Abstract
Autism Spectrum Disorder (ASD) is a complex collection of related sub-conditions (autism, Asperger’s syndrome, pervasive developmental disorder – not otherwise specified) whose manifestation primarily affects socio-cognitive ability, resulting in differing severities of executive, functional language, and interpersonal social dysfunction. A fundamental hypothesis underlying the phenotypic expression of these disorders involves the presence of an irregular pattern of connectivity between brain regions, resulting in abnormal central coherence correlated activity between involved neural networks, allowing for the integration of brain processes for coherent conceptual thought. People exhibiting lesions in specific regions of the cerebellum display neuropsychological deficits with striking similarity to those seen in ASD, suggesting a possible association. Several non-motor areas of the cerebral cortex, including the dorsolateral prefrontal cortex (DL-PFC), frontopolar prefrontal cortex (FP-PFC), and inferior parietal lobule (IPL), have been implicated in cognitive processes underlying deficits in ASD; interestingly, those regions also communicate directly with the cerebellum. Using a circumscribed population of adults with high-functioning autism this thesis sought to investigate the integrity of these connections using Diffusion Tensor Imaging (DTI) to ascertain whether abnormalities in structural connectivity might exist (in comparison with a matched control group). We also applied Voxel-based Morphometry (VBM) to investigate grey matter density of the involved regions of interest. Based on findings of all analyses, a specific aim was to postulate plausible hypothetical links between areas displaying significant effects in the structural analyses and the symptomatic manifestation of high-functioning autism (based on well-known findings in existing literature), with the goal of proposing avenues for further research that links function with structure. Specific findings include greater integrity of neural tracts projecting from left cerebellum to right FP-PFC (increased fractional anisotropy: 0.408 ± 0.010 vs. 0.358 ± 0.014 (arbitrary units), p = 0.009, d = 1.44) and from right DL-PFC to left cerebellum (decreased mean rate of diffusivity: 0.731 ± 0.004 vs. 0.754 ± 0.007 μm2s-1, p = 0.027, d = 1.22) in the high functioning autism group compared with matched controls. In the high-functioning autism group we found clusters of significantly lower grey matter density in the right DL-PFC and left IPL (p = 0.016 and 0.041, respectively) and clusters with significantly higher density in the left cerebellum (p = 0.041). Incorporating the extant literature to assimilate the present results with previously published neuropsychological autism research, testable hypotheses were then generated around the following two themes: (1) There is a link between right DL-PFC/left cerebellum connectivity and working memory in high-functioning autism; and (2) There is a link between left cerebellum/right FP-PFC connectivity and the theory of mind (ToM), reality monitoring and higher-order planning in high-functioning autism. These themes form testable hypotheses, ripe for continuing work that directly assesses structural brain properties with functional deficits on behavioural tasks in people with high-functioning autism. This thesis can therefore be viewed as a proof of concept for the idea of generating plausible functional hypotheses using structural data in an age where the use of remotely accessed information is an increasing trend.