BACKGROUND: People with Parkinson's disease (PD) do not differ from neurologically healthy individuals in obstacle circumvention during walking, therefore they are able to use visual feedback adequately to control motor behavior in this task. However, individuals are often distracted by the secondary task when circumventing an obstacle. An increased cognitive load can require prolonged gaze fixation time on a location of interest to compensate for longer information processing duration. RESEARCH QUESTION: To investigate the effects of cognitive dual tasking (DT) on gaze behavior during waking with obstacle circumvention in people with PD and control group, and to determine the impact of gaze behavior on motor strategy. METHODS: Fifteen individuals with PD (PD-group) and 15 neurologically healthy individuals walked at a self-selected speed over a walkway and circumvented an obstacle centered in the walkway. The experimental conditions (5 trials each one) included obstacle circumvention without DT (OC) and obstacle circumvention with DT (OCDT). In the cognitive task, the participant mentally counted the number of times a target number appeared in an audio recording. We analyzed gaze behavior (i.e. number of gaze fixations and duration on the ground and obstacle), standard gait measures and DT cost. Two-way ANOVAs were completed for gait parameters and moment of fixation. RESULTS: There was no significant difference in DT cost between groups and no obstacle contacts. The participants performed a longer mean duration of fixations on the ground during OCDT compared to OC. Group x condition interactions indicated that the PD-group delayed the obstacle fixation relative to the NHI for OCDT (p < 0.001) and presented greater medial-lateral body clearance (p < 0.001) and longer double support time (p < 0.001) during OCDT compared to OC. SIGNIFICANCE: The results of this study suggest that deficits in locomotion during DT in PD-group may be caused, at least in part, by a reduced ability to fixate gaze at appropriate times during walking.