Autism is a neurodevelopmental disorder characterized by difficulties in social interaction, communications and unusual repetitive behaviours. Studies have documented superior performance in autism in visuo-spatial tasks suggesting perceptual differences in this group. This has been suggested to result from enhanced processing of local details and reduced integration of global information. Contour integration tasks are one way of studying visual integration where participants have to detect a contour made of discrete elements (usually Gabor patches) against a background of noise. However, contour integration studies in autism have yielded mixed findings which possibly due to the closed shapes and longer stimulus duration used in these studies. Recently, Jachim et al (2015) compared contour detection using open and closed contours at a shorter duration reported reduced closure in the autistic group. Closed shapes are generally easier to detect than open shapes, termed the closure effect which is thought to result from more involvement of top-down input. Therefore, the main aim of this thesis was to further explore visual integration in autism using contour integration tasks, to understand the reasons for these variations in autism and neurotypicals. We presented six contour integration experiments using open (eg: lines) and closed (eg: square) contours. The first two experiments explored the influence of autistic traits in the general population and attentional feedback on contour integration in neurotypicals using a dual-task which interferes with processes concurrently using attention. The autistic traits showed a weak effect on contour integration, and the dual task produced a non-significant reduction to closure, that requires further exploration with larger participant numbers. The next two experiments investigated the influence of stimulus duration on contour integration in both groups to test the suggestion of slower global processing in autism. Results showed no effect of durations on closure in both groups, and that reduced closure in the autistic group did not resolved with longer durations. The final two experiments explored the relationship between closure and symmetry in both groups; another process that is thought to involve global processing and received input from similar areas that process closure. The control group showed better symmetry detection with closed contour, suggesting closure benefits symmetry detection. Despite comparable performance to the control group with the closed contour, the autistic group showed better thresholds with the open contour leading to lesser benefit of closure in symmetry detection. In summary, we reported possible variations in contour integration in general population. Additionally, results suggested that the autistic group processed contour integration and symmetry using different mechanisms. This converging evidence is suggestive of enhanced perception at low-level processing in autism, in support of Enhanced Perceptual Functioning Theory (EPF).