Producing mental pictures: a study of visual-image generation processes in brain-lesioned patients

Abstract

This thesis investigates the neural mechanisms that underpin the image-generation component of the visual-imagery process. A review of theory and data relevant to the topic is presented. Methodological issues facing researchers working with neurological and neurosurgical patients are discussed, addressing questions of ethics, the characteristics of brain lesions of different aetiologies and implications for research design. A rationale is presented for the design and selection of tests used in this research. A study involving the assessment of image generation processes in a series of 75 patients with focal brain lesions is described, with the pattern of performance of each participant individually evaluated. The performance of 3l hospitalized-control subjects is used to establish the normal range of performance on the tasks. Seven different tasks were performed. Two tasks involved visualizing letters of the alphabet and answering questions about their shapes, and performing a perceptual equivalent of the task. The third task involved visualizing the appearance of times on an analogue clockface and judging whether the angle between the hands was greater or lesser than a right angle, plus an equivalent perceptual control task. The fourth task involved drawing objects from a verbal command, naming line drawings and copying four items. Subjects were also required to draw a floor plan of the layout of their house. Where possible a comparison house plan was drawn by a family member. Finally a mental rotation and a mental paper-folding task were performed. Eight subjects had deficits of the image-generation process. Six of these had lesions in the parietal or posterior regions of their brains, with the other two having lesions of the central region. Two subjects had a bilateral lesion, two had lesions of the right hemisphere and four lesions of the left hemisphere. One additional subject with a left frontal lesion, had difficulties with the image-generation process but also with recognition and naming tasks, particularly of plants and fruit. It is concluded that the image-generation Process is bilaterally represented in the posterior regions of the brain, although there might be specialization of the process in the two hemispheres. Analysis of how the drawing task may have influenced the relative involvement of the two hemispheres is discussed, including suggestions for future research.