Technology-Based Training

Abstract

Training is increasingly becoming technology-based, swapping classroom time and live instructors for distance learning, serious games, and simulation exercises. This presents both challenges and opportunities for tailoring training to accommodate differences in trainees' backgrounds, prior knowledge, and abilities. Our interest is in building comprehensive technology-based instructional environments that adapt to differences and changes in cognitive factors-experience, knowledge, skills, and attitudes. We describe a space of techniques for adapting instruction in terms of (1) the aspects of a student model that inform individualized instructional decision-making, and (2) the aspects of the instructional experience that can be adapted based on those factors. Student model elements include: (1a) information on experience extracted from background questionnaires, (1b) records of student exposure to instruction and exercise s, (1c) assessments of student performance during exercises, and (1d) overall student mastery estimates for system learning objectives. Instructional adaptations include: (2a) choices of didactic instruction, (2b) choices of exercise scenarios, (2c) choic es affecting selection and delivery of exercise performance hints and feedback, and (2d) choices controlling pedagogically significant behaviors of simulated agents within scenarios. We give examples from a problem-based learning environment intended to train U.S. Army Battle Captains on how to supervise current operations in battalion Tactical Operations Centers (TOCs). A prototype implementation provides a unified environment combining instructional presentations, a scenario-driven TOC simulation, and machinery for controlling simulation behavior, student assessment, and instructional interventions. We describe the student modeling and instructional control components, emphasizing the breadth of instructional adaptation supported. We highlight how a control rule language, in the context of the overall system, will enable experimentation with alternate adaptation strategies. Such an environment is an essential tool for establishing an empirical basis for guiding future deployment of adaptive instructional systems.