Adjustments of preplanned steps are essential for fall avoidance and require response inhibition. Still, inhibition is rarely tested under conditions resembling daily living. We evaluated the ability of young and older adults to modify ongoing walking movements using a novel precision step inhibition (PSI) task combined with an auditory Stroop task.Healthy young (YA, n=. 12) and older (OA, n=. 12) adults performed the PSI task at 4 individualized difficulty levels, as a single and dual task (DT). Subjects walked on a treadmill by stepping on virtual stepping stones, unless these changed color during approach, forcing the subjects to avoid them. OA made more failures (40%) on the PSI task than YA (16%), but DT did not affect their performance. In combination with increased rates of omitted Stroop task responses, this indicates a "posture first" strategy. Yet, adding obstacles to the PSI task significantly deteriorated Stroop performance in both groups (the average Stroop composite score decreased by 13% in YA and 27% in OA). Largest deficit of OA was observed in rates of incorrect responses to incongruent Stroop stimuli (OA 35% and YA 12%), which require response inhibition. We concluded that the performance of OA suffered specifically when response inhibition was required.