Exploration on Evaluation Methods Combining Psychological Algorithms and Deep Learning Models

In recent years, significant progress has been made in the interdisciplinary field of psychology and computer science. This article first explored the integration process of psychological theory and deep learning models, and then provided specific explanations for the mechanism extraction process of cognitive theory and behavioral theory, as well as the manifestation of cognitive bias in deep learning. Subsequently, a personalized and adaptive analysis method was designed by combining neural networks and backpropagation algorithms. Through two sets of simulation experiments, the following conclusions were drawn: compared to traditional methods, the improved analysis method combining psychology and deep learning models achieved a comprehensive average improvement of about 13.3% in prediction accuracy indicators, and an overall average improvement of 24.85% in emotional continuity assessment. This analysis method not only has high effectiveness in simulating human emotional and cognitive processes, but also provides a deeper and broader perspective for psychological research.