No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Over the past several decades, experimental studies of the organisation of spatial frequency (SF) preference in mammalian visual cortex (V1) have reported a wide variety of conflicting results. A consensus now appears to be emerging that in the superficial layers SF is mapped continuously across the cortical surface. However, other evidence suggests that SF may differ systematically with cortical depth, at least in layer 4, where the magnocellular (M) and parvocellular (P) pathway afferents terminate in different sublaminae. It is not yet clear whether the topographic organisation for SF observed in the superficial layers is maintained throughout the input layers as well, or whether there is a switch from a laminar to a topographic organisation along the vertical dimension in V1. I present results from two alternative self-organising computational models of V1 that receive natural image inputs through multiple SF channels in the LGN, differing in whether they develop laminar or topographic organisation in layer 4. Both models lead to topographic organisation for orientation (OR) and SF preference in upper layers, consistent with current experimental evidence. The results suggest that in either case separate sub-populations of neurons are required to obtain a wide range of SF preference from Hebbian learning of natural images. These models show that a laminar organisation for SF preference can coexist with a topographic, columnar organisation for orientation, and that the columnar organisation for orientation is dependent upon inter-laminar feedback. These results help clarify and explain the wide range of SF results reported in previous studies.

Potential differences in developmental trajectory were investigated in autism at both the macro- and micro-scopic scale, using regional volumetric measurements from in-vivo scans and measurements of minicolumnar organisation of the cortex in post-mortem tissue. In addition, a study was carried out to investigate the sensitivity of measures of cortical diffusion to cortical architecture. Three key regions of interest were studied throughout this thesis, orbital frontal cortex (BA11), primary auditory cortex (BA41) and part of the inferior parietal lobe (BA40). Subjects with ASD showed increases in grey matter in left parietal cortex and decreases in left BA11 compared to controls. In addition, subjects with ASD showed increased grey matter volume with age in both BA41 and the inferior parietal lobe, whereas controls only showed a negative correlation between grey matter volume in BA41 and age. Wider minicolumns were found in ASD in all regions, suggesting pathology is not restricted to higher order association areas. Differences seemed more pronounced at younger ages suggesting an altered developmental trajectory in ASD. Such an increase in minicolumnar width arguably underlies the feature-based processing style seen in ASD. A pilot study using post-mortem DTI scans of MS brains revealed a relationship between measures of the directionality of diffusion and the width of axonal bundles in the cortex, an aspect of the minicolumnar arrangement. When extending this investigation to a set of ASD and control brains, evidence was found for different relationships between axon bundle width and measures of the directionality of diffusion in the cortex, suggesting that although differences in axon bundle width were not seen between groups, there may be differences in the composition of the axon bundles between ASD and control groups.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Can the brain register the emotional valence of brief exposures of complex natural stimuli under conditions of forward and backward masking, and under conditions of attentional competition between foveal and peripheral stimuli? To address this question, three experiments were conducted. The first, a behavioural experiment, measured subjective valence of response (pleasant vs unpleasant) to test the perception of the valence of natural images in brief, masked exposures in a forward and backward masking paradigm. Images were chosen from the International Affective Picture System (IAPS) series. After correction for response bias, responses to the majority of target stimuli were concordant with the IAPS ratings at better than chance, even when the presence of the target was undetected. Using functional magnetic resonance imaging (fMRI), the effects of IAPS valence and stimulus category were objectively measured on nine regions of interest (ROIs) using the same strict temporal restrictions in a similar masking design. Evidence of affective processing close to or below conscious threshold was apparent in some of the ROIs. To further this line of enquiry, a second fMRI experiment mapping the same ROIs and using the same stimuli were presented in a foveal (‘attended’) peripheral (‘to-be-ignored’) paradigm (small image superimposed in the centre of a large image of the same category, but opposite valence) to investigate spatial parameters and limitations of attention. Results are interpreted as showing both valence and category specific effects of ‘to-be-ignored’ images in the periphery. These results are discussed in light of theories of the limitations of attentional capacity and the speed in which we process natural images, providing new evidence of the breadth of variety in the types of affective visual stimuli we are able to process close to the threshold of conscious perception.