THE EXPLOITATION OF TASK CHARACTERISTICS DURING SKILL ACQUISITION AS REFLECTED BY TEMPORAL EMG CHANGES

G.D. Heise (1), L. Caillouet (2), A. Cornwell (2), and B. Sidaway (2)
(1) School of Kinesiology and Physical Education
University of Northern Colorado, Greeley, CO 80639
(2) Department of Kinesiology
Louisiana State University, Baton Rouge, LA 70803

Presented at the 20th Annual Meeting of the American Society of Biomechanics
Atlanta, Georgia. October 17-19, 1996

INTRODUCTION

The purpose of this study was to examine if subjects learn to time the activation of muscles to take advantage of the spring characteristics of a ski simulator apparatus. Task performance improved as indicated by increased lateral displacement and cycle frequency. After considerable practice, durations of muscle activity in knee flexors and extensors and the duration of muscle coactivation decreased during knee flexion. The results suggest that subjects economize muscle activity by exploiting characteristics of the task.

REVIEW AND THEORY

Traditional skill acquisition research has focused on the outcome or goal of a movement as the primary dependent variable (e.g., Lee et al., 1990). Conclusions from these studies have implications for skill training concerns (e.g., training schedules, types of information feedback), however, little is added to our understanding of the underlying neuromuscular control processes. More recent research has focused on how the learner acquires control for a certain skill.

From a "dynamical systems" perspective, a learner will acquire the most economical coordination strategy available. That strategy incorporates principles such as mastering the redundant neuromuscular degrees of freedom and exploiting reactive forces present in the link-segment system (Bernstein, 1967). Vereijken (1991) suggested that learners discover the dynamical constraints imposed by the task and exploit them to their advantage. For example, she found that subjects altered the timing of their force application on a ski simulator apparatus as they became more proficient at the task. Subjects had to push the apparatus laterally against an elastic band that acted as a resistance (see Fig. 1). Vereijken suggested that learners began to exploit the elastic characteristics of the ski simulator apparatus and thus economized the timing of force application. The present study was designed to test this hypothesis from a neuromuscular level.

Presumably, subjects will learn to time the activation of muscles to take advantage of the spring characteristics of the ski simulator apparatus. Specifically, we hypothesized that, as subjects become more skilled and are able to displace the platform farther and at a higher frequency, the duration of muscle on-time of knee flexors and extensors and the coactivation between those muscle groups would decrease. We additionally hypothesized that this would be most noticeable during the time the apparatus springs back toward the center (i.e., during knee flexion of the right leg when the platform is moving right-to-left).

PROCEDURES

Six healthy men who were unfamiliar with the ski simulator volunteered as subjects (mean body mass = 78.5 ?13.2 kg; mean body height = 179 ?6 cm). Each subject attended four test sessions in which 25 trials were performed during each session. A trial consisted of a 30-sec attempt to go "as fast and as far as possible" on the ski simulator. Subjects applied force in a lateral direction against the movable platform which sat on curved rails (see Fig. 1). Once the platform reached the most lateral position, it sprang back towards the center because of elastic bands attached to it.

Prior to sessions 1 and 4, and after appropriate skin preparation, surface EMG electrodes were positioned over the bellies of muscles vastus lateralis (VL) and the long head of biceps femoris (lateral hamstrings - LH). A goniometer was placed over the right knee of each subject to monitor knee joint angle. A photoelectric cell was aimed at a reflector attached to the ski simulator platform so that an event signal was produced whenever the platform crossed top dead center (TDC). The goniometer signal, the event signal, and the EMG signals were digitized at 780 Hz for a duration of 3 s. All channels were then stored in digital format on a microcomputer.

Variables that assessed task performance were lateral platform displacement and cycle frequency. Maximum platform displacement from TDC was recorded with a video camcorder and subsequently quantified with a motion analysis system. Cycle frequency was determined from the photocell signal. A cycle was defined as the time it took a subject to go from TDC, to the far right, and back to TDC.

EMG data were full-wave rectified and low-pass filtered (fcutoff = 15 Hz). Muscle onset and cessation were identified using an interactive, computer-graphics program that plotted the submaximal, linear envelope of each channel against time. Muscle on-time durations were calculated as a percent of knee flexion and extension depending on when the muscles were active during the simulated skiing cycle. Durations of coactivation were determined by calculating, as a percent of flexion and extension, the common durations of muscle on-time between VL and LH. For the above variables, the means of trials 3, 4, and 5 were compared to the means of trials 98, 99, and 100 using dependent t-tests.

RESULTS AND DISCUSSION

Task performance improved as represented by an increase in lateral platform displacement (mean [early in practice] = 60.2 cm; mean [late in practice] = 81.7 cm) and an increase in cycle frequency (mean [early] = 1.68 Hz; mean [late] = 2.09 Hz). Fig. 2 shows portions of knee flexion and extension and temporal EMG results from early and late practice trials. Knee extension took place as subjects pushed the apparatus laterally and flexion occurred when the platform came back toward TDC. As subjects became more proficient at the task, the duration of knee extension increased and flexion decreased, as a percent of cycle time (see Fig. 2).

Figure 2. Results of knee flexion (flx) and extension (ext) durations and temporal EMG measures during knee flx and ext. (VL - vastus lateralis; LH - lateral hamstrings; coact - duration of coactivation; * p < .05)

It was during knee flexion that subjects appeared to economize their motion after considerable practice. This is evidenced by significant decreases in the duration of VL activity and VL-LH coactivation. Although not statistically significant, LH duration also decreased during flexion. This confirms Verijken's suggestion, that learners exploit the characteristics of the task. In other words, the elastic return of the ski simulator platform must aid knee flexion since the muscles are active for a shorter duration. The decrease in duration of knee flexion does not confound this finding because muscle activity durations are expressed as percents of flexion and extension, not as a function of cycle time.

Additionally, the increase in coactivation during extension suggests that subjects tuned the neuromuscular system to the characteristics of the ski simulator. Increased displacement requires knee extension against an increasing force in the elastic bands. An increase in stiffness through coactivation in knee flexors and extensors may be the mechanism of control late in the knee extension phase of the cycle.

REFERENCES

Bernstein, N. The Co-ordination and Regulation of Movements, Pergamon: Oxford, 1967.

Lee, T.D. et al. J. Motor Behav., 22, 191-208, 1990.

Vereijken, B. The Dynamics of Skill Acquisition (thesis), Vrije Universiteit, 1991.