AIM: Previous experiments have suggested that the CNS may coordinate muscle activations through a linear combination of muscle synergies. Neurological diseases influence motor recruitment and, consequently, modify muscle synergies. Our purpose was to understand the modifications induced by Functional Electrical Stimulation (FES) in post stroke patients during upper limb rehabilitation. To this aim we compared muscle synergies pre and post intervention. METHODS: Five chronic stroke participants undertook nine tracking tasks in a planar robot in which trajectory (orientation and length), duration, speed and resistance were varied. Error in target tracking was corrected by electrical stimulation of the triceps muscle. Timing and level of stimulation was adjusted using iterative learning control algorithms. EMG signals were collected at 1500 Hz from 7 upper arm and shoulder muscles (Biceps BIC; Anterior Deltoid DEL_A; Pectoralis major PEC_M; Trapezius, upper TRAP_UPP, medial TRAP_MED, lower TRAP_LOW; Triceps TRI), both before and after 18 treatment sessions. The same EMG data was collected under the same conditions with 8 neurologically intact participants on a single occasion. Raw data were rectified, low-pass-filtered (Butterworth, 4th ord., 20 Hz) and normalized to their maximal values. Muscle synergies were extracted using the Non-Negative Matrix Factorization algorithm. Differences in weight coefficients and timing patterns between the groups have been respectively evaluated by means of their scalar product and their correlation coefficients. RESULTS: Preliminary results showed statistically significant differences between muscle synergies extracted from impaired and healthy subjects. However, despite of improved motor performance following the intervention, there were no statistically significant changes in muscle synergies. CONCLUSION: The present study shows that muscle synergies in chronic stroke patients differ from normal but are not modified by this intervention. Concomitant improvement in motor performance in the absence of significant normalisation of muscle synergies implies reinforcement of residual motor strategies acquired after stroke. Improvement in functional may be through compensation rather than true recovery and potentially demand a greater energy cost.