Enterprise AI Analysis
A causal machine learning approach for investigating learners' mental states through electroencephalography (EEG)
Authors: Anupama Jawale, Ruta Prabhu, Amiya Kumar Tripathy | Publication Date: 08 October 2025
Executive Summary
The non-invasive technique of electroencephalography (EEG) has become increasingly popular in recent years as a way to use computers to understand the complex psychological behavior of the human mind. Further fascinating insights into human psychology can be gained from the study of brain-computer interfaces using EEG. However, the extremely low band frequency and high sensitivity of the signals to noise and muscle activity make it challenging to use EEG in brain contact successfully, resulting in complex computation and machine learning tasks. For two distinct scenarios, this work introduces a novel combination of statistical causal inference and machine learning to improve brain-computer interaction performance. A publicly accessible dataset in the first case categorizes confused learners' answers, revealing their degree of confusion with Massive Open Online Course stimuli, while the second case categorizes a learner's level of stress while watching videos with multiple stimuli. Weights of balance for non-EEG variables are determined using propensity score matching, and they are then integrated into the causal model using machine learning techniques for brain wave feature engineering. The suggested approach is validated using algorithms from SVM, Random Forest, Gradient Boosting for Additive Model, Bagged CART Model, and Bayesian Generalized Linear Model. LightGBM outperforms other models with SVM (≈96%), Gradient Boosting for Additive Model (≈91%), Bagged CART Model (≈97%), and Bayesian Generalized Linear Model (~58%) with an accuracy of 98–4%, according to the results. By directing confounding factors that are usually overlooked in conventional methods, Propensity Score Matching, which incorporates causal inference, the suggested method dramatically increases classification accuracy.
Key Metrics at a Glance
0
Classification Accuracy
Achieved with the causal machine learning approach using Propensity Score Matching.
0
Algorithms Outperformed
LightGBM outperformed SVM, Gradient Boosting, Bagged CART, and Bayesian GLM.
0
Causal Inference Impact
Propensity Score Matching dramatically increased classification accuracy by addressing confounding factors.
Strategic Takeaways for Your Enterprise
- Novel integration of causal inference with machine learning significantly boosts EEG classification accuracy.
- Propensity Score Matching is crucial for balancing covariates and enhancing performance in brain-computer interfaces.
- The approach is validated across distinct scenarios: learner confusion with MOOCs and stress levels during video stimuli.
- LightGBM emerged as the top-performing algorithm with 98.4% accuracy.
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
This paper falls under the Neuroscience category, exploring the intricate relationship between brain activity (EEG) and mental states, particularly confusion and stress in learners. It leverages advanced machine learning and causal inference techniques to accurately interpret neural signals, offering significant implications for AI applications in understanding and enhancing human cognition and learning.
Enterprise Process Flow
Dataset Loading & Preprocessing
→
Estimate Causal Effects with Propensity Score Matching
→
Assign Weights to the Dataset
→
Build Classification Model
→
Data Splitting for Cross-Validation
→
Grid Search for Hyperparameter Tuning
→
Retrieve Optimal Hyperparameters
0
Classification Accuracy
The proposed causal machine learning approach achieved a classification accuracy of 98.4%, significantly outperforming conventional methods by effectively directing confounding factors.
| Algorithm | Accuracy (Before PSM) | Accuracy (After PSM) |
|---|---|---|
| SVM | 54.94% | 97.45% |
| Random Forest | 49.49% | 98.99% |
| LightGBM | 49.54% | 99.48% |
| XGBoost | 49.44% | 99.48% |
| DecisionTree | 49.49% | 98.62% |
Impact on Educational Neuroscience
Context: The research directly addresses a fundamental challenge in educational neuroscience: accurately detecting mental states like confusion and stress with minimal EEG hardware.
Approach: By combining machine learning classifiers with causal modeling, the framework offers implications for AI-Brain-IT interfaces in learning and teaching, and early intervention strategies.
Results: The findings are pivotal for understanding learner engagement and cognitive load, enabling personalized educational content and timely support for students.
Advanced ROI Calculator
Estimate the potential return on investment for integrating causal machine learning into your operations.
Estimated Annual Savings
$0
Productive Hours Reclaimed Annually
0
Phased Implementation Roadmap
A typical project timeline for integrating this advanced causal machine learning approach into your enterprise.
Phase 1: Data Acquisition & Preprocessing
Gather raw EEG data and preprocess for noise reduction and feature extraction, focusing on frequency domain analysis (PSD, HNR, Spectral Entropy).
Duration: 2-4 WeeksPhase 2: Propensity Score Matching & Weighting
Apply PSM using logistic regression and CBPS to balance covariates, generate propensity scores, and assign weights to the dataset for causal inference.
Duration: 3-5 WeeksPhase 3: Model Training & Evaluation
Train and validate various machine learning models (LightGBM, SVM, RF, etc.) on the weighted dataset, perform hyperparameter tuning, and assess performance metrics.
Duration: 4-6 WeeksPhase 4: Causal Impact Validation & Deployment
Validate the causal impact of the approach on classification accuracy and prepare for potential deployment in AI-Brain-IT learning interfaces.
Duration: 2-3 WeeksReady to Transform Your Operations?
Discover how causal machine learning can unlock deeper insights and drive unprecedented accuracy in your enterprise.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat