Action-Oriented Spatial Reference Frames in Cortex

Where are you? To answer the question you have to know what kind of information is being requested. Does "where" mean where relative to the room, to the building, or to New York? The spatial reference frame that you use determines the answer, and you can imagine an endless variety of such reference frames. What reference frame does the brain use? The traditional view, strongly supported by subjective experience, is that the brain constructs a single map of the world in which the self, objects, and actions are all represented in a unitary framework. The reference frame for this map is usually thought to be in "real world" coordinates, as though we had access to some kind of absolute spatial representation. To create and use such a representation, the brain would have to transform sensory information from the coordinates of several receptor surfaces (e.g., retina, cochlea, or skin surface) into this ultimate spatial representation and later read the information back out into the motor coordinates needed for each effector system (e.g., eye, head, limb, trunk). It has proven difficult to understand how the brain combines the spatial information contained in topographic maps within each sensory system into a single, coherent representation of space, much less how this spatial information could be used to guide motor action (see Stein, 1992, for discussion). A new view holds that the brain constructs multiple spatial representations.