The search for shape-centered representations.

If you were to ask someone in the street what they find most fascinating about the human brain the chances are they will tell you about a strange neuropsychological case study in which the patient presents with an odd and intriguingly specific deficit. Perhaps they lost the ability to recognize faces, tasted in colors or could no longer navigate familiar environments. Neuropsychological case studies, it turns out, are thus intrinsically fascinating to both scientists and the general public alike. Consistent with this general tenet, the case of ‘Davida’ reported by Vannuscorps et al. (Vannuscorps et al., 2021) is no exception. Davida, a right-handed young woman who is unremarkable in terms of her medical, neuropsychological, and neurological history nevertheless presents with a specific deficit in which she perceives 2D bounded shapes, defined by sharp edges of high contrast, as either inverted, mirror-reversed or plane-rotated by 90 or 180 degrees. Remarkably, her ability to judge the shape, size, location, distance, and tilt of 2D stimuli is no different from controls and her perception of 3D, strongly blurred or very low contrast shapes is intact—a curious case if ever there was one. When reading case-studies one can be left questioning the depth and breadth of the testing, wishing that further tests had been conducted to rule out all manner of possible explanations. Of course, time is tight when conducting these studies, but that sense of a missed opportunity can persist, nevertheless. That is not the case when reading Vannuscorps et al. Over the course of two years, Davida completed six experimental phases comprising over fifty individual experiments. Within these, many of the conceivable adaptations and modifications to the experimental set-up were tested including, but not limited to, visual, auditory, and somatosensory perception, visual illusions, object naming, object matching and drawing/tracing—a monumental achievement for both the researchers and Davida herself. Collectively, these data provide a benchmark for comprehensive behavioral profiling in such casestudies. The fact that Davida systematically mis-perceives 2D shapes defined by sharp and high contrast edges as either reversed or plane-rotated and yet perceives low contrast 2D shapes and 3D shapes correctly, suggests a dissociation between the visual information originating from the parvocellular and magnocellular pathways, pointing to a failure to integrate parvocellular information correctly. But, perhaps the most intriguing implication of Davida’s behavioral profile is that the implied cortical locus of her deficit lies at an intermediate stage of visual processing, somewhere between retinotopic based frames of representation in early visual cortex (e.g., V1-V3) (Wandell et al., 2007), and more abstract spatiotopic and bodycentered coordinate systems that are presumably computed more anterior in the brain (Groen et al., 2021). Vannuscorps et al. term this type of representation “intermediate shape-centered representation” (ISCR) and argue that these are distinct from subsequent and deeper levels of representation that underpin conscious perception and action. A critical and obvious question emerging is thus, where is the most likely cortical locus for such ISCRs within the visual hierarchy? Although no neuroimaging results were reported here, Vannuscorps et al. argue that cortical areas LO1-LO2 (Larsson & Heeger, 2006) are suitable candidates, and several known features of LO1LO2 support their claim. For instance, LO1-LO2 play