Psychological stress during Examination and its estimation by handwriting in answer script

2025-11-08

AI Summary

Conventional grading systems fail to capture students’ cognitive and affective states during high-stakes examinations, hindering objective, non-intrusive assessment of psychological stress. Method: We propose the first explainable handwriting-based stress analysis framework for academic forensics, integrating multimodal features—high-resolution handwriting image analysis with TrOCR-based text transcription and RoBERTa-derived emotional entropy computation—and introducing a five-model ensemble voting mechanism coupled with unsupervised anomaly detection to enhance robustness. Contribution/Results: Evaluated on real-world examination data, our framework reliably generates a numerical stress index, accurately identifying high-stress conditions and atypical test-taking behaviors. It ensures interpretability through transparent feature attribution and maintains practical deployability via lightweight, end-to-end processing. This work bridges affective computing and educational assessment, offering a scalable, privacy-preserving solution for real-time stress monitoring in academic settings.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Evaluation and Analysis
• Cognitive Modeling & Cognitive Systems: Affective Computing

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

This research explores the fusion of graphology and artificial intelligence to quantify psychological stress levels in students by analyzing their handwritten examination scripts. By leveraging Optical Character Recognition and transformer based sentiment analysis models, we present a data driven approach that transcends traditional grading systems, offering deeper insights into cognitive and emotional states during examinations. The system integrates high resolution image processing, TrOCR, and sentiment entropy fusion using RoBERTa based models to generate a numerical Stress Index. Our method achieves robustness through a five model voting mechanism and unsupervised anomaly detection, making it an innovative framework in academic forensics.

Problem

• Quantifying psychological stress through handwriting analysis
• Using AI to assess cognitive states during examinations
• Developing data-driven stress measurement beyond traditional grading

Innovation

• Fuses graphology with AI for stress analysis
• Uses TrOCR and RoBERTa models for sentiment fusion
• Implements voting mechanism and anomaly detection