On the basis of a computational and neurodynamical model, we investigate a cognitive impairment in stroke patients termed visual neglect. The model is based on the "biased competition hypothesis" and is structured in several network modules which are related to the dorsal and ventral pathway in the visual cortex. By damaging the model, visual neglect can be simulated and explained as an unbalanced neurodynamical competition. We predict that acquiring knowledge of objects can increase the frequency of saccades to previously ignored object parts. This prediction is confirmed in a single case study by monitoring eye movements of a neglect patient. Key words: Computational Neuroscience, Visual Neglect, Prediction, Single Case Study, Eye Movements
Dietmar Heinke, Gustavo Deco, Josef Zihl, Glyn W.