Abstract
Introduction. We tested the hypothesis that varied training, as compared to specific training, enhances motor learning. A well-known theory in learning, the schema theory, suggests that variation during practice enhances long-term learning, generalization, and retention. A particularly interesting prediction from this theory is that people who practice throwing alternately at two target distances (varied group) yield better learning and generalization than people who practice only at one target distance (specific group). This prediction was supported by results from Kerr and Booth (1978). Methods. We trained our varied group at 5 and 9 ft target distances, and our specific group at 7 ft. Participants threw a 9.05 oz. bean bag backward over their shoulders to one of the following targets: 3, 5, 7, 9, and 11 ft. These targets were marked on the floor for the participants to view at the beginning of each block. Participants either received or did not receive feedback after each throw. In our short-term study (n = 255), all pre-test, practice (120 trials), post-test 1, and post-test 2 were completed within an hour. In our long-term study (n = 30), the practice had 10 sessions (120 trials each) that lasted for approximately six weeks. We measured signed errors, absolute errors, and variances. Results. Only limited evidence of generalization was found as a result of varied training. One week after the long-term training at post-test 1, we found that the varied group better generalized to 11 ft than the specific group. However, this advantage ceased to be statistically reliable one week later at post-test 2. Conclusion. Overall, the learning could be characterized as generalizing locally to nearby distances (±2 ft), but not beyond, regardless of the training schedule or feedback. These results suggest similarities between motor learning and classic perceptual learning.
Meeting abstract presented at VSS 2017