The study aims to investigate whether neuropsychological function (particularly cognitive flexibility and executive function), functional (assessed by resting functional MRI, rfMRI) and structural connectivity (assessed by DSI), and electrophysiological function (assessed by event-related potential [ERP]: mismatch negativity, MMN and P50) can be effective cognitive endophenotypes (biomarkers) for Autism spectrum disorders (ASD).

Autism spectrum disorders (ASD) is a common severe, multi-factorial, highly heritable, clinically and genetically heterogeneous, life-long impairing childhood-onset neurodevelopmental disorder. Due to its high prevalence and severe lifelong impairment without effective prevention and pharmacological treatment, this disastrous disease has been prioritized for epidemiological, molecular genetic and biomarker studies in the world.

Specific aims:

1. To validate the structural and functional connectivity in fronto-temporal, and cortico-striato-thalamic circuitry as effective imaging endophenotypes by demonstrating the differences between ASD probands with CNVs findings (n=22) and their unaffected siblings (n=22), probands without CNVs and known genetic markers related to ASD (n=22) and their unaffected siblings (n=22), and matched neurotypicals (n=22 for each);
2. To validate the neuropsychological functioning (particularly set-shifting and executive function) as effective neuropsychological endophenotypes by demonstrating the differences among the six groups;
3. To validate the electrophysiological functioning assessed by ERP as effective neurophysiological endophenotypes by demonstrating the differences among the 6 groups; and
4. To correlate the data from structural and functional connectivity, neuropsychology, and electrophysiology involving altered brain functioning.

The investigators anticipate that probands with CNVs may have higher level of decreased structural and functional connectivity, impaired ERP and neuropsychological functioning than probands without CNVs. The alterations in the structural and functional connectivity, neurophysiological and neuropsychological functioning would be observed in the unaffected siblings as compared to neurotypical participants. If CNV in the probands is proved to be de novo mutation and their unaffected siblings did not have such results, the likelihood of different functioning between their unaffected siblings and neurotypical participants would be decreased. The genetic dosage (CNV, rare mutation with moderate to large clinical effect, versus multiple common variants with very small effects, with regards to unaffected siblings, and neurotypicals) is anticipated to pose the strongest effects on the microstructural integrity of white matter, followed by functional connectivity and electrophysiological function, and neuropsychological function with the least effect.

Inclusion Criteria:

1. subjects have a clinical diagnosis of autistic disorder or Asperger disorder defined by the DSM-IV and ICD-10 criteria, which was made by a board-certificated child psychiatrists at the first visit and following visits (probands only, exclusion criteria for unaffected siblings and school controls);
2. their ages range from 10 to 25 (because all the probands with CNV aged > 10);
3. both parents are Han Chinese; and
4. subjects and their parents consent to participate in this study for complete phenotype assessments (3 visits of assessments).

Exclusion Criteria:

• if they currently meet criteria or have a history of the following condition as defined by DSM-IV: Schizophrenia, Schizoaffective Disorder, or Organic Psychosis.

The sample (6 groups) consists of 44 individuals with clinical diagnosis of ASD confirmed by the ADI-R and ADOS assessments (22 with CNVs and 22 without CNVs), their unaffected siblings (n=22 for each group) and age-, sex-, handedness-, and IQ-matched school comparison groups (22 for each group).