Adding Directions to Simulation-Based Training
Marieke M. M. Peeters, Karel Van den Bosch, John-Jules Meyer, Neerincx · TNO Repository · 2010
Current views on education emphasize that learning involves more than the transmission of knowledge from teacher to student. Students should actively engage in a learning environment that offers them the opportunity to really experience the learning task and the outcomes of certain choices or actions, preferably in teams (1). They do this by participating in functional and realistic learning tasks (2,3,4,5,6). Learning tasks should be designed in such a fashion that they systematically call upon all knowledge and skills formulated in the training objectives (7). Moreover, these tasks need to be situated within a real and relevant context or environment. Scenario-based training (SBT) meets these requirements (8). In SBT, trainees prepare, execute, and evaluate real tasks in a simulated environment. Training within simulated environments is beneficial over training in real situations, because it reduces the risks involved and because it offers the possibility to exhibit a certain amount of control over the training. Simulated environments may vary in their fidelity. They may be the real task environment, a high-fidelity simulation, but may very well be a highly symbolic representation of the real task. What they typically have in common is that they require the effort of other people besides the trainee, to deliver training. Staff or team members may for example be needed to play the people with whom the trainee has to interact. In addition, an instructor is needed to evaluate the performance of the trainee and to deliver feedback. Recent developments in the gaming industry give rise to new opportunities to implement SBT in virtual environments in the form of serious games. The additional advantage of this medium is that, with the use of Artificial Intelligence techniques, the other players (e.g. team members, opponents, instructor) can be played by intelligent agents (also called virtual characters). Examples of such agent-based serious training games already exist (9). Combining agent technology and gaming will continue and become more advanced, yielding enormous benefits for training. Trainees can practice more often (and thus achieve higher competence levels), requiring no or fewer resources, which results in lower costs. However, in order to facilitate optimal progression of the trainee within the training, agent-based training needs to be (learning) goal-directed. A director is necessary to create learning opportunities within the game, thereby ensuring that the scenario develops in service of the learning goals. This director must be able to intervene in the events or agent-behaviours within the game in order to bring about the desired situations associated with those learning goals. The assumption here is that the quality of training improves by adding a director. Trainees should experience directed scenarios as more motivating and better adapted to their skill level. To investigate whether this is so, we conducted the present study.