Bridging AI and Psychology: A RAG-Driven Approach to Inhibitory Control and ADHD Interventions

This study explores the integration of Retrieval-Augmented Generation (RAG) into Artificial Intelligence (AI) systems to enhance psychological support, particularly in ADHD interventions and Inhibitory Control assessment. To achieve this, the GPT-4o model was used in conjunction with structured databases of therapeutic dialogues to improve mental health intervention, cognitive evaluation, and professional training. In addition, prompt engineering was used as an inference technique through text-based instructions, and an experimental evaluation was conducted using a predefined set of prompts designed to measure contextual consistency, ethical compliance, and technical accuracy. The results indicate that integrating retrieval mechanisms improves AI-generated responses by providing more contextually relevant and accurate information, demonstrating the model’s enhanced capability to process complex contextual queries.