Plasticity refers to susceptibility of an organism to change. Cognitive training is an intervention approach based on the notion of plasticity. It entails the repeated exercise of a set of higher-order cognitive abilities over several weeks after which performance gains are expected on the trained as well as untrained tasks. Cognitive training has produced successful results in various clinical groups, but its benefits have not been explored in Autism Spectrum Disorder (ASD). The present study will develop a software-based training program tailored to the cognitive deficits in ASD. The investigators will also examine possible training-induced functional changes within the brain using functional Magnetic Resonance Imaging (fMRI). Fifty children with ASD 3-7 years will be recruited and randomly assigned to the control (n=25) or the training group (n=25). A subgroup of these samples will carry out the response inhibition and set-shifting tasks in the fMRI scanner. The study will consist of a pre-post design and a four-month follow up. Repeated measures Analysis of Variance (ANOVA) will be carried out with group (training, control) as the between subjects factor and Time (pre- post-training, follow-up) as the within subjects factor to identify training induced cognitive improvements. To determine training-induced biological changes within the brain, activity maps associated with response inhibition and set-shifting at pre-training and post-training sessions will be entered into a group ANOVA and contrasted for differences within- and between groups.

1. INTRODUCTION Autism Spectrum Disorder (ASD) is a severe neurodevelopmental disorder characterized by deficits of social and communicative skills, restricted set of interests, activities and/or repetitive behaviours that are typically observed in early childhood. The prevalence of ASD is about 1 in 68 children and it is the fastest growing developmental disability. The cost of diagnosing and treating ASD is estimated to be around $236-262 billion annually.

Plasticity broadly refers to the susceptibility of an organism to change. The human brain shows incredible plastic capacity. Multiple neuronal networks can sustain the same cognitive function (a mental ability) with different systems supporting the same function in different individuals. This "many-to-one" structure-function relationship is one form of plasticity. Examining neural networks responsible for a cognitive process in different individuals can provide insight into the plastic capacity of that process.

Cognitive training is centered on the notion of plasticity. It entails the repeated exercise of a specific cognitive process (or multiple processes) over several weeks, after which performance gains are expected on the trained task as well as various untrained tasks that directly or indirectly involve the targeted cognitive process(s). The generalization of performance gains to untrained tasks is termed "transfer" and is essential to the efficacy of the training. Cognitive training has been used to remediate deficits in adults with strokes, multiple sclerosis, Schizophrenia, children with working memory deficits , children with attention deficit hyperactivity disorder (ADHD), healthy pre-kindergarten children, as well as to enhance cognitive performance in healthy young adults, and healthy older adults. Yet, the benefits of this intervention for children with ASD have received relatively little attention.

Recent research indicates that disrupted patterns of cortical development in ASD may lead to its clinical manifestation. More specifically earlier reports have shown a pattern of reduced long-range cortical connectivity and increased localized functional connectivity in ASD. This pattern has been recently verified using highly stringent imaging analysis methods. This altered functional connectivity may be especially disruptive to cognitive functions that demand integrative information processing such as executive functioning (higher order cognitive functions that control other cognitive processes eg. Conscious control of thought and action), theory of mind, face processing, language and communication, all of which have been previously established as impaired processes in ASD.

Moreover, approximately 50-70% of children with ASD are diagnosed with intellectual disability, which manifests as cognitive impairments. Intellectual disability has been shown to place children with ASD at risk for a "low functioning" trajectory throughout life and at risk for having more severe symptoms. Cognitive training has been shown to enhance executive functions underlying intellectual capacity such as working memory, fluid intelligence, executive attention as well as academic achievement . Cognitive training delivered at an early age may strengthen the processes that are important for intellectual capacity and therefore improve the clinical trajectory of ASD.

Although these recent theories of ASD point to deficits of executive functions, this field is currently lacking an evidence-based intervention, which directly tackles deficits of executive functions. The existing interventions for ASD (mainly consisting of behavioural skills development) are complex in administration and require highly trained staff. As a result, these approaches have placed an extremely high demand on clinics and practitioners creating long wait lists. These approaches have poor accessibility for many families and schools. The most striking and consistent limitation of the existing interventions is an apparent lack of transfer of learned skills to other conditions and contexts. Meaning, new learned behaviours are limited to the specific context in which they are trained. Generalization to other tasks and contexts is a distinguishing strength of cognitive training, this approach can be used at home via a personal computer providing more accessibility, and it can be used in conjunction with the behavioral skills development to improve the child's receptiveness to learning. Thus, cognitive training may have the potential to provide promising results in the areas where the existing treatments have shown limitations.

The present project intends to first develop a software based cognitive training program tailored to cognitive needs of children with ASD (Cognitive Assessment and Video-game Intervention Solutions, CAVINS)(phase 1) and then examine the program's efficacy through clinical trials and imaging of the brain (Clinical Trials and Imaging Phases). The imaging component will provide the opportunity to learn about the neural framework of some of the targeted cognitive processes as well as training-induced changes in each process. This intervention will target several functions implicated in ASD such as the ability to shift attention to a different aspect of the task, inhibitory control, working memory, planning, reasoning, selective attention, and face processing. During the imaging phase, as our first step, the investigators will use functional Magnetic Resonance Imaging (fMRI) to identify training-induced changes in the brain associated with set-shifting and response inhibition.

Response inhibition consists of two distinct forms, the restraint of a response and the cancellation of a response. Preventing a response from being initiated characterizes the restraint process whereas termination of a response that is already underway represents the cancellation process. Additionally the ability to monitor, detect and adjust behaviour following an erroneous response represents error processing and is an inherent component of response inhibition. Difficulties of response inhibition have been hypothesized to be responsible for the stereotyped and repetitive behaviour observed in ASD, predictive of "theory of mind" performances in preschoolers, and associated with altered connectivity between the frontal cortex and the striatal and parietal regions as well as volume differences and altered development of the striatum. Similarly, lower accuracy on set-shifting tasks is associated with reduced activation in frontal, striatal, and parietal cortexes and is hypothesized to be responsible for mental inflexibility, restricted and repetitive behaviours observed in ASD.

Several imaging studies have demonstrated modifications in the underlying neural network following completion of cognitive training in healthy adults. However, similar studies in children are very scarce. Currently, there are no studies that have examined training-induced changes within the brain in children with ASD. Findings from the present project will reveal benefits of cognitive training in ASD, generalization and persistence of potential benefits, identify biological changes associated with training, and provide much needed insight into the plasticity of the systems supporting two cognitive functions implicated in ASD.

Inclusion Criteria:

• A diagnosis of ASD
• Age between 3-7 years

Exclusion Criteria:

• History of head injury
• Current medical problems that would preclude their participation in the study