Aims: Non-invasive brain stimulation (NIBS) modalities such as transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) are powerful tool to study brain physiology and to aid rehabilitation of stroke patients (SP). Different studies showed clinical improvements after repetitiveTMS (rTMS) or tDCS on SP with motor and cognitive deficits and reported electrophysiological changes in evoked related potentials (ERPs) as N100 and P300. Few studies unified NIBS, ERPs and quantitative analysis of electroencephalographic frequencies (qEEG) in stroke rehabilitation. In the current study, we applied NIBS to inhibit the unharmed hemisphere of SP and we evaluated cognitive ERPs and qEEG as measures of the rehabilitation outcome, in a crossover trial (rTMS versus tDCS). Materials and methods: Eight patients (2/6 Males/Females, age 51 ± 10y) with chronic right brain lesions were enrolled. Patients received 10 daily sessions of low frequency 1 Hz rTMS, 80% resting motor threshold, over the intact left motor cortex (MAG&more, München); after a six months wash-out period, the same patients received 10 daily sessions of cathodal tCDS, 1,5mA intensity, over the intact left motor cortex (HDCStim, Newronica, EBN), during 20 minutes mirror box therapy. Electrodes were applied to participants’ scalp according 10-20IS. Outcome measures included P100, P300, N100, N200, N400 (ERPs) and alpha, beta1, beta2, theta, delta frequencies (qEEG). The complete evaluation was performed before and after the end of each treatment rTMS/tDCS. Analysis were conducted using Wilcoxon test for paired samples (p<0.05). Results: After both rTMS and tDCS protocols, we observed significant latency reductions for N200, P300 and N400 potentials on affected and unaffected hemisphere. Also in left hemisphere, we observed a P200 latency reduction after tDCS and a N2-P3 complex amplitude increase after TMS. After the wash-out period we observed significant latency increase for P300 and N400 potentials. We observed no significant differences in ERPs of SP who received rTMS and tDCS, even if analysis conducted on single scalp electrode shows a tendency toward better performances in tDCS patients. No significant differences were observed in the patients’ qEEG before and after NIBS. Conclusions: Worsening in ERPs latencies and amplitudes has been described in stroke chronic pathological effects. Our findings suggest that inhibitory NIBS over the intact hemisphere enhanced cerebral activity, probably affecting the inter-hemispheric functional connectivity balance. Moreover we observed similar effect in both rTMS/tDCS, even if we have to consider the cumulative effect of the two sequential techniques in the same patient.