Author(s): NAKAJIMA, K., MURAKAMI, H., YAMADA, N., Institution: CHUKYO UNIVERSITY, Country: JAPAN, Abstract-ID: 1743

INTRODUCTION:
Human movement follows the speed–accuracy trade-off principle, whereby increased accuracy is generally accompanied by reduced speed. In contrast, sports performance requires maintaining both speed and accuracy at high levels under temporal constraints. Previous studies using whole-body movements such as vertical jumping have shown that increasing accuracy often reduces speed; however, whether repeated practice restores or improves speed and leads to higher-level optimization of the speed–accuracy balance remains insufficiently explored. By contrast, our previous work demonstrated that accuracy can improve without sacrificing speed following repeated practice. However, motor learning does not progress uniformly across individuals and may not be sufficiently captured by short-term practice alone. The present study therefore extended the measurement to three stages to examine temporal changes in the speed–accuracy trade-off accompanying motor learning.
METHODS:
Five participants performed maximal vertical jumps under normal and adjustment conditions; detailed analyses are reported for one illustrative participant showing a typical learning pattern. Each condition consisted of 10 trials in Section 1, followed by Sections 2 and 3 with the same protocol. Take-off position, landing position, jump height, and information entropy were calculated from two-dimensional landing coordinates. Entropy was defined as zero when all landing points fell within the same spatial bin and served as an index of spatial dispersion (landing consistency).
RESULTS:
In Section 1, jump height was higher in the normal than in the adjustment condition (42.3±1.8 vs. 39.1±1.0 cm; p < .001), while information entropy was identical (1.57). In Section 2, jump height remained higher in the normal condition (41.3±1.0 vs. 39.9±0.6 cm; p < .05), whereas information entropy was lower in the adjustment condition (1.76 vs. 0.88). In Section 3, jump height did not differ between conditions (37.4±1.0 vs. 36.4±1.2 cm; p = .09), and information entropy was lower in the adjustment condition (1.29 vs. 0.92). This decrease in information entropy is consistent with improved movement consistency through learning. Information entropy may be sensitive to changes in movement control strategies with learning.
CONCLUSION:
In Section 1, the conventional speed–accuracy trade-off was observed. As learning progressed across Sections 2 and 3, changes in information entropy were observed, and the between-condition difference in jump height showed a decreasing trend, becoming smallest in Section 3. These findings suggest that the speed–accuracy trade-off is not altered transiently but is reorganized with learning, potentially reflecting a reduced cost of accuracy constraints across learning stages and a stage-wise optimization process.
Future studies should analyze all participants to assess generalizability.