Fig. 3: Weights on internal states show subjects use different motor strategies based on session performance. | Nature Communications

Fig. 3: Weights on internal states show subjects use different motor strategies based on session performance.

From: Internal states as a source of subject-dependent movement variability are represented by large-scale brain networks

Fig. 3: Weights on internal states show subjects use different motor strategies based on session performance.The alternative text for this image may have been generated using AI.

A comparison between the internal state weights and session performance reveals that higher performers learn from error and become more vigilant after perturbations. The larger the magnitude of the weight, the larger the impact the internal state has on behavior. The sign of the weight determines the impact the internal state has on behavior. Each marker is labeled by the subject. Average session performance (51%) is marked by the vertical gray dotted line. A weight of 0 (i.e., internal state does not impact behavior) is marked by the horizontal gray dotted line. The least-squares line is marked as the gray dashed line. All relationships were quantified using a two-tailed Pearson correlation without adjusting for multiple comparisons. a There is a significant relationship (two-tailed Pearson correlation: r = − 0.63, p = 0.05) between session performance and weight of error state on RT. Higher performers countered their error by reacting faster than average (0 > βSE) and lower performers maintained their error by reacting slower than average (0 < βSE) after moving too slow. b There is a significant relationship (two-tailed Pearson correlation: r = − 0.74, p = 0.01) between session performance and weight of error state on speed error (SE). Higher performers tended to move too fast (0 > βSE) and lower performers tended to move too slow (0 < βSE) after moving too slow. c There is not a significant relationship between session performance and weight of perturbed state on RT. Most subjects hesitated after perturbation trials by reacting slower (0 < βP). d There is not a significant relationship between session performance and weight of perturbed state on SE. Half of the subjects hesitated by moving slower (0 < βP) and the other half hastened by moving faster (0 > βP) after perturbations trials. Refer to Supplementary Fig. 6 and Supplementary Table 11 for all the weights. Source data are provided as a Source Data file.

Back to article page