Motor task variation induces structural learning

Abstract

When we have learned a motor skill, such as cycling or ice-skating, we can rapidly generalize to novel tasks, such as motorcycling or rollerblading [1-8]. Such facilitation of learning could arise through two distinct mechanisms by which the motor system might adjust its control parameters. First, fast learning could simply be a consequence of the proximity of the original and final settings of the control parameters. Second, by structural learning [9-14], the motor system could constrain the parameter adjustments to conform to the control parameters' covariance structure. Thus, facilitation of learning would rely on the novel task parameters' lying on the structure of a lower-dimensional subspace that can be explored more efficiently. To test between these two hypotheses, we exposed subjects to randomly varying visuomotor tasks of fixed structure. Although such randomly varying tasks are thought to prevent learning, we show that when subsequently presented with novel tasks, subjects exhibit three key features of structural learning: facilitated learning of tasks with the same structure, strong reduction in interference normally observed when switching between tasks that require opposite control strategies, and preferential exploration along the learned structure. These results suggest that skill generalization relies on task variation and structural learning.

Figure 1

Schematic Diagram of Structural Learning (A) The task space is defined by two parameters, but for the given task, only certain parameter combinations occur (black line). This relationship is indicated by the curved structure, which can be parameterized by a one-dimensional metaparameter μ. However, a parametric learner that is ignorant of the structure has to explore the full two-dimensional space when readjusting the parameter settings. (B) A structural learner, in contrast, takes the relationship between the parameters into account. By adjusting only the metaparameter μ, the learning problem is effectively one-dimensional.

Figure 2

Structural Learning of Visuomotor Rotations (A) Learning curves for a block of +60° rotation trials performed by a group that had experienced random rotations before (Rot-learner, red), a control group that had only experienced movements with veridical feedback (blue), and a group that experienced random linear transforms (green). The rotation group shows strong facilitation. The initial angular error over all subjects is shown with double-exponential fits. (B) Learning curves for a subsequent block of −60° rotation trials performed by the same groups. The interference effect that can be seen in the control group is strongly reduced in the rotation group. (C) Learning curves for a subsequent block of +60° rotation trials performed by the same groups. Again, the random rotation group shows a performance advantage in the first ten trials. (D–F) The same effects are much more pronounced for the cumulative error computed over the entire trajectory. Facilitation (D), interference reduction (E), and facilitation of relearning (F) are significant. The median error over all subjects and the pertinent interquartile confidence interval are shown.

Figure 3

Structural Learning of Rotations versus Shearings (A) Mean trajectories over all subjects in 60° rotation probe trials performed by a group that experienced random rotations before (red) and another group that experienced random shearings before (black). The two groups react differently to the same perturbation. The trajectories to the different targets have all been rotated so that the cursor target is vertically above the starting location. (B) Mean trajectories in 1.5 shearing probe trials performed by the same groups. (C and D) Speed profiles for the same trials. (E and F) Variances in the same probe trials. The variance is reduced when subjects face a probe trial that is compatible with the structure of their previously experienced task.

Figure 4

Structural Learning of 3D Rotations (A) Angular error in probe blocks of horizontal (red) and vertical (blue) 45° rotations experienced by a group that experienced random horizontal rotations before. There is a clear facilitation for learning the horizontal rotation. The black line indicates performance in the block of null-rotation (washout) trials preceding the probe block. (B) Performance error in the same probe blocks for a group that experienced random vertical rotations before. The facilitation pattern is reversed. (C and D) Movement variance shortly before trial end for both kinds of probe blocks. The variance in the task-irrelevant direction—perpendicular to the displacement direction—is significantly reduced for isostructural probe blocks (ellipses show the standard deviation). This suggests that subjects explored less outside the structure they had learned during the random rotation blocks. (E and F) Circular histograms of initial movement adaptation from the first trial of the probe block to the second trial. Subjects responded to probe blocks from the same structure in a consistent way, correcting toward the required target. In the case of probe trials for a different structure, subjects also showed components of learning in the direction of the previously learned structure.