Shape-centered representations of bounded regions of space mediate the transformation of retinotopic representations into conscious perception of objects

The primary visual cortex represents the retinotopic orientation of visual primitives (edges, blobs, bars), but our conscious perception is of orientated objects (e.g., dogs, forks) in the environment. How this transformation operates remains unknown. We report here the study of a young woman presenting with an extraordinarily clear and informative visual disorder that affects highly specific aspects of object perception allowing precise inferences about the type and properties of visual representations that mediate this transformation. Davida perceives sharp-edged 2D bounded regions of space of medium to high contrast as if they were plane-rotated by 90, 180 or 270 degrees around their center, mirrored across their own axes, or both. In contrast, her perception of strongly blurred or very low contrast shapes, and of compound shapes emerging from a collection of bounded elements, is intact. The nature of her errors implies that visual perception is mediated by a representation of each bounded region of space in a shape-centered coordinate system aligned on either the shapes most elongated part or on the shapes axis of symmetry and centered either at the midpoint of the shapes most elongated part or at the shapes centroid. The selectivity of her disorder to sharp-edged medium to high-contrast stimuli additionally suggests that duplicate shape-centered representations are computed in parallel from information derived from the parvocellular and magnocellular subcortical channels and integrated precisely at the level at which shape representations must be mapped onto a behaviorally relevant frame of reference.