Relevant raw outcome data.

Cognitive training techniques such as motor imagery (MI)–cognitive simulation of movement, has been found to successfully facilitate skill acquisition. The MI literature emphasizes the need to accurately imitate key elements of motor execution to facilitate improved performance outcomes. However, there is a scarcity of MI research investigating how contemporary approaches to motor learning, such as nonlinear pedagogy (NLP), can be integrated into MI practice. Grounded in an ecological dynamics approach to human movement, NLP proposes that skilled action is an emergent process that results from continuous interactions between perceptual information of the environment and movement. This emergent process can be facilitated by the manipulation of key task constraints that aim to encourage learners to explore movement solutions that satisfy individual constraints (e.g., height and weight) and achieve successful performance outcomes. The aim of the present study was to explore the application of a NLP approach to MI approach for skill acquisition. Fourteen weightlifting beginners (two female and 12 male) participated in a 4-week intervention involving either NLP (i.e. analogy-based instructions and manipulation of task constraints) or a linear pedagogy (LP; prescriptive instructions of optimal technique, repetition of same movement form) to learn a complex weightlifting derivative. Performance accuracy, movement criterion (barbell trajectory type), kinematic data, and quantity of exploration/exploitation were measured pre-mid-post intervention. No significant differences (p = .438) were observed in the amount of exploration between LP (EER = 0.41) and NLP (EER = 0.26) conditions. Equivalent changes in rearward displacement (R×D) were observed with no significant differences between conditions for technique assessments 1, 2, or 3 (p = .13 - .67). Both NLP and LP conditions were found to primarily demonstrate ‘sub-optimal’ type 3 barbell trajectories (NLP = 72%; LP = 54%). These results suggest that MI instructions prescribing a specific movement form (i.e., LP condition) are ineffective in restricting available movements to a prescribed technique but rather the inherent task constraints appear to ‘force’ learners to explore alternative movement solutions to achieve successful performance outcomes. Although MI instructions prescribing specific techniques have previously supported improved skill development, the current findings indicate that learners may self-organise their movements regardless of MI instructions to satisfy individual and task constraints while achieving improved performance. Therefore, it may be beneficial to consider scripts that are more outcome focused and incorporate task constraints to facilitate learners’ inherent exploration of individual task solutions.

诸如运动表象（motor imagery, MI）这类认知训练技术——即对动作进行认知模拟——已被证实可有效促进技能习得。现有运动表象研究均强调，需精准复刻动作执行的核心要素，方能优化训练效果。然而，当前关于运动表象的研究中，鲜有探讨如何将当代动作学习范式（如非线性教学法，nonlinear pedagogy, NLP）融入运动表象训练的相关成果。
非线性教学法以人类动作的生态动力学研究路径为理论根基，提出熟练动作是一种涌现过程，由个体对环境的知觉信息与自身动作之间的持续交互所催生。该涌现过程可通过调控核心任务约束得以优化，此类约束旨在引导学习者探索契合自身约束（如身高、体重）且能达成良好训练效果的动作解决方案。
本研究旨在探索将非线性教学法应用于运动表象训练以促进技能习得的可行性。十四名举重新手（2名女性、12名男性）参与了为期四周的干预实验，被随机分配至非线性教学组（即基于类比的指导与任务约束调控）或线性教学组（linear pedagogy, LP：对最优动作技术的指令性指导、固定动作模式的重复练习），学习一项复杂的举重衍生动作。
分别在干预前、干预中、干预后对以下指标进行测量：动作准确性、动作准则（杠铃轨迹类型）、运动学数据，以及探索与利用的量化值。结果显示，线性教学组（EER=0.41）与非线性教学组（EER=0.26）的探索量无显著差异（p=0.438）。在第1、2、3次技术评估中，两组的向后位移（R×D）变化幅度相当，组间无显著差异（p=0.13~0.67）。两组均主要呈现‘次优’的3型杠铃轨迹（非线性教学组占72%，线性教学组占54%）。
上述结果表明，要求固定动作模式的运动表象指导（即线性教学组）并未能将学习者的动作限制在指定技术框架内，相反，任务固有的约束似乎会‘迫使’学习者探索其他动作解决方案以达成良好训练效果。尽管此前已有研究证实，指定具体技术的运动表象指导可助力技能提升，但本研究结果显示，学习者可自主调整动作模式，无需依赖运动表象指导，即可在满足个体与任务约束的同时优化训练表现。因此，在设计运动表象训练脚本时，可考虑更聚焦于训练目标，并融入任务约束要素，以引导学习者自主探索适配自身的动作解决方案，这或许能带来更佳的训练收益。