Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Technology and Training Training personnel is a major task for the U.S. military: when people are not engaged in military operations, their most important activity is training. Military personnel are predominantly high school graduates in the enlisted force and college graduates in the officer corps. Although a few military jobs have civilian counterparts and can be selected rather than trained (e.g., doctors, lawyers, and aircraft mechanics), most service jobs have few civilian counterparts, most obviously combat skills. In addition, the military promotes from within and develops its own leaders rather than selecting them. The training of service men and women throughout their military careers is expensive, both in terms of dollars and personnel. Since the military (like industry) pays its members to be trained, a reduction in training time results in savings of time and money. CURRENT TRAINING APPROACHES Training is deeply embedded in military culture as a core mission (Bratton-Jeffrey, Hoffman, and Jeffrey, 2007). Today, the training needs of the military are expanding from the need to teach skills in isolation, such as how to trouble-shoot an electronic device, to the need to teach problem- solving strategies and concepts in the context of complex and ever-changing task environments, such as how to negotiate in different cultural contexts or how to rapidly integrate information from multiple sources to make an on-the-ground decision (van Merrienboer, 2007). Fortunately, recent advances in educational technology offer highly promising ways to meet this new kind of instructional need, but basic research and research-based 39
40 HUMAN BEHAVIOR IN MILITARY CONTEXTS theory are needed in order to determine how to use educational technology productively (Mayer, 2005). This will require both short- and long-term research approaches. The military has successfully implemented an instructional systems design approach to guide the development of training of isolated skills, but there are indications that this classic approach is not well suited to the emerging new demands for training of strategies and concepts in complex contexts (Reiser, 2007). Research is needed to support the development of both a research base and a research-based theory of instructional design for these new kinds of learning, which include decision making, informa- tion integration, communication in cultural context, and problem solving in unexpected situations. FUTURE APPROACHES In future military training, one can envision learners sitting in front of computer screens at school, home, or a job site and having the opportunity to learn with the help of an on-screen agent who offers useful job-related practice tasks within realistic simulations. One can also envision military personnel being able to play serious games that promote learning. Graesser and King (in this volume) describe 10 advanced learning environments that hold potential for technology-based training: computer-based train- ing, multimedia training, interactive simulation, hypertext and hypermedia, intelligent tutoring systems, inquiry-based information retrieval, animated pedagogical agents, virtual environments with agents, serious games, and computer-supported collaborative learning. Such technology-based environ- ments can support the learning of both individuals and teams. Advances in computer and information communication technology have potential for greatly increasing the efficiency and effectiveness of training, and there is encouraging preliminary evidence of the efficiency and effectiveness of technological approaches under appropriate conditions (Andrews, Nullmeyer, Good, and Fitzgerald, in press; Breuer, Molkenthin, and Tennyson, 2006; Chipman, 2006; Clark and Mayer, 2003; Cuevas, Fiore, Bowers, and Salas, 2004; Fletcher, 2003; Jonassen, 2004; Mayer, 2001, 2005; Moreno, 2006; OâNeil, 2005; OâNeil and Perez, 2003, 2006; Pearson, Ferdig, Blomeyer, and Moran, 2005; Wulfeck and Wetzel-Smith, in press). In general, reviews of the literature with respect to efficiency conservatively indicate a 30 percent reduction in training time when the same objectives are taught on computers in comparison with conventional instruction (Fletcher, 2003; Kulik, 1994; Sitzmann, Kraiger, Stewart, and Wisher, 2006). With respect to effectiveness, the critical issues are the instructional strategies and assessments embedded in the computer-based systems, not the
TECHNOLOGY AND TRAINING 41 medium per se (Clark, 2001; Kirschner, Sweller, and Clark, 2006; Sitzmann et al., 2006). When appropriate instructional strategies have been embed- ded in technology-based training systems, the systems have been shown to be 19 percent more effective than conventional instruction for teaching declarative knowledge (Sitzmann et al., 2006) and to have average effect sizes of 1.05 for modern intelligent tutoring systems (Fletcher, 2003). The latter finding represents an improvement of performance of 50th percentile students to the 85th percentile. Findings also show that some forms of tech- nological support, particularly the use of computer simulations for training, offer the opportunity to train skills safely, efficiently, and effectively that are either impossible or very expensive to train without such support (e.g., pilot training, combat skills, medical skills; OâNeil and Andrews, 2000). THE SCIENCE OF LEARNING AND THE SCIENCE OF INSTRUCTION Although the hardware and software technologies for implementing advanced learning environments are being developed, the work often takes place without an understanding of how people learn. To complement hard- ware and software development efforts in creating new training technolo- gies, basic research on how to use such training technologies to improve human learning is necessary. Currently, there is a small research base on the topic, but a serious investment of research support could significantly increase its pace and usefulness for the military in the near term. Advances in educational technology are outpacing advances in an un- derlying science of learning with technology in part because the field is vendor driven, not science driven. Thus, decisions about how to design technology-based training are often based on intuitions and opinions of persons with technological development skills rather than on research evi- dence and a research-based theory of how people learn. As a result, training programs may not reach the optimal levels of effectiveness and efficiency. For example, there is only limited scientific research evidence that computer games facilitate the learning of adults (OâNeil and Fisher, 2004; OâNeil, Wainess, and Baker, 2005), yet they continue to be promoted and used in the military context. Given the centrality of training for the military, it is critical that technology-based training is based on research evidence and research-based theory. Training programs should be based on an understanding of how people learn and how instructional methods affect learning. It is tempting to focus on the tremendous technological advances in education, including web- based training, without sufficient attention to the people who need to be trained. For example, in taking a technology-centered approach, instruc- tional designers begin with a cutting-edge technology and try to build learn-
42 HUMAN BEHAVIOR IN MILITARY CONTEXTS ing environments for users rather than starting with the user and trying to determine what technology can meet their needs. In his review of the his- tory of educational technology during the 20th century, Cuban (1986) has shown that cutting-edge technologies of each era have failed to have much impact on improving educationâincluding motion pictures in the 1920s, educational radio in the 1930s and 1940s, educational television in the 1950s, and computer-based programmed instruction in the 1960s. In con- trast, in taking a learner-centered approach, instructional designers begin with an understanding of how people learn and seek to use technology as a cognitive tool to aid learning (Mayer, 2001; Sweller, in press). For example, technology-supported instruction that was based on learner-centered design principles (in this case, the cognitive theory of multimedia learning), test performance improved by 0.6 to 1.3 standard deviations as compared to conventional practice (Mayer, 2001, 2005). Research is needed to develop a theory of learningâor science of learningâthat is relevant to learning with technology, particularly in the military. How do people learn from words (such as spoken or printed text) and graphics (such as illustrations, photos, animation, or video)? How do people learn from on-screen agents? How does interactivity influence learning? How do people use and learn self-regulatory skills in technology- based environments? These are the kinds of basic questions that need to be addressed in building a science of learning with technology that can benefit the training needs of the military. Training programs should also be based on an understanding of how instructional methods affect learning. In spite of stunning advances in com- puter and information technology, the way to incorporate these technolo- gies in the service of human learning requires behavioral, not technological, research. As noted above, there is broad consensus that learning results from instructional methods rather than instructional media (Clark, 2001). Using a particular mediumâsuch as a computer-based multimedia lesson or a serious gameâdoes not ensure an improvement in learning. Rather, the medium needs to follow from knowledge about instruction and learn- ing. Research can identify the effects of technology-supported instructional methods and clarify the conditions under which the methods are effective (e.g., Mayer, 2001, 2005; Sweller, in press). One topic that crosses disciplin- ary boundaries and is of considerable significance to the military is the role of feedback in computer-based simulations used for training and assessment of teams. This type of feedback is called after-event review (Ellis, Mendel, and Nir, 2006) or after-action review in the military (Meliza and Goldberg, in press). The goal of research on such issues should be the establishment of a set of research-based principles for how to design technology-based training to meet military needs. Training programs should incorporate specification of the knowledge to
TECHNOLOGY AND TRAINING 43 be learned and assessment of what learners know. Many technology-based training applications focus on increasing learnersâ cognitive knowledge (National Research Council, 2001), but motivational or affective learning is also of importance. For example, much of the assumed effectiveness of serious games is attributed to motivational effects. There is, to date, very limited research on assessing the cognition and motivation for individuals, teams, and organizations. Much of the assessment in training technology relies on single formats such as multiple-choice testing, which is most useful for recalling facts (Anderson et al., 2001), not motivation or other affective components of learning. Some of the most interesting technological training applicationsâsuch as simulation, games, team training, and after-action reviewsâdo not include assessments of knowledge that permit real-time diagnosis of learning and prescriptions for improvement. Currently, the most frequently used assessments are think-aloud pro- tocols, behavioral observation systems for teamwork training, and face validity for simulation applications. Yet, none of these techniques permits real-time feedback. And, these approaches seldom address issues of reli- ability or validity. What is needed is a model-based approach to assessment and a psychometrics of simulation. A model-based assessment approach (Baker and OâNeil, 2006; Baker and Mayer, 1999) would focus on a model of learning, not a model of content. It would (a) draw on elements from learning and assessment scientific knowledge, (b) be empirically developed, (c) have both domain-independent and domain-dependent aspects, (d) have reusable components that would result in time and cost savings, and (e) give evidence of technical quality obtained. Multiple purposes of assessment would also be supported (e.g., program evaluation, system monitoring, in- dividual/team certification, selection and classification, individual and team diagnosis and prescription). A psychometrics of simulation approach would deal with traditional psychometric issues in the simulation domain (e.g., difficulty, norming and equating, reliability, and validity). Finally, affective and motivational models of learning, as well as social capital ideas for or- ganizational improvement, would drive new assessment methodology. TECHNOLOGY TO MEET MILITARY NEEDS Basic research on technology and training can contribute to solving problems for each of the five major needs for the military identified in this report: leadership, training, personnel, social interactions, and organiza- tional structure. It can improve instruction in leadership skills, including computer-based simulations to enable decision-making practice in simu- lated settings. Promising approaches in leadership training include the AXL, ELECT, and Vector projects (see Zbylut, Metcalf, Kim, Hill, Rocher, and Vowels, 2007; Hill, Gordon, and Kim, 2004; Hill, Belanich, Core, Lane,
44 HUMAN BEHAVIOR IN MILITARY CONTEXTS Dixon, Forbell, Kim, and Hart, 2006; Zachary, Le Mentec, Miller, Read, and Thomas-Meyers, 2005). Clearly, basic research on technology and training can improve the ef- fectiveness and efficiency of training in the military, even in the near term. In particular, the committee believes that the application of scientifically tested instructional principles can improve performance by at least 0.8 standard deviations, which is considered a large effect. We also expect that technology-supported testing itself can enhance learning, as reflected in the test enhanced learning effect, in which taking repeated tests improves student learning (Roediger and Karpicke, 2006). With respect to personnel, basic research on technology and training is needed to determine the role of individual differences in learning, includ- ing how to design adaptive learning environments based on assessments of learnersâ characteristics and progress during instructional lesson. Basic research on technology and training is also needed to determine how to use social cues and group-based methods to foster better learning and also to determine how best to teach social interaction skills, such as negotiation. Finally, basic research on technology and training is needed to determine how to embed training programs in existing organizational structures and practices. This type of research will yield results in both the short and long terms. Overall, the goal of basic behavioral research on technology and train- ing is to create a science of learning (National Research Council, 1999) that is relevant to issues in military training, builds a research base that can guide instructional design of technology-based training in the military, and identifies and assesses instructional methods that have large effects on learning. Basic behavioral research on technology and training will enable military trainers to take an evidence-based approach to designing training for individuals and for teams. TECHNOLOGY AND OTHER RECOMMENDED RESEARCH Knowledge about technology and training interacts with each of the other research topics identified in this report: intercultural competence, teams and complex environments, nonverbal communication, emotion, and neurophysiology. For example, the appropriate design of collaborative learning environments depends on research on intercultural competence training programs aimed at teaching intercultural skills, and it is similarly dependent on research on teams operating in complex environments (Salas and Priest, 2005). Research on nonverbal processes is directly relevant to the design of on-screen agents, including the role of gesture, expressions, and voice. Understanding affective processing in learning, including ways
TECHNOLOGY AND TRAINING 45 in which learners react to frustration or contentment during learning, is useful in designing training programs. Neurobiological markers can be used to measure physiological states of learners during the learning process (see Chapter 7) and can be particularly useful in understanding the role of stress in learning and performance.