多媒体学习中表征认知负荷的眼动指标研究

Research on Eye-tracking Indicators for Cognitive Load Representation in Multimedia Learning

通过眼动设备进行认知负荷的测量已成为认知负荷测量的重要方法,尽管相当多的研究依赖眼动仪输出的测量数据进行认知负荷表征,但究竟哪些眼动数据能够准确表征学习者认知负荷状态并没有明确的指标。研究采用循证研究的元分析方法,对62篇国内外实验研究文献进行量化分析,系统检验认知负荷对眼动各项指标的影响,探究可以表征多媒体学习中认知负荷状态的眼动指标。研究发现:(1)纳入研究的总效应值为0.547(k=318),表明整体上眼动指标受认知负荷影响较大;(2)将不同测量指标作为调节变量分析,发现瞳孔直径等15项具有较大效应值的眼动指标能够较好地表征学习者的认知负荷状态;(3)在其他调节变量分析中,发现低于60Hz的采样率更适用于多媒体学习中认知负荷的测量,而知识性质和认知负荷来源不存在调节效应。

Measuring cognitive load through eye-tracking devices has become an important method for cognitive load measurement.Although many studies have relied on the measurement data output by eye-tracking devices to characterize cognitive load,there are no clear indicators of which eye-tracking data can accurately characterize the cognitive load status of learners.This study used a meta-analysis method of evidence-based research to quantitatively analyze 62 experimental studies at home and abroad,to systematically examine the impact of cognitive load on eye-tracking indicators and explore eye-tracking indicators that can characterize cognitive load in multimedia learning.It is found that:(1)the overall effect size of the samples included in the study is 0.547(k=318),indicating that eye-tracking indicators are significantly affected by cognitive load;(2)analyzing the different measurement indicators as moderating variables,it is found that 15 eye-tracking indicators with significant effect sizes,such as pupil diameter,can better characterize learners'cognitive load status;(3)in the analysis of other moderating variable,it is found that the eye-tracking device with a sampling rate below 60Hz is more suitable for the measurement of cognitive load in multimedia learning,while there is no moderating effect for the nature of knowledge and the source of cognitive load.