(A) Four experimental conditions: perceptual uncertainty is shown in rows and costs of mistakes are shown in columns. U represents the magnitude of perceptual (visual) uncertainty (+U: higher uncertainty, –U: lower uncertainty), and C represents the magnitude of costs of mistakes during movement execution (+C: higher cost, –C: lower costs). (B) Example hand movement trajectories for the real & binocular condition normalized and averaged over participants separately for the four different obstacle heights (3.4, 13.0, 22.6, and 32.2 mm). The x-axis shows the distance from the start position (triangle) to the target position (square) in millimetres. The y-axis shows the height of the movements (mm). The grey-shaded horizontal bars indicate the height and extent of the different obstacles. Curves show the averaged trajectories for the different obstacle heights as indicated by the grey level. The filled circles indicate the peak height of the trajectories (mm). (C) Peak height (mm) averaged over 21 participants separately for the four experimental conditions. (red: real & monocular, magenta: real & binocular, cyan: mirror & monocular, blue: mirror & binocular). The x-axis shows the height of the obstacles (mm) and the y-axis shows the peak height of the reaching movements (mm). Error bars show ±1 SEM. Straight lines show linear regression lines. (D) Mean intercepts (baseline safety margin) and slopes (sensitivity to changes in obstacle height) for each of the four combinations of experimental conditions (as shown in A). Baseline safety margins were largest for the monocular & real condition (red), and lowest for the binocular & mirror condition (blue). Error bars show ±1 SEM.