摘要
学习场域中的认知负荷测评存在缺乏过程性数据、测评维度单一、测评精确性不足等问题.多模态数据融合与分析技术从多维时空角度揭示认知负荷的表征机制,将认知负荷测量问题置于数据驱动范式中重新审视,有助于形成理解学习认知和相关规律的更为有效的方法.基于此,文章从认知负荷测评研究的现状出发,梳理分析了融合生理数据驱动认知负荷测评的动因.深入分析了相关研究开展所涉及的认知负荷可计算、认知负荷表征模型的可解释、认知负荷要素权重的计算等关键问题,明晰了融合生理数据驱动认知负荷测评的多维性、过程性、精确性特征,并在此基础上基于"理论模型—数据采集—模型构建—计算分析与模式识别"思路构建了教育生理计算框架.
Abstract
Cognitive load assessment in the learning field suffers from the problems of lack of process data,single dimension of assessment,and insufficient accuracy of assessment.Multimodal data fusion and analysis technology reveals the representation mechanism of cognitive load from multidimensional spatial and temporal scales,and re-examines the cognitive load measurement problem in a data-driven paradigm,which can help to form a more effective method to understand the learning cognition and related laws.Based on this,the article analyzes the motivations for integrating physiological data to drive cognitive load measurement from the overview of the current state of cognitive load measurement research.It also analyzes the key issues involved in the development of related research,such as the computability of cognitive load,the interpretability of cognitive load representation model,and the computation of cognitive load element weights,and clarifies the multidimensional,process,and accuracy characteristics of the fusion of physiological data-driven cognitive load assessment.Finally,the article analyses and explains the educational physiological computational framework based on the idea of"theoretical model—data collection—model construction—computational analysis and pattern recognition".
NETL
NSTL
万方数据