Spatial neglect is a strong and negative predictor of general functional outcome after stroke, and its therapy remains a challenge. Whereas inhibitory non-invasive brain stimulation (NIBS) over the contralesional, intact hemisphere has generally been shown to ameliorate neglect on a group level, a conspicuous variability of the effects at the individual level is typically observed. We aimed to comprehensively assess the characteristics and determinants of the effects of inhibitory NIBS in neglect, identifying which patients would respond to this therapeutic approach and which not. To this end, we prospectively included sixty patients with a subacute right-hemispheric stroke. In thirty patients with spatial neglect, continuous Theta Burst Stimulation (cTBS) was applied over the left posterior parietal cortex (PPC) in a randomized clinical trial, either in 8 or 16 trains, or as sham stimulation. Thirty patients without neglect served as control group. Neglect severity was measured with a neuropsychological test battery and the Catherine Bergego Scale (CBS), at admission to and at discharge from inpatient neurorehabilitation, as well as at 3 months follow-up. General functional outcome was assessed by means of the Functional Independent Measurement (FIM) and the Lucerne ICF-based Multidisciplinary Observation Scale (LIMOS). The impact of clinical and demographic factors was evaluated, and the influence of lesion location and extension was assessed by means of voxel-based lesionsymptom mapping (VLSM). On a group level, both cTBS protocols (i.e., 8 and 16 trains) significantly reduced neglect severity in both the CBS and the neuropsychological tests, at discharge and 3 months later. Furthermore, cTBS significantly improved general functional outcome. On an individual level, hierarchical cluster and VLSM analyses revealed that the variability in the responses to cTBS is determined by the integrity of inter-hemispheric connections within the corpus callosum, in particular parieto-parietal connections. In cTBS responders, in whom neglect and general functional outcome were significantly improved, the corpus callosum was intact, whereas this was not the case in cTBS non-responders. Moreover, analyses based on the Proportional recovery rule and the Maugeri predictive stroke recovery model showed that the recovery ! ! 3! ! of neglect and of the activities of daily living (ADL) was accelerated only in cTBS responders. Furthermore, the level of ADL recovery of these neglect patients was brought close to the one of right-hemispheric control patients without neglect. Hence, in neglect patients with intact interhemispheric connectivity, cTBS over the contralesional PPC significantly improves and accelerates neglect recovery and, associated with it, general functional outcome.