importance and relevance of melatonin to human biological rhythms. Journal of Neuroendocrinology, 15, 427–431.

Audiffren

André

(2015). The strength model of self-control revisited: Linking acute and chronic effects of exercise on executive functions. Journal of Sport and Health Science, 4, 30–46.

Barendse

E. M.

Hendriks

M. P.

Jansen

J. F.

Backes

W. H.

Hofman

P. A.

Thoonen

. . . Aldenkamp

A. P.

(2013). Working memory deficits in high-functioning adolescents with autism spectrum disorders: Neuropsychological and neuroimaging correlates. Journal of Neurodevelopmental Disorders, 5, 14.

Brand

Jossen

Holsboer-Trachsler

Pühse

Gerber

(2015). Impact of aerobic exercise on sleep and motor skills in children with autism spectrum disorders—A pilot study. Neuropsychiatric Disease and Treatment, 11, 1911–1920.

10.

Calhoun

S. L.

Fernandez-Mendoza

Vgontzas

A. N.

Liao

Bixler

E. O.

(2014). Prevalence of insomnia symptoms in a general population sample of young children and preadolescents: Gender effects. Sleep Medicine, 15, 91–95.

11.

Corsi

P. M.

(1972). Human memory and the medial temporal region of the brain. Dissertation Abstracts International, 34, 819B.

12.

Cortesi

Giannotti

Ivanenko

Johnson

(2010). Sleep in children with autistic spectrum disorder. Sleep Medicine, 11, 659–664.

13.

Curcio

Ferrara

De Gennaro

(2006). Sleep loss, learning capacity and academic performance. Sleep Medicine Reviews, 10, 327–337.

14.

de Vries

Prins

P. J. M.

Schmand

B. A.

Geurts

H. M.

(2015). Working memory and cognitive flexibility-training for children with an autism spectrum disorder: A randomized controlled trial. Journal of Child Psychology and Psychiatry, 56, 566–576.

15.

Dewald

J. F.

Meijer

A. M.

Oort

F. J.

Kerkhof

G. A.

Bogels

S. M.

(2010). The influence of sleep quality, sleep duration and sleepiness on school performance in children and adolescents: A meta-analytic review. Sleep Medicine Review, 14, 179–189.

16.

Diamond

Lee

(2011). Interventions shown to aid executive function development in children 4 to 12 years old. Science, 333, 959–964.

17.

DiLorenzo

T. M.

Bargman

E. P.

Stucky-Ropp

Brassington

G. S.

Frensch

P. A.

LaFontaine

(1999). Long-term effects of aerobic exercise on psychological outcomes. Preventive Medicine, 28, 75–85.

18.

Efird

(2011). Blocked randomization with randomly selected block sizes. International Journal of Environmental Research and Public Health, 8, 15–20.

19.

Faul

Erdfelder

Lang

A.-G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191.

20.

Fitts

P. M.

Posner

M. I.

(1967). Human performance. Belmont, CA: Brooks/Cole Publishing.

21.

Gilotty

Kenworthy

Sirian

Black

D. O.

Wagner

A. E.

(2002). Adaptive skills and executive function in autism spectrum disorders. Child Neuropsychology, 8, 241–248.

22.

Goldman

S. E.

Richdale

A. L.

Clemons

Malow

B. A.

(2012). Parental sleep concerns in autism spectrum disorders: Variations from childhood to adolescence. Journal of Autism Developmental Disorder, 42, 531–538.

23.

Gomez-Pinilla

Hillman

(2013). The influence of exercise on cognitive abilities. Comprehensive Physiology, 3, 403–428.

24.

Gong

Y. X.

Cai

T. S.

(1993). The Wechsler Intelligence Scale for children revised in China (C-WISC). Changsha, China: Hunan Map Press.

25.

Gregory

A. M.

Sadeh

(2012). Sleep, emotional and behavioral difficulties in children and adolescents. Sleep Medicine Review, 16, 129–136.

26.

Hartescu

Morgan

Stevinson

C. D.

(2015). Increased physical activity improves sleep and mood outcomes in inactive people with insomnia: A randomized controlled trial. Journal of Sleep Research, 24, 526–534.

27.

Hopkins

Yuill

Branigan

H. P.

(2017). Inhibitory control and lexical alignment in children with an autism spectrum disorder. Journal of Child Psychology and Psychiatry and Allied Disciplines, 58, 1155–1165.

28.

Khan

N. A.

Hillman

C. H.

(2014). The relation of childhood physical activity and aerobic fitness to brain function and cognition: A review. Pediatric Exercise Science, 26, 138–146.

29.

Krakowiak

Goodlin-Jones

Hertz-Picciotto

Croen

L. A.

Hansen

R. L.

(2008). Sleep problems in children with autism spectrum disorders, developmental delays, and typical development: A population-based study. Journal of Sleep Research, 17, 197–206.

30.

Kredlow

M. A.

Capozzoli

M. C.

Hearon

B. A.

Calkins

A. W.

Otto

M. W.

(2015). The effects of physical activity on sleep: A meta-analytic review. Journal of Behavioral Medicine, 38, 427–449.

31.

Lakes

K. D.

(2013). Measuring self-regulation in a physically active context: Psychometric analyses of scores derived from an observer-rated measure of self-regulation. Mental Health and Physical Activity, 8, 189–196.

32.

Lakes

K. D.

Hoyt

W. T.

(2004). Promoting self-regulation through school-based martial arts training. Journal of Applied Developmental Psychology, 25, 283–302.

33.

Lee

Kim

(2014). Effects of exercise with or without light exposure on sleep quality and hormone responses. Journal of Exercise Nutrition & Biochemistry, 18, 293–299.

34.

Liu

Hubbard

J. A.

Fabes

R. A.

Adam

J. B.

(2006). Sleep disturbances and correlates of children with autism spectrum disorders. Child Psychiatry and Human Development, 37, 179–191.

35.

Malow

B. A.

Marzec

M. L.

McGrew

S. G.

Wang

Henderson

L. M.

Stone

W. L.

(2006). Characterizing sleep in children with autism spectrum disorders: A multidimensional approach. Sleep, 29, 1563–1571.

36.

Marrin

Drust

Gregson

Morris

C. J.

Chester

Atkinson

(2011). Diurnal variation in the salivary melatonin responses to exercise: Relation to exercise-mediated tachycardia. European Journal of Applied Physiology, 111, 2707–2714.

37.

Maski

Owens

J. A.

(2016). Insomnia, parasomnias, and narcolepsy in children: Clinical features, diagnosis, and management. Lancet Neurology, 15, 1170–1181.

38.

Mazurek

M. O.

Sohl

(2016). Sleep and behavioral problems in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46, 1906–1915.

39.

McEwan

Ginis

K. A. M.

Bray

S. R.

(2013). The effects of depleted self-control strength on skill-based task performance. Journal of Sport and Exercise Psychology, 35, 239–249.

40.

Morgan

David

Gasoigne

(2007). Daily sleep diary. Retrieved from https://www.nhs.uk/Livewell/insomnia/Documents/sleepdiary.pdf

41.

Newell

K. M.

(1991). Motor skill acquisition. Annual Review of Psychology, 42, 213–237.

42.

Ozonoff

(1995). Executive functions in autism. In Schopler

Mesibov

G. B.

(Eds.), Learning and cognition in autism (pp. 199–219). Boston, MA: Springer.

43.

Pan

C.-Y.

Chu

C.-H.

Tsai

C.-L.

Sung

M.-C.

Huang

C.-Y.

W.-Y.

(2017). The impacts of physical activity intervention on physical and cognitive outcomes in children with autism spectrum disorder. Autism, 21, 190–202.

44.

Pesce

Croce

Ben-Soussan

T. D.

Vazou

McCullick

Tomporowski

P. D.

Horvat

(2016). Variability of practice as an interface between motor and cognitive development. International Journal of Sport and Exercise Psychology. Advance online publication. doi:10.1080/1612197X.2016.1223421

45.

Psychology Software Tools, Inc. (2012). E-Prime 3.0. Pittsburgh, PA. Retrieved from http://www.pstnet.com

46.

Reed

Watts

Truzoli

(2013). Flexibility in young people with autism spectrum disorders on a card sort task. Autism, 17, 162–171.

47.

Richdale

A. L.

Schreck

K. A.

(2009). Sleep problems in autism spectrum disorders: Prevalence, nature, & possible biopsychosocial aetiologies. Sleep Medicine Review, 13, 403–411.

48.

Rossignol

D. A.

Frye

R. E.

(2011). Melatonin in autism spectrum disorders: A systematic review and meta-analysis. Developmental Medicine & Child Neurology, 53, 783–792.

49.

Russell

Jarrold

Henry

(1996). Working memory in children with autism and with moderate learning difficulties. Journal of Child Psychology and Psychiatry, 37, 673–686.

50.

Russo

Flanagan

Iarocci

Berringer

Zelazo

P. D.

Burack

J. A.

(2007). Deconstructing executive deficits among persons with autism: Implications for cognitive neuroscience. Brain Cognition, 65, 77–86.

51.

Sachse

Schlitt

Hainz

Ciaramidaro

Schirman

Walter

. . . Freitag

C. M.

(2012). Executive and Visuo-motor function in adolescents and adults with autism spectrum disorder. Journal of Autism and Development Disorders, 43, 1222–1235.

52.

Sadeh

Sharkey

K. M.

Carskadon

M. A.

(1994). Activity-based sleep-wake identification: An empirical test of methodological issues. Sleep, 17, 201–207.

53.

Souders

M. C.

Mason

T. B.

Valladares

Bucan

Levy

S. E.

Mandell

D. S.

. . . Pinto-Martin

(2009). Sleep behaviors and sleep quality in children with autism spectrum disorders. Sleep, 32, 1566–1578.

54.

Souders

M. C.

Zavodny

Eriksen

Sinko

Connell

Kerns

. . . Pinto-Martin

(2017). Sleep in children with autism spectrum disorder. Current Psychiatry Reports, 19, 34.

55.

Szuhany

K. L.

Bugatti

Otto

M. W.

(2015). A meta-analytic review of the effects of exercise on brain-derived neurotrophic factor. Journal of Psychiatric Research, 60, 56–64.

56.

Taylor

M. A.

Schreck

K. A.

Mulick

J. A.

(2012). Sleep disruption as a correlate to cognitive and adaptive behavior problems in autism spectrum disorders. Research in Developmental Disability, 33, 1408–1417.

57.

Thorndike

R. L.

Hagen

E. P.

Sattler

J. M.

(1986). Stanford-Binet Intelligence Scale (4th ed.). Chicago, IL: Riverside.

58.

Uzefovsky

Allison

Smith

Baron-Cohen

(2016). Brief report: The Go/No-Go Task online: Inhibitory control deficits in autism in a large sample. Journal of Autism and Developmental Disorders, 46, 2774–2779.

59.

Verté

Geurts

H. M.

Roeyers

Oosterlaan

Sergeant

J. A.

(2006). The relationship of working memory, inhibition, and response variability in child psychopathology. Journal of Neuroscience Methods, 151, 5–14.

60.

Wachob

Lorenzi

D. G.

(2015). Brief report: Influence of physical activity on sleep quality in children with autism. Journal of Autism and Developmental Disorders, 45, 2641–2646.

61.

Wallace

G. L.

Yerys

B. E.

Peng

Dlugi

Anthony

L. G.

Kenworthy

(2016). Assessment and treatment of executive function impairments in autism spectrum disorder: An update. In Hodapp

R. M.

Fidler

D. J.

(Eds.), International review of research in developmental disabilities (Vol. 51, pp. 85–122). New York, NY: Academic Press.

62.

Wirojanan

Jacquemont

Diaz

Bacalman

Anders

T. F.

Hagerman

R. J.

Goodlin-Jones

B. L.

(2009). The efficacy of melatonin for sleep problems in children with autism, fragile X syndrome, or autism and fragile X syndrome. Journal of Clinical Sleep Medicine, 5, 145–150.

63.

World Health Organization (2016, March). Questions and answers about autism spectrum disorders (ASD). Retrieved from http://www.who.int/features/qa/85/en/

64.

Wright

Lipszyc

Dupuis

Thayapararajah

S. W.

Schachar

(2014). Response inhibition and psychopathology: A meta-analysis of go/no-go task performance. Journal of Abnormal Psychology, 123, 429–439.

65.

Xiao

Hong

Yang

. . . Liu

(2012). Response inhibition impairment in high functioning autism and attention deficit hyperactivity disorder: Evidence from near-infrared spectroscopy data. PLoS ONE, 7, e46569.

66.

Zimmerman

B. J.

Kitsantas

(1997). Developmental phases in self-regulation: Shifting from process goals to outcome goals. Journal of Educational Psychology, 89, 29.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.50 MB

0.02 MB