Popular theories of motor learning (e.g., Adams, 1971; Schmidt, 1975) rely heavily on formation of new skills through refinements of pre-existing ones. Dynamic Pattern Theory has the advantage of being able to assess initial individual differences on the required task so that the subject becomes the important unit of measure. The general purpose was to identify the reciprocal influence of intrinsic patterns and learning a new pattern. In two experiments subjects were required to practice a rhythmic bimanual coordination task of the forearms using linear sliding devices. In the first experiment, 7 subjects practiced a 90° relative phase pattern for 45, 15 s trials on each of 6 practice days. In-phase and anti-phase trials were performed pre-and post-practice. Subjects were provided terminal feedback with a Lissajou figure after each practice trial and augmented feedback was provided after every 5th practice trial. Mean constant error (CE) for individual subject data and absolute CE (|CE|) for group data were used as measures of accuracy. Standard deviation of relative phase was used as a measure of stability (VE). Subjects were able to learn the 90° pattern and performance plateaued by the fourth practice day. Neither intrinsic pattern showed any destabilization, although a temporary decrease in accuracy (CE) within days was found. The four week retention test revealed no change for any pattern. Experiment 2 compared two groups practicing either 45° or 135° relative phase. It was predicted that the 135° relative phase pattern would be easier to learn because of the reduced competition from the less stable 180° intrinsic pattern. The procedure was similar to Experiment 1 except that subjects practiced for only four days. Performance of the practiced patterns was never as accurate as 0° and 180° but variability of performance was not different for both practiced and intrinsic patterns by Day 4. There was no difference in either accuracy or stability between the two groups on the practiced patterns. As in Experiment 1, there was no change from the last day of practice to the four week retention test. Individual subject data revealed numerous different paths to learning the required pattern. Constant error and VE values for intrinsic patterns were not particularly good predictors of ability to learn the practiced pattern. Additionally, a low VE was not indicative of a low CE or vice-versa when practicing the required pattern. The results from both experiments show that early in learning, competition biases performance away from the intrinsic attractors. Later in practice, subjects stabilize their performance of the new pattern and the intrinsic patterns do not destabilize. Differences in performance of the required pattern may have depended, not only on dynamic principles, but also on motivation, handedness, and conceptualization of the task. / Thesis / Master of Science (MSc)
Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/22634 |
Date | 04 1900 |
Creators | Fontaine, Robert |
Contributors | Lee, Tim, Human Biodynamics |
Source Sets | McMaster University |
Language | English |
Detected Language | English |
Type | Thesis |
Page generated in 0.0016 seconds