Spatial representations for perception and action in the human brain

How the brain represents space and uses this information to generate actions is a central question in the cognitive neurosciences and an essential step toward an understanding of brain disorders. A critical aspect of this question is the reference frame with respect to which object locations for actions are encoded. There is considerable evidence that the encoding of intended actions in visual space is represented in a dynamic, eye-centered map, such that each eye movement requires an internal updating of this information. This process, however, is made more complex if we also consider the need to remember object locations designated for possible future actions while navigating in a three-dimensional (3-D) world. If and how this updating process works under these additional constraints is currently unknown. Recently, I have designed a model that elucidates the computational principles required under these conditions. A main objective of this project is to obtain psychophysical and fMRI data that will test various predictions of this model. However, the storing and updating of object locations is not the whole story. To generate actions in the world, this (updated) spatial information needs to be integrated with effector information, for example, hand position. While different theoretical schemes have been proposed for this process, it is surprisingly difficult to distinguish between them experimentally. We will use a novel approach to resolve this issue. We will first simulate the various theoretical alternatives in 3-D. This is a vital step, which is designed to reveal different predictions of these schemes. To further clarify this issue, experimental evidence (in the form of psychophysical and fMRI data) will be collected which will form an empirical test between these theoretical concepts.