We wished to contrast cortical activation during hand movements in profoundly weak patients with motor neuropathy and in normal controls using a paradigm that is behaviourally matched between the two groups. Previous work has suggested that a passive movement task could be appropriate. Using functional magnetic resonance imaging (fMRI), we first characterised patterns of brain activation during active and passive index finger movements in healthy controls (n=10). Although the relative activation differences were highly variable, there was a trend for the mean number of significantly activated voxels in the primary motor cortex contralateral to the hand moved (CMC) to be lower for the passive than for the active task (40% relative decrease, P=0.09). There was a small posterior shift in the centre of mass of the CMC (mean, 8 mm, P<0.02) and of the ipsilateral sensorimotor cortex (IMC) (mean, 11 mm, P<0.05). No activation with passive movement was found in the patients with severe distal sensory neuropathy (n=2), suggesting that activation with passive movements is dependent on sensory feedback and unlikely to be due to mental imagery alone. In contrast, patients with severe pure motor neuropathies (MN, n=2) showed substantial increases in the volumes of activation compared to controls. The relative increases in numbers of voxels activated above threshold in different regions of interest for both the active (MN/controls: CMC, 2. 1; IMC, 8.1; supplementary motor area [SMA], 5.2) and passive (CMC, 2.6; IMC, 8.0; SMA, 5.1) tasks were similar. These results confirm expansion of cortical representation for finger movement in patients with motor neuropathy and demonstrate central reorganisation as a consequence of the motor nerve loss. An expanded representation for finger movement in the primary motor cortex with peripheral weakness suggests the possibility that the primary motor cortex may encode motor unit activation rather directly.