Of frames and mirrors: reflections on neglect. a reply to McCarthy (2002).

Vision, like physics, requires a reference frame within which the position and motion of an object can be defined (Tadin et al., 2002). The brain initially receives information in retinocentric coordinates but successive neural computations can develop and integrate such information so as to transform the nature of the visual reference frame into another representation (e.g., head-centred, body-centred, object-centred). The main reason for the brain to perform such transformations of reference frame could be to achieve a more efficient representation. An example from astronomy illustrates the argument: a planet’s position can be determined either in a heliocentric or geocentric reference frame but astronomers have found the former representation to be more efficient than the latter. The study of the syndrome of spatial neglect appears to provide a unique opportunity for the understanding of the nature of the reference frames used by the brain. These patients show a remarkable anomaly in their interactions with surrounding space, as well as with their own body. They show remarkable omissions in recognition, gross mistakes in spatial judgments, and impaired navigational abilities often incompatible with everyday functionality. To a first approximation, these patients can be described as unresponsive to stimuli in the contralesional part of space. Some issues arise immediately from this description. If the lesion (typically located within the right hemisphere of the patient) and the neglected object each represent two locations in space, both of these locations could be described in relation to one another in different ways, depending on the specific frame of reference used. Thus, investigating whether neglect patients ignore either contralesional stimuli, stimuli on the left side of their retinae or of their body midline, one side of each object, stimuli within or outside reach, etc., might allow us to understand which frames of reference are relevant for human vision. Impairments in neurological patients that can best be described within a specified spatial frame support the idea that such a spatial frame is actually used by the brain. In this commentary we review work on reference frames in neglect, focussing particularly on some recent studies that have used mirrors to disentangle frames of reference, and discuss some of the issues raised by McCarthy (2002) in her commentary on Laeng et al. (2002b).