Robert T. Hays

Simulation Fidelity as an Organizing Concept

Although fidelity has typically referred to training devices, the fidelity concept can be used to facilitate and improve the design of total training systems. Based on the literature and an analysis of the issues surrounding fidelity, a working definition of fidelity and a strategy for acquiring the appropriate empirical information necessary for making fidelity requirements decisions is provided. The fidelity definition and strategic approach is basic to understanding the available research and guidance in later chapters. Task analysis is an important first step in accumulating information for designing training systems. A survey of task analysis techniques shows that no one approach is sufficient to provide this information. Fidelity is used as the organizing concept to determine how to conduct task analyses and use task analysis outputs to make training system design decisions.

Research Issues in Training Device Design

The Army Research Institute (ARI) has an ongoing research program to systematically accumulate the data necessary to provide design guidance for the development of training simulators. As part of this research program, an extensive literature review was conducted. On the basis of this literature review, a series of research questions, which require empirical investigation was accumulated. This presentation is designed to bring these research issues to the largest possible audience. The goal of this paper is to sensitize researchers to the types of questions that have the highest payoff for the Army and the training community. The research questions are organized around a model of the Instructional Systems Development (ISD) process. Each ISD section is further subdivided into specific subareas where empirical research is needed.

Training Effectiveness Evaluation

Any training device or training system must be evaluated to insure that the training provided by the system is effective. As discussed in Chapter One, evaluation is a major phase of training system development and is the basis for determining when a training program requires modification, or when whole new training approaches are required. Some evaluations can also be used in selecting training approaches for new training programs. Much of the research to be discussed in subsequent chapters was designed to evaluate training device design and use. This chapter provides an overview of training effectiveness evaluation so that the subsequent discussions of research will be more meaningful to the reader. One should not, however, think of training effectiveness evaluation as a one-time effort. It is rather an ongoing process with the goal of constantly improving the effectiveness of the training system.

A Conceptual History of Training Fidelity

As new capabilities are provided by emerging technologies, the job of the training system developer becomes more difficult. How is the designer to determine what characteristics to incorporate into the design of new training devices and how should these devices be most effectively used in a training system? One traditional design method has been to determine the degree of fidelity necessary in the training device or training system. Unfortunately, there has never been a consensus on what fidelity means or on the training effects of various levels of fidelity.

Fidelity Research on Fault Diagnosis and Other Maintenance Tasks

We have summarized an organized research effort which sought to systematically specify the training effects of both the physical and functional aspects of fidelity for different tasks. Other research efforts have also investigated the training effects of fidelity. In one series of experiments, low physical/medium to high functional fidelity computer simulations were used to investigate human decision making in fault diagnoses. These tasks have high cognitive requirements that are prime candidates for training with high functional fidelity.

Training for Cognitive Skills

The research discussed previously dealt primarily with tasks that required interaction with equipment. Training for those tasks has, for the most part, used some form of training device. The material to be discussed in this chapter involves cognitive tasks and often does not require any form of equipment. Nevertheless, many of these tasks have used computers and other types of training devices primarily to train the functional requirements of those tasks. Other training approaches, such as role playing, emphasize the functional requirements of the persons involved in the tasks. It is possible and important to analyze the training requirements for these cognitive tasks from a fidelity perspective. The emphasis in these fidelity analyses typically will center on the functional fidelity required for training rather than the physical characteristics of the task. The areas of cognitive skills training discussed include management training, logistics training, health care training, and training for international relations.

A History of Device-based Training Systems

The modern training device did not arise full grown, like Athena. Rather, it has a long history of developmental applications. The capabilities of modern training devices are the result of three major historical developments: the teaching machines of the 1920’s and 1930’s, the military training devices developed during World War II, and the programmed learning machines, made popular by B. F. Skinner during the 1950’s. The modern computer now offers a means of incorporating the lessons learned in each of these training approaches into training devices which enhance the effectiveness of training. Computer technology also provides a basis for new training options and approaches. These recent developments are based in a long history of training devices. If training system designers are to take advantage of the wide variety of training options afforded by computers, it is helpful to understand how these options evolved historically.

Operator and Maintenance Training Research

Research on training for non-aviation systems has not been as extensive as research in flight simulation. Non-aviation research has focused on two major areas: operator training, and maintenance training. It should be noted that to a great extent these two areas overlap. For example, a maintenance task usually requires some knowledge of equipment operation and operators may be required to conduct maintenance procedures to determine equipment operational readiness.

Current Trends in Training Systems: Computer-based Instruction, Embedded Training, and Artificial Intelligence

It should be evident from the preceding chapters that modern training developers have increasingly used computers for delivery and management of training. This reliance has led to three major new trends in training systems: computer-based instruction, embedded training and artificial intelligence. The first trend, the use of computers, has been growing ever since the early days of flight simulation and programmed learning (see Chapter four). New aspects of computer-based instruction are due to the increasing power, availability, and portability of computers. The second trend, embedded training, uses the computers that are part of operational equipment to deliver training. This is a very new approach and is still evolving. Artificial intelligence is a development based in the cognitive and computer sciences. It takes advantage of the memory capacity of new computers to archive and structure the knowledge of experts in a form that can be easily used for both training and job aiding. The three trends are not mutually exclusive and considerable overlap exists in training strategies using these techniques.

Definitions and Problems in Training System Design

Designers of training systems have a difficult job deciding what characteristics they should incorporate into the training media used in those systems. Their job is made more difficult by the variety of terms used to describe training equipment. The media and techniques that may be used to assist in the delivery of training are diverse. Many terms are used by training system developers to define these media and techniques. Before proceeding, a standard set of definitions will be provided for the terms used in the rest of the book.