Enhancing students' self-directed learning through emotion recognition systems and intelligent collaboration platforms: Evidence from AI-enabled primary schools in Chengdu, China

Ziwei Ouyang, Alan Robert White and Supot Rattanapun

African Educational Research Journal
Published: September 22 2026
Volume 14, Issue 3
Pages 760-771
DOI: https://doi.org/10.5281/zenodo.22897576

Abstract

The relationships among students' perceived Emotion Recognition Systems (ERS), Intelligent Collaboration Platforms (ICP) and Self-Directed Learning Ability (SLA) were explored in six private primary schools in Chengdu, China. A questionnaire survey was carried out, and 402 Grade 4 - 6 students were surveyed cross-sectionally. Descriptive statistics, reliability analysis, Pearson correlations, regression analyses and bootstrapped mediation analysis were performed. ERS and ICP were positively associated with SLA ((β = .754) and (β = .812), respectively, (p < .001)). ICP was also positively associated with ERS (β = .680, p < .001). In the joint regression model, ERS (β = .375, p < .001) and ICP (β = .557, p < .001) remained positively associated with SLA, explaining 73.5% of its variance (R2 = .735). The bootstrapped indirect association of ICP with SLA through ERS was significant (ab = .144, Boot SE = .026, 95% CI [.095, .196]). The results indicate that the positive attitudes of students toward CLD are correlated with their SLL. The results must be considered as associations and not as causal effects due to the cross-sectional design of the study. Longitudinal and experimental studies are needed in the future that will look at the temporal relationship, quality of implementation, and data and privacy protection for children.

Keywords: AI in education, emotion recognition systems, intelligent collaboration platforms, primary education, self-directed learning, self-regulated learning.

Full Text PDF

This article is published under the terms of the Creative Commons Attribution License 4.0