华东师范大学学报(教育科学版) ›› 2026, Vol. 44 ›› Issue (11): 64-78.doi: 10.16382/j.cnki.1000-5560.2026.11.006

• 课程与教学 • 上一篇    下一篇

超越“认知减负”:“理想困难”视角下的认知负荷再思考

王哲, 程超(), 朱天润, 楼婧怡   

  1. 教育部人文社科重点研究基地华东师范大学课程与教学研究所,上海 200062
  • 出版日期:2026-11-01 发布日期:2026-10-24

Beyond Cognitive Load Reduction: Rethinking Cognitive Load from the Perspective of Desirable Difficulties

Zhe Wang, Chao Cheng(), Tianrun Zhu, Jingyi Lou   

  1. Institute of Curriculum and Instruction, East China Normal University, Shanghai 200062, China
  • Online:2026-11-01 Published:2026-10-24

摘要:

在信息过载、教育技术快速发展与课堂“减负”话语不断强化的背景下,认知负荷长期被视为需要控制和降低的对象。然而,理想困难框架表明,某些在学习过程中增加主观努力、降低即时表现的困难条件,反而可能促进长期保持和知识迁移。本文尝试在认知负荷理论与理想困难框架之间建立理论对话,重新思考认知负荷在学习中的功能。文章首先梳理内在认知负荷、外在认知负荷和生成性认知负荷的基本内涵及其与学习结果的关系,指出认知负荷并非天然具有消极意义,其作用取决于负荷是否超过学习者的认知承载上限,以及是否能够引发目标导向的有意义加工。随后,文章结合通道效应等典型学习情境,分析不同类型认知负荷在何种条件下可能引发理解困难。本文认为,认知负荷能否发挥理想困难功能,主要取决于负荷强度、加工性质、动机状态与成功反馈:负荷强度须处于学习者可成功加工的区间,困难类型应指向有意义加工过程,学习者需要具备愿意投入的动机状态,并能够获得可感知的成功反馈。基于此,文章主张应超越“负荷越低越好”或“困难越多越好”的二元对立,将认知负荷理解为一种既可能阻碍学习,也可能促进深度加工的动态条件。教学设计的关键不在于单纯降低或增加负荷,而在于根据学习者先验知识、任务复杂度和学习目标,对认知负荷进行精细化调节,使适度困难转化为促进深度学习与迁移的生产性摩擦。

关键词: 认知负荷, 理想困难, 生产性摩擦, 生成性加工, 深度学习

Abstract:

Against the backdrop of information overload, the rapid development of educational technology, and the growing emphasis on “reducing burden” in classroom discourse, cognitive load has long been regarded as something that should be controlled and reduced. However, the framework of desirable difficulties suggests that certain difficult conditions that increase subjective effort and reduce immediate performance during learning may instead promote long-term retention and knowledge transfer. This article attempts to establish a theoretical dialogue between cognitive load theory and the desirable difficulties framework, thereby reconsidering the function of cognitive load in learning. It first reviews the basic meanings of intrinsic, extraneous, and germane cognitive load and their relationships with learning outcomes, arguing that cognitive load is not inherently negative. Rather, its effects depend on whether the load exceeds learners’ cognitive capacity and whether it can elicit goal-directed, meaningful processing. The article then draws on typical learning contexts, such as the modality effect, to analyze the conditions under which different types of cognitive load may give rise to desirable difficulties. It argues that the transformation of cognitive load into desirable difficulties is mainly constrained by four conditions: the intensity of load must fall within a range that learners can successfully process; the type of difficulty should involve meaningful processing; learners need to be motivationally willing to invest effort; and they should receive perceptible feedback indicating successful progress. Based on this analysis, the article argues for moving beyond the binary opposition between “the lower the load, the better” and “the more difficulty, the better.” Instead, cognitive load should be understood as a dynamic condition that may either hinder learning or promote deeper processing. The key to instructional design is not simply to reduce or increase load, but to finely regulate cognitive load in light of learners’ prior knowledge, task complexity, and learning goals, so that moderate difficulty can be transformed into productive friction that supports deep learning and transfer.

Key words: cognitive load, desirable difficulties, productive friction, generative processing, deep learning