Objective:To investigate the value of event-related brain potentials (ERPs) for evaluating the effect of repeated high-frequency transcranial magnetic stimulation (rTMS) in treating post-stroke depression.Methods:Sixty-four depressed stroke survivors were divided at random into an observation group and a control group, each of 32. Thirty-five healthy volunteers constituted a healthy control group. All of the patients were treated with 150mg/d of venlafaxine for 6 weeks. The observation group was additionally given rTMS five times a week for 6 weeks. Before and after the treatment, both patient groups were evaluated using the Hamilton depression scale (HAMD-17) as well as the visual P300.Results:After the treatment the average HAMD-17 scores of the two patient groups had decreased significantly, but with significantly greater improvement in the observation group. The total effective rate of the observation group after treatment (87.5%) was significantly higher than the control group′s rate (62.5%). Before the treatment the latency and amplitude of Cz and Fz in both patient groups was significantly delayed and lower than in the healthy group. After the treatment the average Fz amplitude in the observation group had risen and the latency had moved forward significantly compared with the other two groups. No significant differences were observed among the control group before and after the treatment. Before the treatment the average P2 and P3 latencies of the two patient groups were significantly longer than in the healthy group, while the amplitudes were significantly lower. After the treatment the average latency of P2 and the average P3 latency and amplitude of the observation group were significantly better than before the treatment. No significant differences were observed in the healthy control group.Conclusions:High-frequency rTMS can affect post-stroke depression. The MMN and visual P300 instruments can be used for rehabilitation evaluation.