Michèle Brouchon Yves Joanette Madeleine Samson
Study of gestural organization within the central nervous system has benefited considerably from at least two sets of knowledge in the last 2 decades. The first one is the now accepted distinction between two visual systems: one devoted to spatial vision, and the other to identification of objects ( Held, 1968; Schneider, 1969; Trevarthen, 1968). The second concerns, on the one hand, results obtained in the current quest for central nervous pathways involved in each of these visual systems and, on the other hand, the current progress in the related physiology of motor control ( Mishkin, Ungerleider, & Macko, 1983).
The problem of spatial vision has been studied through a set of observable sequential operations involving each component of the eye-limb system in a purposeful movement. These operations are detecting, orienting to, locating, and reaching a visual object. It can be argued, though, that these operations can be distinguished according to the (spatial) field in which they occur. In fact, these operations can occur either in immediate surroundings in which manual reaching and grasping are achieved without locomotion, or in the visual field outside of direct manual reaching and grasping in which visually guided locomotion and throwing are probably the most usual spatial actions. Such a descriptive approach of spatially oriented behaviors has generally led to a postulate that these are only attainable by reference to an internal map where the spatial relationships between extracorporal space and body space are stored ( Paillard, 1971). This model of a spatial internal map is generally represented as a correlation storage of the spatial cues between the sensory input, the motor command and the movement reafferences.
The hypothesis of a central nervous system constituting a spatial map in terms of the nature of the action to be achieved in a given field is supported by different