Method for assessing creativity psychology based on artificial intelligence and reinforcement learning model

Abstract

The application provides a creativity psychological evaluation method based on artificial intelligence and a reinforcement learning model, and relates to the technical field of creativity psychological evaluation.The application calculates attention entropy values through eye movement data, combines standardized processing of frontal lobe Gamma band electroencephalogram signals, constructs a multi-dimensional feature vector, realizes quantitative representation of the collaborative state of attention distribution and brain nerve activity, breaks through the one-sidedness of traditional single signal evaluation, simultaneously, a reinforcement learning task decision engine dynamically matches evaluation tasks according to real-time cognitive states, so that the task type is highly adapted to individual cognitive characteristics, ensures that more real creativity performance is captured, a reinforcement learning model driven by a multi-dimensional reward mechanism is continuously optimized, the evaluation not only reflects surface behavior results, but also excavates the internal rules of the cognitive process, and provides a quantitative basis for creativity psychological mechanism research.

Images