Spatial representations and multiple-visual-systems hypotheses: Evidence from a developmental deficit in visual location and orientation processing

AH, a young, well-educated woman, has a developmental deficit in processing visual location and orientation information. Her deficit manifests itself in a wide range of visual tasks, including visually-guided reaching, copying pictures and words, and responding verbally to the location or orientation of visual stimuli; however, her performance in non-visual localization tasks is intact. AH's visual location and orientation errors are systematic left-right or up-down reflections (e.g., reaching to the far right for an object on the far left). More specifically, the errors involve reflection across the point where AH's attention is focused, regardless of where her eyes are fixated. These results imply that at some level(s) of the visual system, locations and orientations of visual stimuli are represented in a spatial coordinate system with an origin defined by the focus of attention. In these attention-centered representations location is specified in terms of distance and direction of displacement from the attentional focus along horizontal and vertical reference axes. AH's errors, I argue, result from misrepresentation of displacement direction (e.g., left rather than right, down rather than up) along a reference axis. Several visual variables dramatically affected AH's performance in visual location and orientation tasks: She was much more accurate for stimuli that were brief, moving, flickering, low in contrast, or high in eccentricity, than for those that were long in duration, stationary, continuous, high in contrast, and low in eccentricity. These results suggest that location and orientation are computed in each of two visual subsystems, which I call transient and sustained, and that AH's deficit affects only the sustained subsystem. I argue that AH's performance poses challenges to multiple-visual-subsystems hypotheses proposed by Ungerleider and Mishkin (1982) and by Milner and Goodale (1995).