E-CODEM: Interactive Learning for Complex Decision Making

点击翻译按钮获取中文摘要

This project investigates how decision making in complex environments may be enhanced through behavioural training, intelligent tutoring and cognitive support using an interactive learning environment. The current proposal includes three phases. Phase 1 aims to investigate cognitive support requirements, i.e., to better characterize what makes complexity so difficult to deal with and behaviours responsible for poor outcomes. This phase involves human-in-the-loop experimentation using a simulated testbed and a systematic manipulation of various attributes of the complex system (factors of complexity). Results will allow deriving a quantitative model of the individual and behavioural factors that explain success/failure in these situations. In Phase 2, we will derive from that model a context-based prescriptive tool for improving decision making behaviours. The prescriptive tool will take the form of an intelligent tutor that provides corrective feedback on the decision maker's behaviour and include a cognitive support module to promote the use of more comprehensive strategies without overloading cognitive capacity. This second phase will involve the development and validation of a simulation-based training tool and procedure called E-CODEM (which stands for Computer-Based Training for Complex Decision Making). Phase 3 involves extending this investigation to a team context in order to characterize the teamwork challenges and identify key support requirements. This project will advance the scientific understanding of human training requirements for decision making in complex situations and test approaches (interactive learning environments, intelligent tutoring, cognitive support tools) to improve cognitive readiness. The prescriptive model and interactive learning environment combined with the decision support module will be valuable tools for researchers and practitioners in a variety of domains such as defense, security, education and management.