Objective
To understand the brain processes that lead to behavior will be one of the greatest scientific challenges of the new millennium. The aim of the present research proposal is to examine the basic mechanisms of visual perception for the purposes of scene recognition and action planning. We propose to address this aim along three complementary themes of research. First, we shall study the perceptual attributes for recognition by looking at luminance and chromatic cues for scenes, the attention to diagnostic information, and the binding problem of the different visual attributes. Second, we shall focus on the perceptual attributes for action by studying the visual segmentation problem, which three- dimensional cues are being used for action, and the underlying cerebral pathways for these visuo-motor tasks. Finally, we shall study the interaction between recognition and action by focusing on the imitation of human movement, the eye movements used during scene exploration, and the high-level representations for the planning of actions. These three research themes will involve three more specific research projects and each research project will correspond to two or three experiments. Each research project will be investigated with a variety of experimental paradigms borrowed from psychology, motor control and brain imaging. For instance, the perception of motion transparency will be addressed both from a psychophysical point of view to determine the efficiency of human observers in this task, but also with the help of fMRI studies to reveal the brain structures involved in this task. Our ultimate goal is to increase our understanding of the perception for recognition and action through active collaboration and shared expertise of eight leading laboratories in Europe in these fields. We therefore anticipate major scientific breakthroughs from this unique joint effort.
Fields of science
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
Topic(s)
Data not availableCall for proposal
Data not availableFunding Scheme
NET - Research network contractsCoordinator
G12 8QQ GLASGOW
United Kingdom