Neuroscience-Ai Framework For Predicting Human Decision-Making Under Uncertainty

Abstract

Understanding how humans make decisions under uncertainty is a longstanding challenge in cognitive neuroscience and artificial intelligence. This research proposes a Neuroscience-AI integrated framework that models and predicts human decision-making behavior by combining neurophysiological insights with machine learning architectures. The proposed framework leverages neural data—including electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) signals—to identify brain activation patterns associated with risk perception, reward evaluation, and cognitive bias. These features are processed using advanced deep learning models, such as recurrent neural networks (RNNs) and attention-based transformers, LSTM Framework to capture temporal dynamics and contextual dependencies in decision-making processes. Experimental results demonstrate that the framework achieves superior predictive accuracy compared to traditional behavioral models, providing interpretable mappings between neural activity and decision outcomes. The integration of neuroscience-driven features with AI algorithms enhances the transparency, adaptability, and biological plausibility of computational predictions. This interdisciplinary approach opens new pathways for applications in behavioral economics, neuropsychology, adaptive human-computer interaction, and cognitive diagnostics, offering a robust foundation for understanding and forecasting human decisions under uncertain and dynamic conditions.