Tracking children’s handwriting learning process using EEG: A system development and validation

Background: Handwriting is a fundamental component of school education, especially in the early phases. It is thus important to develop a scientific approach to refining the cost-efficiency of handwriting learning in young children. 

Aims: In this study, we proposed that an integration of behavioral and neural data tracking during the real-time process of handwriting learning can reveal the learning process and thus inform the design of handwriting training. The main rationale is that the two complementary information channels can reveal the dynamic learning curve during repetitive practices. 

Sample: Participants were 50 typically developing schoolchildren (aged 6–7) who had limited orthographical knowledge of Chinese characters and handwriting training. 

Methods: Synchronized EEG and handwriting kinematics data were collected when participants were performing a Chinese character copying task. Six unfamiliar Chinese characters at three different complexity levels were selected, and the participants copied each character repetitively for 15 times. The representative behavioral and neural features related to handwriting fluency were quantified, including writing duration, velocity, and event-related potentials (ERPs) extracted from the copying process of each character. By applying linear mixed models (LMMs), we found significant behavioral improvement and neural adaptation effect across the repetitive copying practices; and the observed behavioral and neural effects showed a systematic pattern of dependence on character complexity. 

Conclusions: These findings validated the cognitive association of the non-invasively collected neural signals, demonstrated the feasibility of combing behavioral and neural signals to track the process of children's handwriting learning, and informed the design of handwriting training programs. Copyright © 2023 The Authors. Published by Elsevier Ltd.