| Summary: | 碩士 === 國立成功大學 === 物理治療研究所 === 98 === Objective: Fast and slow movements employ different control regimes. The former favors the open-loop mode, while the closed-loop mode is prevalent to the latter. Because of the discrepancy of the control strategies, kinematic or kinetic properties, error-correction strategy, and brain activation regions differ between the motor process of slow and fast movements. It is hypothetically assumed that subsequent learning transfer varies with rate in the practice sessions. The purpose of this study was to compare the transfer effect to tracking tasks in the frequency modulation (FM) and amplitude modulation (AM) conditions, following intensive practices of sinusoidal tracking at fast and slow target frequencies. The potential neural mechanisms underlying the frequency-dependent learning transfer were discussed based on corresponding changes in physiological tremor characteristics.
Methods: Thirty-two healthy subjects, who were randomly assigned to the fast or slow group, participated in this study. The subjects in the fast and slow groups practiced a total of 15 trials of 1.4 Hz and 0.2 Hz sinusoidal force-tracking, respectively. Each practicing trial consisted of 24 seconds. Three tracking paradigms were conducted before and 30 min after the practice sessions (i.e., pre-test and re-test), including simple task (the learning task itself), FM task, and AM task. Tracking error was characterized with root mean square (RMS) of the mismatch between the force profile and target signal. The simple learning and associated transfer effect after fast and slow practices were analyzed with change in tracking error for all testing paradigms from the re-test to pre-test. The amount of learning transfer in the FM and AM testing paradigms after fast and slow practices was assessed with standardized change in tracking error, defined as the difference in tracking error between the re-test and pre-test divided by that of the pre-test. Coefficient of variance of tracking error (CVE) was used to characterize underlying error-correction strategy for each tracking paradigm. Other physiological measures included the RMS of the electromyographic (EMG) activities of first dorsal interosseous (FDI), the limb tremors (from index finger and the 2nd metacarpal bone), force tremor, and 8-12 Hz coherence peak between force tremor and EMG (CohFT-EMG).
Results: For both the fast and slow groups, tracking errors for all the simple task, FM task, and AM task significantly decreased after practices. In the FM task, the slow group exhibited a greater standardized change in tracking errors than the fast group, in support of differential transfer effect between the two groups. On the other hand, the standardized change in tracking errors in the AM task was not different for the two groups. The error-correction strategy varied conditionally to the target rate of practice. CVE of the simple test in the fast group decreased with practices, whereas it was conversely potentiated in the slow tracking group. CVE in the FM test for the fast group remained unchanged after practice, but it increased for the slow group; CVE in the AM test consistently reduced for both of the two groups. For all testing paradigms, the RMS of the EMG of the FDI muscle did not significantly differ, but the RMS of segment tremors and force tremor demonstrated a decreasing trend for practice effect. Both groups showed a significantly lower CohFT-EMG in the re-test than in the pre-test, except that the CohFT-EMG of the FM task was not mediated by practice of fast -tracking.
Conclusion: Both learning of tracking maneuver at fast and slow rates could be effectively transferred to the FM and AM tasks. However, owing to the discrepancy of error-correction strategy inherent with the fast and slow practices, slow-tracking practice led to a greater amount of learning transfer in the FM task than fast-tracking practice. Among all physiological measures, the CohFT-EMG was the most illustrative to advantageous learning transfer after slow-tracking practice in the FM task. We considered that the rate-dependent transfer effect might be ascribed to modulations of the oscillatory circuits in the central neural system and their descending common drive in control of muscle activities.
|
|---|
