教学代理真的能促进学生学习吗?——基于对国内外31项实验与准实验的元分析

Can Pedagogical Agent Really Promote Student Learning?——A Meta-analysis Based on 31 Experimental and Quasi-experimental Studies

教学代理是多媒体学习环境中整合语言及视觉线索等设计而成的虚拟人物形象,在交互式教学中希望通过发挥其引导、反馈等功能进而促进学生学习。为探究其实际应用效果,采用元分析方法,对国内外31项实验与准实验研究结果进行分析,发现:(1)教学代理能够促进学生学习,且对认知学习的促进效果强于非认知学习;(2)教学代理能够提升学生学习兴趣和学习效率;(3)教学代理能在一定程度上降低认知负荷,但不利于学习者心理努力及自我效能感的发展;(4)教学代理的应用效果受调节变量(学段、媒介组合方式、代理形象、代理手势)的影响而存在差异。因此,今后在教学代理的研究、设计与应用中,应强调其非认知效果,重视学生真实学习表现;设计多样化教学代理,兼顾学生认知基础及使用期待;注重调节变量优势,发挥教学代理的积极效力。

Pedagogical agents are virtual characters designed by integrating language and visual cues in a multimedia learning environment,and it is hoped to facilitate students'learning by providing guidance and feedback in interactive teaching.In order to explore the effect of its practical application,a meta-analysis method was adopted to analyze the results of 31 experimental and quasi-experimental studies at home and abroad,and finds that:(1)Pedagogical agents can promote students'learning,and have a stronger effect on cognitive learning than non-cognitive learning;(2)Pedagogical agents can improve students'learning interest and learning efficiency;(3)Pedagogical agents can reduce cognitive load to a certain extent,but are not conducive to the development of learners'mental effort and self-efficacy;(4)The application effect of pedagogical agents varies depending on the moderating variables(key stage,media combination method,agent image and agent gesture).Therefore,in the future research,design and application of pedagogical agents,emphasis should be placed on their non-cognitive effects and on students'authentic learning performance.Various pedagogical agents can be designed,taking into account the students'cognitive base and usage expectations.Attention should be paid to moderating variable advantage to bring into play the positive efficacy of pedagogical agents.