Head centered but no body centered modulation of horizontal space misrepresentation in neglect and in hemianopia.

Neglect patients fail to attend to stimuli or parts of stimuli in the space contralateral to their unilateral brain damage. Visual neglect is doubly dissociated from impairments due to damage of retinotopic representations of space in primary visual pathways (i.e. hemianopia) and is thought to depend on the disruption of extra-retinotopic multimodal representations. Anisometric representation of horizontal space with relative underestimation of contralesional object size or between-object distances and corresponding relative overestimation of equivalent ipsilesional ones, was recently hypothesised to be a relevant component of neglect (Bisiach et al., 1994; Milner et al., 1993; Halligan and Marshall, 1991; Ishiai et al., 1994). Bisiach et al. (1994), specifically emphasised that anisometric misrepresentation can explain the relative contralesional overextension and the ipsilesional underextension showed by neglect patients in the reproduction of horizontal distances. This finding seems indeed difficult to explain both by contralesional hypokinesia or representational scotoma and by ipsilesional attentional bias or defective disengagement. However, misreproduction of horizontal distances appears selectively marked only when neglect is accompanied by concomitant damage of retinotopic representations of contralesional space in occipital and parietal areas (Doricchi and Angelelli, 1999; Doricchi et al., 2002a and 2002b), whereas it can be completely absent in neglect patients without concomitant visual field defects (Karnath and Ferber, 1999). Compared with neglect patients affected by concomitant hemianopia, hemianopic patients without neglect have an opposite pattern of misrepresentation (i.e. contralesional overestimation/ ipsilesional underestimation; Zihl and Von Cramon, 1986; Barton et al., 1998; Kerkhoff, 1993). The aim of the present study was to identify the spatial coordinate system (or systems) whose disruption is responsible for space misrepresentation in neglect patients (with or without hemianopia) and in pure hemianopic patients. We used a visual version of the endpoint task (Bisiach et al. 1994). In this task any asymmetry in reproduction can be attributed to genuine space misperception because the patient is forced by the examiner to look at and check the endpoints of the distance being reproduced at each step of each trial. The same could not happen in size comparison or landmark tasks with a segment on either side of a central reference. In this case, attention and inspection could be preferentially attracted by the center of “mass” of stimuli rather than by their endpoints, leaving open the possibility of mistaking the effects due to defective attention and representation of the most contralesional section of stimuli for those genuinely linked to space misrepresentation. In experiment 1 we investigated whether changing the position of test stimuli with respect to the patient’s body midsagittal plane (corresponding to manipulations of proprioceptive information signalling the position of the trunk under the head) produces any change in the degree of horizontal space misrepresentation. In experiment 2 we evaluated the effect of changing the headcentered position of the test stimuli (separated and combined manipulation of eye and neck position) in order to define the influence of eye-neck proprioceptive inputs on space misrepresentation.