Investigating the relationship between CSAT scores and prefrontal fNIRS signals during cognitive tasks using a...

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Intelligence

Background:

• Academic achievement is linked to cognitive abilities, necessitating research into predictive markers.
• Neuroimaging techniques like functional near-infrared spectroscopy (fNIRS) provide insights into brain activity and its relation to cognitive performance.
• Understanding the neural correlates of academic success is crucial for educational and cognitive research.

Purpose of the Study:

• To investigate the association between prefrontal brain activity, measured by fNIRS, and academic achievement (CSAT scores).
• To apply a novel quantum annealer (QA) algorithm for feature selection in fNIRS data to identify relevant cognitive tasks.
• To compare the performance of QA-based feature selection with classical machine learning algorithms like XGBoost.

Main Methods:

• Extracted twelve signal features from prefrontal fNIRS data across two time windows (10-s and 60-s).
• Utilized a quantum annealer for feature selection to identify cognitive tasks correlated with CSAT scores.
• Validated feature selection performance using XGBoost and multiple linear regression models, assessing model fitness (adjusted R ² ) and prediction error (RMSE).

Main Results:

• Quantum annealing demonstrated comparable performance to XGBoost for feature selection in fNIRS data.
• Identified significant correlations between specific cognitive tasks (verbal fluency, recognition, Corsi block tapping) and academic achievement.
• Found stronger associations between Tower of London and N-back tasks with higher CSAT scores.

Conclusions:

• Quantum annealing presents a promising approach for feature selection in neuroimaging data, particularly fNIRS.
• Specific cognitive tasks are demonstrably linked to academic achievement, offering potential avenues for assessment and intervention.
• This research opens new possibilities for exploring cognitive predictors of academic success using advanced computational methods.