The role of Indoor Environmental Quality (IEQ) factors in shaping student behavior and emotional states in the classroom, which have been observed as potentially diminishing performance, necessitates objective and continuous assessment to overcome the limitations of subjective methods. This study addressed this need by utilizing a case study approach. We deployed an AI-powered behavioral observation system to anonymously estimate aggregate student behavior metrics (Engagement, Attention, Interaction) in real-time, synchronized with data collected from a custom-built multi-sensor device monitoring IEQ factors, including temperature, humidity, equivalent carbon dioxide (eCO₂), total volatile organic compounds (TVOCs), air quality index (AQI), light variations, and oxygen volume (O₂). Comprehensive statistical and causality analyses included nonparametric correlations, Cross-Correlation Function (CCF) analyses to assess lagged effects, Time-Varying Granger Causality (TV-GC) tests, and categorical analysis with Chi-squared tests. The results revealed that thermal and humidity extremes correlate with increased behavioral volatility. Temperature is the most consistent predictor of student attention; Chi-squared and violin plot analyses demonstrated that attention levels are significantly higher at slightly lower temperatures, specifically below 30. 9^ C, within the reported range 29. 62 - 31. 35^ C. Besides, a significant association between humidity and attention was observed, although it was significant at the =0. 05 level rather than the more stringent =0. 01 threshold applied to core findings. Additionally, the study identified a critical TV-GC relationship between O₂ volume and engagement, pinpointing specific causal bursts that global correlation measures failed to capture. Standard CCF analyses suggested that lower light levels may be associated with higher interaction levels; however, this pattern was not statistically significant after pre-whitening and bootstrapping the CCF, nor was it supported by the TV-GC analyses. These findings advocate for responsive, automated classroom systems that dynamically adjust IEQ parameters to synchronize with the temporal demands of the learning process.
Mena-Martinez et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: