This research explores how Multimodal Large Language Models can augment human reasoning for visual perception analysis, drawing on principles from psychology and cognitive science.

• Integrates established cognitive science principles to guide AI visual perception capabilities
• Provides interpretable analysis of complex visual information
• Creates frameworks for human-AI collaboration in visual reasoning tasks
• Offers applications across educational settings with learning-focused implementations

For education professionals, this research offers powerful tools to enhance visual learning, support diverse learning styles, and develop more effective teaching methodologies around visual content comprehension.

Multimodal LLM Augmented Reasoning for Interpretable Visual Perception Analysis