This RCT study aims to compare effectiveness physical exercise and melatonin supplement on treating sleep disturbance in children with ASD. A four-arm randomized controlled trial with equal allocation ratio to the three intervention groups (i.e., jogging; melatonin supplement, combine jogging and melatonin supplement) and one placebo-control group will be conducted . Actigraph device and sleep log will be used for sleep assessment. 6-sulfoxymelatonin from 24-h and first morning urinary samples. The investigators will monitor the changes of four sleep parameters (sleep onset latency, sleep efficiency, wake after sleep onset and total sleep duration) and the 24-h and first morning melatonin level throughout the whole study.

Children with ASD are characterized by deficits to some extent - from mild to severe - in social interaction, stereotyped behavior and limited interest. Apart from these core symptoms, sleep disturbance is also commonly found in this population. To cope with sleep problems, medication such as supplemental melatonin is commonly used. Melatonin is the primary hormone produced by the pineal gland for properly timing circadian sleep/wake rhythms to enhance sleepiness during night. Previous research indicated that children with ASD tend to have abnormal melatonin secretion and this may contribute to the abnormal development of sleep-wake cycles. Over the past decade, effect of supplemental melatonin on sleep disturbance treatment has been studied widely in children with autism and the efficacy is well recognized.

Apart from supplemental melatonin, researchers have also explored other intervention strategies that could ameliorate the symptoms of sleep disturbance in children with ASD. One intervention strategy that receives growing attention is physical exercise. Previous studies showed that physical exercise could improve sleep quality in children with ASD. Then it is natural to ask which intervention - physical exercise or melatonin supplement is more effective to promote sleep health in the population. This proposed study aims to compare the effectiveness of three interventions, namely, physical exercise, melatonin supplement and the combination of the two, on treating sleep disturbance among children with ASD via a melatonin mediated mechanism model. Two significant impacts will be expected. First, if the three interventions were equally effective, then physical exercise, which is more naturalistic and brings other health benefits, can be an alternative option for practitioners and parents to treat sleep problems for their patients and children with ASD. Second, if there was a difference between the three interventions on treating sleep problems, it will then inform further research on the relationship between physical exercise, supplemental melatonin and sleep. For example, if the combination was the most effective intervention to ameliorate the sleep disturbance in children with ASD, it would imply the investigators may need to further investigate other factors such as neurological factors (e.g. brain derived neurotrophic factor) or psychological factors (e.g. happiness, motivation) that may play a role on mediating the relationship between physical exercise and sleep. Each participant will attend 3 one-week-long assessments in their respective schools, where the investigators will assess their habitual sleep patterns and endogenous melatonin level before the intervention (T1), in the mid of the study (5 weeks after the commencement of the study) (T2) and after the 10-week intervention (T3). The mid-assessment is valuable to know how much variance in sleep behaviors is accounted by variance in melatonin, as well as to assess the adherence of the interventions.

The findings of this study will allow the investigators to reaffirm if melatonin mediates the relation between physical exercise and changes in sleep quality, which would further strengthen the current limited evidence on the efficacy of physical exercise for sleep problems in children with ASD. The findings of the proposed study will ultimately lead to optimal treatment interventions for sleep disturbance not only in children with ASD, but also in any population suffering from sleep disturbances.

Inclusion Criteria:

• age 8 - 12 years
• clinical diagnosis of ASD from a physician or clinical psychologist based on the Diagnostic and Statistical Manual of Mental Disorders, 5th edition, (DSM-5) and confirmed with the Autism Diagnostic Interview- Revised (ADI-R)
• pre-puberty as indicated by Tanner stage I through screening by a physician
• being given an average of 8 hours of sleep per night by their parents over the past 3 months
• parents reported sleep onset delays of 30 minutes of longer on three or more nights per week
• free of psychotropic medications (allergy medications and medications for constipation are allowed)
• non-verbal IQ over 40
• able to perform the requested physical intervention.

Exclusion Criteria:

• with one or co-morbid psychiatric disorders identified with a structured interview based on DSM-5
• with other medical conditions that limit their physical activity participation and sleep (e.g., asthma, seizure, cardiac disease etc)
• with a complex neurologic disorder (e.g., epilepsy, phenylketonuria, fragile X syndrome, tuberous sclerosis)
• participants who are currently meeting physical activity guidelines (more than or equal to 60 minutes of moderate and vigorous physical activity each day).