The mystery of patient DF
Patient DF is arguably the most important patient in the field of visuomotor research. The finding that DF could make accurate visually-guided movements despite her poor performance in tasks testing her spatial vision suggested that vision for action and vision for perception are based on different neural substrates. This prompted the development of the perception/action model by David Milner and Mel Goodale and initiated a whole new area of research devoted to the exploration of the visual attributes and cortical structures that define the perceptual and motor components of the human visual system. Despite the fact that the perception/action model is based to a considerable extent on experiments with DF, there are aspects of her behaviour, which are difficult to explain within that model. Firstly, it is a surprise that she is not making more use of the information in her visuomotor stream to compensate for her difficulties in perceptual tasks. For example in the case of grasping an object it appears that DF has access to accurate information about the size of the object, but when asked to discriminate perceptually between objects of different sizes, she fails. It appears that the same information is required for both tasks, and it is obvious that this information is represented somewhere in her brain. Thus, why is DF not using the information that guides her in the motor task to solve the perceptual task? Another problem is posed by DF’s performance in handwriting tasks. Handwriting is a classical example of a visually-guided motor act, and therefore according to the perception/action model her performance in this task should be normal. But it is not. Why? In my talk I will present findings from new experiments with DF which suggest answers to those two questions but also a slight modification of the perception/action model.