BACKGROUND: The difficulties in obstacle walking are significant in people with Parkinson's disease (PD) leading to an increased fall risk. Effective interventions to improve obstacle walking with possible training-related neuroplasticity changes are needed. We developed two different exercise programs, complex walking training and motor-cognitive training, both challenging motor and cognitive function for people with PD to improve obstacle walking. AIM: To investigate the effects of these two novel training programs on obstacle walking and brain activities in PD. DESIGN: A single-center randomized, single-blind controlled study. SETTING: University laboratory; outpatient. POPULATION: Individuals with idiopathic PD. METHODS: Thirty-two participants were randomly assigned to the complex walking training group (N.=11), motor-cognitive training group (N.=11) or control group (N.=10). Participants in training groups received exercises for 40 minutes/session, with a total of 12-session over 6 weeks. Control group did not receive additional training. Primary outcomes included obstacle walking, and brain activities (prefrontal cortex (PFC), premotor cortex (PMC), and supplementary motor area (SMA)) during obstacle walking by using functional near-infrared spectroscopy. Secondary outcomes included obstacle crossing, timed up and go test (TUG), cognitive function in different domains, and fall efficacy scale (FES-I). RESULTS: The motor-cognitive training group demonstrated greater improvements in obstacle walking speed and stride length, SMA activity, obstacle crossing velocity and stride length, digit span test, and TUG than the control group. The complex walking training did not show significant improvement in obstacle walking or change in brain activation compared with control group. However, the complex walking training resulted in greater improvements in Rey-Osterrieth Complex Figure test, TUG and FES-I compared with the control group. CONCLUSIONS: Our 12-session of the cognitive-motor training improved obstacle walking performance with increased SMA activities in people with PD. However, the complex walking training did not lead such beneficial effects as the cognitive-motor training. CLINICAL REHABILITATION IMPACT: The cognitive-motor training is suggested as an effective rehabilitation program to improve obstacle walking ability in individuals with PD.