| dc.description.abstract | Eye-movements are essential actions of the human oculomotor system: they are used to
direct high acuity foveal vision to the most informative locations of a scene, and to fix
the gaze on these locations, to maintain high resolution perceptual input. The implementation
of these processes is crucial to the natural behavioural objectives of humans and
animals. However, relatively little is known about the general scheme of gaze allocation
in space and time. This dissertation systematically discusses the existing models of eyemovements,
and proposes a computational model of human eye-movements, which takes
into account various constraints of the human visual and oculomotor systems. In the
model, the choice of the subsequent fixation location is posed as a problem of stochastic
optimal control, which is addressed by reinforcement learning methods. We show that
if biological constraints are taken into account, then the trajectories simulated under a
learned policy share both basic statistical properties and long-range correlations with human
eye movements. The model simulations have been evaluated in relation to human
psychophysical eye-tracking experiments. | en_US |
