Kim J, Chung S, Tennant L K, Singer R N, Janelle C M
Kyungpook National University, Korea.
Percept Mot Skills. 2000 Feb;90(1):253-61. doi: 10.2466/pms.2000.90.1.253.
Traditional one-dimensional error scores are still consistently used in research on motor learning to quantify two-dimensional error; however, the inherent differences in two-dimensional tasks render that application inappropriate and often misleading. Consequently, the purpose of this paper was to propose a novel method of presenting errors, which more precisely represents the accuracy, direction, and variability of error in two-dimensional settings. Although closely related to several alternatives for representing errors, the methodology used and the results obtained provide a more accurate procedure for pinpointing critical trends in what have been commonly referred to as AE (absolute error), VE (variable error), CE (constant error), and E (total variability). The proposed measurements of AVE (adjusted variable error), DE (directional error), TSE (total spread of error), and RE (radial error) provide composite error scores carrying a variety of information about performance on two-dimensional tasks. Formulas and examples are provided to facilitate computation and enhance understanding of the proposed scores.
在运动学习研究中，传统的一维误差分数仍被持续用于量化二维误差；然而，二维任务中的固有差异使得这种应用并不恰当，且常常具有误导性。因此，本文的目的是提出一种呈现误差的新方法，该方法能更精确地表示二维环境中误差的准确性、方向和变异性。尽管与几种表示误差的替代方法密切相关，但所使用的方法和获得的结果为精确找出通常被称为绝对误差（AE）、可变误差（VE）、恒定误差（CE）和总变异性（E）中的关键趋势提供了更准确的程序。所提出的调整可变误差（AVE）、方向误差（DE）、误差总散布（TSE）和径向误差（RE）测量方法提供了包含二维任务表现各种信息的综合误差分数。文中给出了公式和示例，以方便计算并增进对所提出分数的理解。
