DF's visual brain in action: The role of tactile cues Robert L. Whitwell a,b , A. David Milner c , Cristiana Cavina-Pratesi c , Caitlin M. Byrne a , Melvyn A. Goodale a,n a The Brain and Mind Institute, The University of Western Ontario, London, ON, Canada N6A 5B7 b The Graduate Program in Neuroscience, The University of Western Ontario, London, ON, Canada N6A 5B7 c Department of Psychology, Durham University, South Road, DH1 3LE Durham, UK article info Keywords: DF Perception Action Grasping Dorsal stream Ventral stream abstract Patient DF, an extensively-tested woman with visual form agnosia from ventral-stream damage, is able to scale her grip aperture to match a goal object's geometry when reaching out to pick it up, despite being unable to explicitly distinguish amongst objects on the basis of their different geometries. Using evidence from a range of sources, including functional MRI, we have proposed that she does this through a functionally intact visuomotor system housed within the dorsal stream of the posterior parietal lobe. More recently, however, Schenk (2012a). The Journal of Neuroscience, 32(6), 2013–2017; Schenk (2012b). Trends in Cognitive Sciences, 16(5), 258–259. has argued that DF performs well in visually guided grasping, not through spared and functioning visuomotor networks in the dorsal stream, but because haptic feedback about the locations of the edges of the target is available to calibrate her grasps in such tasks, whereas it is not available in standard visual perceptual tasks. We have tested this 'calibration hypothesis' directly, by presenting DF with a grasping task in which the visible width of a target varied from trial to trial while its actual width remained the same. According to the calibration hypothesis, because haptic feedback was completely uninformative, DF should be unable to calibrate her grip aperture in this task. Contrary to this prediction, we found that DF continued to scale her grip aperture to the visual width of the targets and did so well within the range of healthy controls. We also found that DF's inability to distinguish shapes perceptually is not improved by providing haptic feedback. These findings strengthen the notion that DF’s spared visuomotor abilities are driven largely by visual feedforward processing of the geometric properties of the target. Crucially, these findings also indicate that simple tactile contact with an object is needed for the visuomotor dorsal stream to be engaged, and accordingly enables DF to execute visually guided grasping successfully. This need for actions to have a tangible endpoint provides an important new modification of the Two Visual Systems theory. & 2013 Elsevier Ltd. All rights reserved. 1. Introduction 1.1. Two visual systems hypothesis and patient DF Human beings and other primates are capable of reaching out and grasping objects with great skill and precision, and vision plays an indispensable role in this ability. Marc Jeannerod and his colleagues in Lyon pioneered the study of visuomotor control in both humans and non-human primates, and he wrote the first comprehensive account of visuomotor neuroscience (Jeannerod, 1988). Subsequently, Marc was one of the first to argue that “visuomotor coordination relies on a specific mode of visual input processing which is different from that giving rise to visual perception” (Jeannerod & Rossetti, 1993). At about the same time, Goodale and Milner (1992) had independently proposed a similar thesis, identifying the specific cortical visual pathways in the cerebral cortex that might underlie these separable visual functions. Accord- ing to their account, the visual control of action is mediated by pathways that arise in early visual areas and project to the posterior parietal cortex, whereas visual perception is mediated by pathways that also arise in early visual areas but project to the inferotemporal cortex. Although the “two-visual systems” (TVS) hypothesis is strongly supported by a range of evidence from neurobehavioural and neurophysiological studies of human and non-human primates as well as neuroimaging (for reviews, see Goodale, 2011; Milner & Goodale, 2006, 2008), the key observation that led to the genesis of the core concepts of the TVS hypothesis was the striking dissociation between perception and action observed in patient DF (Goodale, Milner, Jakobson, & Carey, 1991). DF, who was a young woman at the time that the first studies were carried out, had developed a profound visual form agnosia as Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/neuropsychologia Neuropsychologia 0028-3932/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropsychologia.2013.11.019 n Correspondence to The Brain and Mind Institute, The University of Western Ontario, The Natural Sciences Centre, Room 205, London, ON, Canada N6A 5B7. Tel.: þ1 519 661 2070; fax: þ1 519 661 3961. E-mail address: mgoodale@uwo.ca (M.A. Goodale). Please cite this article as: Whitwell, R. L., et al. DF's visual brain in action: The role of tactile cues. Neuropsychologia (2013), http://dx.doi. org/10.1016/j.neuropsychologia.2013.11.019i Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎