Zampeta Kalogeropoulou, GRK 1589 / BCCN Berlin / HU Berlin

Feature-based Attention in Active Vision

Humans perceive a stable world, despite head, body and rapid eye-movements changing what falls upon their retinas. Therefore, a natural question emerges: how does the visual system construct the stable percept of the world during eye-movements? Spatial attention (attending to a specific location in the visual field) has been thoroughly investigated in the literature and has been suggested to play a key role in the process of continuous perception of the world. Yet, there is another flavour of attention, whose contribution to visual stability has not so far been investigated: feature-based attention (enhancing a specific feature across the field of view e.g, color ’red’). Feature-based attention is another candidate for supporting visual stability that is independent of space. This feature may be a crucial ingredient for visual stability across eye-movements.
In this thesis, I investigated the contribution of feature-based attention to two fundamental visual processes. First, if feature-based attention has been deployed and performance reaches its highest point, is this high performance maintained across eye-movements? Second, I studied the interplay between feature-based attention and visual short-term memory. My findings suggest that feature-based attention is not disrupted by eye-movements and can be redeployed to internal representations, changing feature priorities held in visual short-term memory. Therefore, I propose that feature-based attention may be another mechanism of stitching together discontinued visual input and hence contributes to the undisturbed perceptual panorama of our visual environment.

Additional Information

PhD defense in the GRK 1589 "Sensory Computation in Neural Systems"

Organized by

Martin Rolfs / Robert Martin