Nowadays with a growing demand of tools and techniques that improve public education,constant learning and industrial training a radical change from the current content-oriented e-learning solutions to user-centered, interactive and collaborative model oflearning is required.The new learning model should change the role of the learner from a passive recorder ofdata and information to a constructor of his knowledge via allowing him to participateinto sophisticated interactions (e.
g. learning by doing, educational games, simulationenvironments).Cognitive interactive agent technology represents the implementation of the new learningparadigm. This technology has the common advantages of distributed and concurrentproblem solving and has the supplementary advantage of complex patterns of interactions.Among the underlying distinguishing features are flexibility and high-level nature ofthese interactions, which support the power of the novice learning paradigm.The potential of game-based learning across a wide range of application fields and domainsincluding but not limited to education, business, military and medical applications wasproven.
One of the possible application areas is the training of the negotiation skills.The Master thesis project implements a unique approach of collaborative creation ofa cognitive interactive agent by the different parties like game designers, end-users(educational and business ones) and players. The project provides an easy creation ofgame-based training environments situated in various gaming scenarios,thereby addressingthe increasing demand for computational intelligence algorithms for learner’s behaviour,motivation, performance monitoring and analysis that developers and training providersare facing.
Thus the project provides a negotiators training environment for a variousdomains. An agent designed to use Theory of Mind for learners and play a role of thepractice partner for complex negotiations. The agent models Theory of Mind to makedecisions about its own actions and the behavior of the player. This serves as a source ofthe agent’s explicit feedback on the opponent’s strategy and behavior.
In this paper wealso discuss several experiments and evaluation results to show the effectiveness of thepresented agent as credible opponent for improving Theory of Mind abilities in learnersand, as a result, increase negotiation performance.
