Figure 5From: The role of feed-forward and feedback processes for closed-loop prosthesis controlGrouped results from Experiment 2. Three metrics are used to compare performance. Error bars denote standard error. Data are from two groups of subjects, one with vibrotactile feedback (N = 6), one without vibrotactile feedback (N = 6). (A) Comparison of within-subject factors of visual feedback condition (red bars), tactile feedback condition (green bars), and object weight (blue bars). There was a significant within-subjects effect of both object weight and visual feedback condition, but not tactile feedback condition. Post-hoc results confirmed these differences (denoted by stars, significance at the p = .05 level.) (B) Comparison of subject's ability to discriminate object weight as a function of feedback condition. Feedback conditions were (left to right): no feedback; vibrotactile feedback only; visual feedback only; and both visual and tactile feedback. The two bars per condition indicate performance with the lightweight object (left) and heavy object (right). Successful discrimination is indicated by a positive slope. Subjects discriminated well in all feedback conditions, including in the absence of any feedback.Back to article page