There is an ongoing debate concerning the perception and neural representation of space in neglect. Four experiments are here reported designed to further investigate the nature of perceptual distortions and their modifiability in patients with neglect. In Experiment 1 it was found that neglect patients, in contrast to left- or right-hemisphere lesioned control patients and normal subjects, show similar distortions of perceived visual space when judging the extension of horizontal distances (space distortion) as compared to the horizontal size of objects (size distortion). Similar deficits were present in most neglect patients in a newly developed space bisection task. These results attest that neglect patients have perceptual distortions related to within-object (size) and between-object (distance) spatial processing in their horizontal plane. Objects were oversized by 33% and distances by 19% horizontally in neglect patients, whereas all control groups showed nearly veridical spatial coding (deviations<5%). In Experiment 2 the modifiability of these distortions was tested by the use of slow visual background motion. Leftward, coherent background motion transiently restored normal horizontal size and distance coding in neglect patients, whereas rightward motion aggravated the deficit significantly in the distance task, but not in the size task. None of the other subject groups showed any influence of background motion on spatial judgments. Experiment 3 evaluated possible effects of simultaneous vs successive stimulus presentation in perceptual distortions, thus modulating attentional factors. Neglect patients performed significantly better - although not normal - with a successive presentation of the spatial stimuli (2 s, 10 s delay) as compared to the simultaneous condition in the size judgment task, but not in the distance task. In contrast, this manipulation had no effects in any of the control groups. Experiment 4 reports more detailed results of subject J.S., a neglect patient with a right mediotemporal lesion, who showed a marked horizontal size distortion, but normal horizontal distance judgments. Despite some fluctuation in J.S.'s size judgments he showed a significant overestimation of horizontal object size by +22% to +40% across several test blocks and testing sessions. Thus, performance fluctuations due to attentional or other reasons cannot fully account for J.S.'s dissociation in size vs distance judgments. He thus shows that the visual coding of horizontal spatial extension within an object can be dissociated from that of the spatial extension between objects along the horizontal plane. Finally, performance in the three spatial tasks (used in Experiment 1) was found to correlate significantly with three typical tests of spatial neglect (line bisection, cancellation, copying, r=0.42-0.77), thereby indicating a significant relationship to the neglect syndrome.Together, the results of the four experiments are interpreted in support of multiple spatial-perceptual distortions in visual neglect, which are influenced by visual motion and attention. Perceptual distortions relating to objects and space between objects are present in most neglect patients, but may dissociate, as in case J.S. It is argued that these might reflect the existence of several, partially overlapping and diverging neural maps for the representation of different spatial attributes in the horizontal plane. Furthermore, it is hypothesized that these perceptual distortions constitute an important element of the spatial-perceptual deficits encountered in the syndrome, contribute to its severity, but are not the key deficit of the disorder.