(A) The experimental, wearable finger-extension device with 10 cm finger extensions and the control device without the extensions. (B) Example of the Jenga stacking training task. Participants wore the finger-extension device and were asked to stack a full Jenga tower; task performance was measured by the time taken to complete the tower. (C) Outline of the experimental design. (D) Group data showing improvement in performance (reduction in completion time) across training blocks.

Proprioceptive judgments across stages of augmentation with wearable finger-extension device.

(A) Comparison of proprioceptive maps from stage 1 (baseline) and stage 2 (initial device fitting). (B) Left: Comparison of biological finger segment between stage 2 and stage 3 (after training). Right: Comparison of artificial finger segment between the same stages. (C) Comparison of biological finger representation from stage 1 to stage 4 (post-device removal). (D) Comparison of biological finger representation from stage 3 (with device) to stage 4 (after removal). In all line plots, perceived location (y-axis) is plotted against actual target location (x-axis). Error bars represent standard error of the mean. The dashed line indicates the identity line (slope = 1). Raincloud plots show individual participant slopes in each stage. The following figure is formatted in the same fashion.

Proprioceptive judgments across stages with training performed using the control device.

(A) Comparison of proprioceptive maps from stage 1 (baseline) and stage 2 (initial finger-extension device fitting). (B) Left: Comparison of biological finger segment between stage 2 and stage 3 (after training with control device). Right: Comparison of artificial finger segment between the same stages. (C) Comparison of biological finger representation from stage 1 to stage 4 (post-device removal). (D) Comparison of biological finger representation from stage 3 (after training with control device) to stage 4 (after removal). In all line plots, perceived location (y-axis) is plotted against actual target location (x-axis). Error bars represent standard error of the mean. The dashed line indicates the identity line (slope = 1). Raincloud plots show individual participant slopes in each stage.

Direct comparison of representational changes across training devices.

Difference scores in proprioceptive localization slopes across stages, shown separately for the spatially augmenting finger-extension condition (Experiment 1A) and the non-augmenting control condition (Experiment 1B). (A) Difference between Stage 1 and Stage 2. (B) Difference between Stage 2 and Stage 3. (C) Difference between Stage 1 and Stage 4. (D) Difference between Stage 3 and Stage 4. In all panels, each boxplot shows the group distribution for one device condition, with individual participant values overlaid.