Modeling Self-Efficacy as a Dynamic Cognitive Process with the Computational-Unified Learning Model (C-ULM)
Self-efficacy is a person's subjective confidence in their capability of effectively executing behaviors and actions including problem solving. Research has shown it to be one of the most powerful motivators of human action and strongest predictors of performance across a variety of domains. This paper reports on the computational modeling of self-efficacy based on principles derived from the Unified Learning Model (ULM) as instantiated in the multi-agent Computational ULM (C-ULM). The C-ULM simulation is unique in tying self-efficacy directly to the evolution of knowledge itself and in dynamically updating self-efficacy at each step during learning and task attempts. Self-efficacy beliefs have been associated with neural and brain level cognitive processes. Because C-ULM models statistical learning consistent with neural plasticity, the C-ULM simulation provides a model of self-efficacy that is more compatible with neural and brain level instantiation. Results from simulations of self-efficacy evolution due to teaching and learning, task feedback, and knowledge decay are presented. Implications for research into human motivation and learning, cognitive informatics, and cognitive computing are discussed.