This paper illustrates the dangers in applying human visual system metrics such as Johnson’s Criteria and the Snellen Chart to visual system design for dynamic, high workload platforms such as fighter aircraft. Johnson’s Criteria is an accepted standard defining the number of optical line pairs required for visual detection and identification of targets displayed by scanned imaging devices. The Snellen Chart contains characters of the alphabet with a calibrated set of sizes. The ability to distinguish characters of a specific size from a distance of 20 feet identifies normal 20/20 vision. This paper is derived from a study that established operational visual cueing requirements for typical missions of a multi-role fighter aircraft. After the visual cueing requirements were derived from an operational perspective, the results were used as a basis for defining a set of visual system performance requirements. When Johnson’s Criteria were applied, the results did not correlate as expected, suggesting additional factors must be considered for detection and identification of observable entities such as other aircraft, ground targets or airfields in a dynamic, high workload environment.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

Building Situation Awareness: Impact of Symbol Type and Readout Location

John Winters, Greg Hildebrand, and Lisa Chavez

Basic Commerce and Industries, Inc.

Dahlgren, Virginia

Dr. Daniel Wallace

Naval Surface Warfare Center

Dahlgren, Virginia

For many of their tasks, Navy CIC operators sequentially select or otherwise browse tracks or contacts of interest. Both the symbology used and the design of the text readouts associated with selected tracks can impact performance. A set of four tasks was used to quantitatively compare operator performance with three different symbol sets - colorized Naval Tactical Data System (NTDS) symbols and modified MIL-STD-2525B symbols in a top-down plan view display, and realistic 3D icons in a perspective view display. For the 2D displays, two locations for text readout were assessed - adjacent to the track in question and in a fixed location. The 3D displays were found to provide no benefit for altitude estimation tasks, and comparison of results across tasks indicates that realistic aircraft symbols in a 3D display are harder to search than are 2D symbols. Including within the symbols more information of relevance to the operator improved response times by 40% to 60% in two different tasks. No consistent differences in performance were found for text readout location.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

VISUAL MOMENTUM: DESIGNING FOR AN EFFECTIVE VIRTUAL ENVIRONMENT FOR MILITARY INTELLIGENCE

Drs. Ir. Jaap Jan van den Bosch and Wilfred Janssen

Atos Origin Nederland B.V. / Technical Automation

Utrecht, The Netherlands

This paper reports on the progress and first results of a research project that has the objective to demonstrate the feasibility and use of virtual environments (VEs) in military decision-making. This technology is developed for the intelligence section of an army battalion or brigade staff. The main purpose is to show how the decision maker's situation awareness can be improved by use of VEs. For the study described in this paper we focused on the pre-battle tasks of the intelligence section. The intelligence members must process large amounts of data in short time periods. The application of other presentation means in the form of a VE possibly improves human information processing and hence performance. We adopted a user-centered approach in the development of the VE. This involved task analysis, early design reviews with military experts and a scenario evaluation with intelligence officers. Task Analysis: We analyzed the pre-battle intelligence tasks with Hierarchical Task Analysis (HTA) and task decomposition. With the results, tasks were selected which are likely to benefit from the use of VEs. Design Virtual Environment: For the selected tasks we designed and developed the VE on basis of guidelines of visual momentum. Visual momentum guidelines maximize the integration of information from various displays and views. Furthermore we used an incremental approach in which prototypes of the VE were built and evaluated with officers who are experts in intelligence and military command and control. Scenario Evaluation: Operational intelligence officers evaluated the actual VE. The officers performed pre-battle intelligence tasks with a typical military scenario. After completion of each task, the officers filled in a questionnaire with which they rated the difference for speed, accuracy and mental workload between the VE and their normal 2D manner of working. All officers foresaw benefits of the VE, especially for the determination of fields of view, determination of obstacles due to steepness and the forming of a mental picture of the shape of the terrain. Discussion: For the design of VEs preservation of visual momentum is an effective design approach. Future Work: We will extend the work to during-battle intelligence tasks to study the effect of VEs on the intelligence officer's situation awareness of the development of the battle.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

Digital Skill Learning Using
Constructivist Training Methods

Jerry Childs, Ph.D.

TRW Systems

Albuquerque, New Mexico

Brooke Schaab, Ph.D.

Army Research Institute

Alexandria, Virginia

Norm Blankenbeckler

TRW Systems

Columbus, Georgia

The Objective Force will present many challenges for tomorrow’s soldier. The purpose of this study was to assess the effectiveness of constructivist learning techniques for training soldiers in Military Intelligence (MI) on the digital battlefield. Exercises and test instruments used in current MI courses at Fort Huachuca were adapted for application by school instructors. The constructivist approach was compared with conventional, lecture-based MI training. Constructivist scenarios were designed to require soldiers to operate the All-Source Analysis System-Remote Workstation (ASAS-RWS) and to achieve tactical dominance via the use of superior awareness. The research team developed Practical Exercises (PEs) requiring soldiers to solve problems, develop/send/receive specialized queries, and develop time-sensitive MI products. A major goal was to train soldiers to acquire and apply digital skills in response to changing and ambiguous battlefield situations. The group (n=18) taught using constructivism performed significantly better than conventionally-trained controls (n=32) on the performance-based PE (F (1,43)=11.59, p<.05). Both groups mastered all terminal learning objectives. As compared with the constructivist-trained group, soldiers trained using the conventional method reported higher levels of mental demand, time stress, and effort. The constructivist group covered more material in less time than those taught using the conventional method yet they did not perceive an increase in workload as measured by the NASA-TLX. Results suggest that constructivist training approaches provide an effective means for teaching digital and analytical skills on the future battlefield.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

STEPS TOWARD DEVELOPING A SIMULATION SELECTION METHODOLOGY

Captain Christopher Huffam

Canadian Forces Training Development Centre

Borden, Ontario

C. Ian Mack, P. Eng

DRDC Toronto

Toronto, Ontario

In the Canadian and other militaries, simulation is used to familiarize personnel with equipment under near-real conditions with minimal risk. Up to this point, no standard methodology for the selection of simulation has existed within the Canadian Forces (CF). As a result of this lack, CF personnel involved in acquisition of simulation are exposed to the priorities of Industry, which may differ markedly from those of the CF. This paper will discuss the findings from focus groups in regard to the acquisition of simulation in capital projects and of a case study in the effectiveness of the pilot simulation selection model.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

SEAMAN RECRUIT TO CNO:

APPLYING HUMAN PERFORMANCE ACROSS NAVY CULTURE

L. Kirk Schultz

Naval Education and Training Professional Development

and Technology Center

Pensacola, Florida

The Human Performance System Model (HPSM) provides the basis for the Navy’s human performance initiative and has been widely adopted by upper-level Navy management. The Task Force for Excellence through Commitment to Education and Learning (Task Force EXCEL) has referenced the HPSM in its efforts to develop a “Five-Vector Model” career path for every Navy Sailor. However, discussions in various workshops and working groups indicate that the Model is still not uniformly understood nor consistently applied by Navy personnel. Confusion exists about what each of the four HPSM quadrants represents and what processes they encompass. While providing Human Performance Seminars and Working Group Guidelines has reduced some of this confusion, the Navy still lacks a published human performance process that can act as a simple framework to help implement the System Model. This underlying process must describe human performance principles in terms that can be uniformly understood and applied, not just by management and the human performance "experts," but by Sailors at all levels of command. The 4X4 Performance Improvement Process (PIP) has been designed to meet this need. This paper describes the HPSM, presents the 4X4 PIP and shows how the 4X4 PIP overlays the HPSM. The paper then discusses the need and a plan to communicate the process to the Fleet. Preliminary indications of its potential to impact and change Navy culture are cited.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

INSERTION OF CROWD BEHAVIOR MODELS INTO THE INIWIC COURSE

John M. Kenny

and

Wendy L. Gilpin

Penn State University

The Individual Non-Lethal Weapons Instruction Course (INIWIC) is designed to teach military units to control crowds using non-lethal methods. Proper selection and use of non-lethal techniques requires an understanding of the dynamics of crowd behavior and crowd control. Under the guidance and funding of the Joint Non-Lethal Weapons Directorate, the authors inserted the latest, validated crowd behavior models into the course curriculum and the methods utilized are discussed in this paper. This effort can be categorized into five steps that were used to accomplish the on-the-fly course revision:

§ Familiarization - The INIWIC faculty received a comprehensive brief on the latest crowd behavior models and reviewed the underlying data.

§ Curriculum Development - A team, composed of Penn State researchers and INIWIC faculty, reviewed the existing curriculum objectives, and identified insertion points for the use of crowd control non-lethal tactics.

§ Implementation - The curriculum was revised and introduced to the classroom setting.

§ Evaluation - Using existing testing methods, the effectiveness of the pilot curriculum was evaluated. Additionally, this information was supplemented by feedback from both instructors and students. A list of problems and shortfalls was compiled.

§ Revision/Expansion - Using the list of problems and shortfalls, researchers revised and expanded the curriculum as necessary to meet the learning objectives. It was necessary to repeat this cycle twice to achieve the desired acceptance level.

This paper discusses the pertinent crowd behavior and crowd control models, the underlying data used to validate these models, and the course implementation methods and results. It reinforces the need to design and embed on-going reviews of existing curriculum to ensure that military personnel are properly trained. Course material must be continually reviewed to ensure incorporation of relevant research and real-word experiences and knowledge.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

MODELING HUMAN ERROR FOR EXPERIMENTATION, TRAINING,

AND ERROR-TOLERANT DESIGN

Scott D. Wood

Soar Technology

Ann Arbor, Michigan

David E. Kieras

University of Michigan

Ann Arbor, Michigan

Human error in computer systems has been blamed for many military and civilian catastrophes resulting in mission failure and loss of money and lives. However, the root cause of such failures often lies in the system’s design. A central theme in designing for human-error tolerance is to build a multi-layered defense. Creating such a robust system requires that designers effectively manage several aspects of erroneous system usage: prevention, reduction, detection, identification, recovery, and mitigation. These also correspond to discrete stages before and after error occurrence where different defensive measures can be taken. Human error models can be used to better understand these stages, the underlying cognitive mechanisms responsible for errors, and ultimately how to design systems and training to reduce the effects of inherent human limitations.

This paper presents a general framework for human error recovery based on five key stages of erroneous performance: the commission of an error, its detection, identification, and correction, and resumption of the original task. These stages constitute the main components of a state model that characterizes human performance, and allows designers and trainers to comprehensively address the most important aspects of error-tolerant design. Furthermore, these performance stages can be modeled computationally, to varying degrees, using standard information processing architectures. This work also demonstrates the effectiveness of a technique using GOMS models to design systems to prevent human error. The technique is applied to WebStock, a realistic web application designed to elicit human error, and the results are used to redesign WebStock's user interface. We compared user performance on the original Web-Stock interface with the interface improved using the technique.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

IMPROVING HUMAN PERFORMANCE THROUGH ADVANCED COGNITIVE SYSTEM TECHNOLOGY

Dylan D Schmorrow, Ph.D.

LCDR MSC USN

Defense Advanced Research Projects Agency

Arlington, Virginia

Amy A Kruse, Ph.D.

Strategic Analysis Inc.

Arlington, Virginia

In this paper, we describe the DARPA Augmented Cognition program and preliminary observations from the first phase of research currently in progress. The goal of this multidisciplinary effort is to enhance the warfighter’s cognitive capacity and capability under complex operational and stressful conditions. This program will develop the means to measure a subject’s cognitive state non-invasively in real-time. By accessing the cognitive state of the individual in real-time, automated computational systems will be able to use that information to modify and mediate cognition. These cognitive systems will provide operational data in a manner specifically targeted to the user – and in a way that will not disrupt the user’s current functions. This new interaction will be significantly more potent than just the simple sum of a brain and a computer system. We will achieve an increase the overall system IQ, capitalizing on the synergistic effect of this new human computer symbiosis. The Augmented Cognition program will move beyond simply redesigning human-computer interfaces by completely recreating them with the state of the human as an integral component. This research will enable development of closed loop human-computer technologies, where the state of the user is measured, analyzed and automatically adapted to by the computational system. Success will improve the way 21st Century warriors interact with computer based systems, advance systems design methodologies, and fundamentally re-engineer military decision making processes. These efforts represent a major commitment to the principles of human systems integration and cognitive engineering.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

HUMAN SYSTEMS INTEGRATION FOR FUTURE COMMAND SYSTEMS:

A FOCUS ON BATTLE COMMAND

Dr. Carl W. Lickteig, Mr. William R. Sanders,

and Dr. James W. Lussier

U.S. Army Research Institute (ARI) at Fort Knox, Kentucky

and

LTC (Ret) Gary Sauer

DARPA Program Manager FCS C²

The Army’s ongoing transformation to Future Combat Systems (FCS) poses an unprecedented alliance of humans and machines. Examples of the alliance include humans working with intelligent agents or “bots” for information processing and decision aiding, and with robotic entities for moving, seeing and shooting. Particularly for battle command, advanced technologies are expected to help commanders visualize the operation, describe it within their intent, and direct subordinates toward mission accomplishment. However, the “science” of enabling technologies severely lags the “art” of battle command. Creating an alliance that actually improves, and does not impede, battle command is a human systems integration challenge for FCS. To address that challenge, the FCS C² program is investigating future battle command concepts at the small echelon level, namely the Unit Cell. The effort, led by DARPA and CECOM with assistance from the U.S. Army Research Institute (ARI), includes an incremental series of command-in-the-loop experiments from Oct 2001 to Mar 2003. Methods and findings from initial experiments that focus on the integration of humans and machines to establish new paradigms for command and control are documented. A description of the participant command group and the Commander’s Support Environment (CSE) developed to help the participants command and control the Unit Cell’s manned and robotic elements is provided. Research methods for assessing human system integration include structured interviews, questionnaires, and the analyses of fully recorded experimental runs. Findings are provided on the tasks and functions performed by the command group in collaboration with their CSE technologies. Battle command functions and command group communications are analyzed by run complexity and METT-TC (Mission, Enemy, Troops, Terrain, Time, Civilians) factors. These research methods and findings provide benchmarks and direction for future efforts to forge a true alliance of humans and machines in support of battle command.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

USER-CENTERED DESIGN AND TESTING FOR A DIGITAL FEEDBACK SYSTEM

Laura Faulkner

Applied Research Laboratories, The University of Texas at Austin

Austin, Texas

A pivotal element of U.S. Army training is the expert feedback that units receive upon completion of battlefield exercises at the Manuever Combat Training Centers (MCTCs). The feedback is provided by Observer/Controllers (O/Cs) and Training, Analysis and Feedback (TAF) Analysts. The feedback is based on direct observations by O/Cs of the unit’s performance in the field, and remote observations by the TAF analysts, via GPS-tracking over an instrumentation system and voice radio communications monitoring. As the Army increases its development of digital communications, use of and reliance upon them, there is a critical need for digital training analysis and feedback. However, the analysts and O/Cs no longer have access to significant amounts of information that allow for accurate situational awareness. Further, the high-pressure environment of the TAF operations during battle exercises, and the multiple inputs that the analysts must currently handle, place them in danger of information overload. This creates a delicate design situation for the delivery of digital information for use in MCTC After Action Reviews (AARs). The Army Battle Command System Integration (ABCSI) is a digital communication collection and analysis system created by employing multiple user-centered design techniques. Along with detailed information regarding MCTC digital data needs, this paper describes three of those techniques, how they were employed in ABCSI development, and what was learned in each instance about the users, their environments and the designs. Information and insights gained in applying these techniques provide a rich data source for general system development and deployment at the MCTCs, specific design and implementation of digital data collection and analysis systems, and lifecycle methodologies for achieving maximum use and usability of delivered products.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DEVELOPING AN AFTER ACTION REVIEW SYSTEM FOR VIRTUAL DISMOUNTED INFANTRY SIMULATIONS

Bruce W. Knerr and Donald R. Lampton

United States Army Research Institute

Orlando, Florida

Glenn A. Martin, Donald A. Washburn, and Duvan Cope

Institute for Simulation and Training

University of Central Florida

Orlando, Florida

The Dismounted Infantry Virtual After Action Review System (DIVAARS) is one component of a research program to develop effective Virtual Environment (VE) training techniques for dismounted Infantry leaders, soldiers, and units. This paper describes the development of the system, the conduct of the soldier evaluations, and the resulting system enhancements. DIVAARS is designed to provide trainees with a common understanding of what happened during an exercise and why it happened, so that they can identify ways to improve their performance, and to facilitate data analysis and presentation to support both training feedback and research and development. Determining what happened during an exercise is particularly difficult in urban terrain, where buildings limit the portion of the battlefield that can be observed by any one person. DIVAARS software runs on a Linux platform and can be hosted on a Pentium III PC. Key capabilities of the system are: digital videodisc (DVD)-like replay with synchronized audio and video, the capability to mark times and views during an exercise and jump to them during the AAR, and tabular and graphic data summaries. Several features are especially useful for exercises conducted in urban terrain: interiors of individual floors of multi-story buildings can be selected for viewing, individual soldiers can be identified, and visual traces of their movements can be presented. An evaluation of the system was conducted at the Dismounted BattleSpace BattleLab, Fort Benning, Georgia. Squads composed of Infantry soldiers and computer-generated forces conducted exercises in VE models of the Shughart Gordon enclave over a 5-day period. Questionnaires and interviews were used to evaluate the system. The system was rated positively. Several areas for enhancements were identified and implemented.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

Exploiting Archival Data to Identify

CRM Training Needs for C-130 Aircrews

Robert T. Nullmeyer

Air Force Research Laboratory

Mesa, Arizona

V. Alan Spiker

Anacapa Sciences

Santa Barbara, California

Given the time, money and effort being invested to satisfy Crew Resource Management (CRM) training requirements, there is surprisingly little empirical data to guide CRM course content. In military programs, CRM training is often organized around a set of elements that are listed in service-wide training regulations, with roughly equivalent attention being paid to each area. If some CRM processes are more closely related to mission outcomes than others, the value received from CRM training might be increased through greater focus on areas of greatest need. Military and commercial aviation research findings and recommendations led the Air Force to require a more data-driven approach to establish specific behaviorally anchored CRM training objectives. The Air Mobility Command, the C-130 Aircrew Training System Program Office, Lockheed Martin, and The Air Force Research Laboratory formed a partnership that is now conducting a series of studies and analyses to identify and prioritize CRM training requirements for C-130 aircrews. This paper discusses two analyses of existing data: (1) instructor observations recorded in student training folders during mission qualification training, and (2) mishap report narratives and associated databases of causal and contributing human factors that are maintained by the Air Force Safety Center. In both cases, the original data were generated for other purposes, but were made available for these CRM analyses. Each proved to be a fertile source of insight regarding specific CRM behaviors that need to be considered in redesigning CRM instruction. We describe these two data sets, our analytical approach, trends identified, and implications for CRM training. We anticipate that these analyses will contribute to the development of observable CRM training objectives that will, in turn, enable compelling transfer of training assessments of resulting changes in CRM training.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

An Applied Training Model for Retention of Women in Aviation

Mary Ann Turney, Ed.D., Associate Professor

Arizona State University

Merrill R. Karp, Ph.D., Associate Professor

Arizona State University

James C. Bishop, Ph.D., Associate Professor

Bryant College

Mary Niemczyk, Ph.D., Assistant Professor

Arizona State University

Ruth L. Sitler, Ph.D., Associate Professor

Kent State University

Mavis F. Green, Ph.D., Associate Professor

Embry-Riddle Aeronautical University

A projected long-term pilot shortage coupled with the low numbers of women pursuing technical careers in aviation prompted a three-year study focused on retention of women. The study, undertaken by a team of university faculty from Embry-Riddle Aeronautical University, Arizona State University, Bryant College, and Kent State University. The goal was to discover the factors that influence women once they have already selected an aviation career, and to discover what could be done to support those women who have demonstrated a serious interest in an aviation career by enrolling in a collegiate aviation program.

This paper reports the results of a statistical analysis of data collected in the first and second years of the study, and introduces a training model to meet the needs of the aviation industry. Extensive data for the study were collected from questionnaires distributed to 382 collegiate aviation students (192 women and 190 men) in nine colleges and universities, representing widely varied geographic areas and including two and four year institutions.

Research results presented in this paper reflect significant areas of concern among women in pilot training. Differences emerged in the responses of the male and female groups; these differences were expected. However, a surprising finding was that women in early stages of pilot training responded differently from women in more experienced stages. These response differences did not occur among the male subjects.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

EVALUATING AND DELIVERING THE

UNITED STATES ARMY AIRCREW COORDINATION TRAINING ENHANCEMENT (ACTE) PROGRAM

Gary Grubb

Dynamics Research Corporation, Enterprise, Alabama

Neal Crossland

Dynamics Research Corporation, Andover, Massachusetts

Dr. Larry Katz

Army Research Institute, Fort Rucker, Alabama

The Aircrew Coordination Training Enhancement program is a continuation of the US Army Research Institute (ARI) priority to promote applied research and development (R&D) of the Army Aircrew Coordination Training (ACT) program. The goal of the ACT Enhancement effort is to provide a web-delivered, interactive aircrew coordination training system that provides Army aircrews worldwide with the knowledge and skill-sets needed to increase flight safety and mission effectiveness in daily operations. The research plan consists of three major phases – upgrade and sustain the existing ACT program, refresh and maintain the upgraded ACT program, and deploy advanced ACT applications.

This paper describes objectives and outcomes of ongoing high performance team training system R&D under the guidance of ARI. The prototype products from the first phase of research include two interactive multimedia courses of instruction with supporting training materials. Development of courseware web application components and production of graphics are achieved with a suite of Macromedia Flash and Dreamweaver UltraDev authoring tools. The Aircrew Course and Instructor Course include a fully integrated Data Management System that tracks student demographics, provides graphic feedback displays during evaluation exercises, and facilitates electronic course critiques. User testing and validation results indicate high levels of acceptance for both the training and performance evaluation components. Initial testing of the prototype courseware on the Army’s distance learning suite supports both the web-based and instructor facilitated delivery strategies for Army-wide implementation. The ACT event-driven scenarios serve as model constructs for integration of ACT into advanced aircraft simulators and multiple aircraft training exercises supported by distributed interactive simulation. Ongoing research activities include developing web-based training support packages and institutional training to support Flight School XXI. Training effectiveness results have led the Army to initiate research into applying the ACTE courseware design and delivery model to other than aviation systems.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DEPARTMENT OF ENERGY (DOE) EMERGENCY OPERATIONS FUNCTIONAL ANALYSIS FOR PERFORMANCE IMPROVEMENT

Laura Kratochvil, M.A.

TRW Systems

Albuquerque, New Mexico

Jerry Childs, Ph.D.

TRW Systems

Albuquerque, New Mexico

The Department of Energy (DOE) facilities are responsible for producing, storing, and handling significant quantities of nuclear materials, weapons, classified information, and equipment, creating a need for extensive initial and ongoing training in the interest of national security. Emergency Operations (EO) staff throughout the entire DOE complex must have the knowledge, skills, and abilities (KSA) to respond to any site/facility/transportation emergency and to any radiological or nuclear crisis or emergency. Therefore, it is essential that EO staff training be designed complex-wide not only to maintain but to improve human performance in responding to a radiological or nuclear emergency. As one of the initial activities in designing a comprehensive staff training program, a first-ever, complex-wide, functional analysis was performed. Spanning several months and involving the Nonproliferation and National Security Institute (NNSI), the Emergency Operations Training Academy (EOTA), DOE, and TRW, this pioneering effort took an in-depth, national approach to data collection, resulting in comprehensive analysis of 15 EO staff performance functions, which were comprised of 613 performance tasks, and subtasks. Quantitative survey results relied on response ratings using 5-point Likert scale and applied within a Training Recommendations Decision Tree. Overall, the data reflected an overwhelmingly large number of recurring training task recommendations within all EO functional areas. Of the 613 total tasks addressed, only two (<1%) were recommended for informal (on-the-job) training. In addition to quantitative data, the survey asked respondents to provide comments regarding training needs for each functional area. Also, table-top interviews with EO staff regarding the functions were performed at various nuclear sites around the country. This allowed an opportunity for further participant input and perspective on the EO performance and training issues. The application of this phased, quantitative/qualitative, systems-based approach provided significant insight into EO training program needs for the future. This comprehensive, complex-wide training analysis supported the development of standard training requirements for effective EO course design and development at DOE sites across the United States.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

FIDELITY REQUIREMENTS FOR SPEECH

OF SYNTHETIC TEAM MEMBERS

James M. Stokes

CHI Systems, Inc.
Lower Gwynedd, Pennsylvania

Speech interaction with synthetic team members has the potential to provide a highly natural task context in situations where a trainee must utilize specialized communication skills for key operational functions. Systems of this type can support team training and provide a high level of skill transfer to the operational context, with fewer personnel than currently needed. Because speech is a critical behavioral component for synthetic team members, the issue of speech fidelity is central to their employment. Given existing speech technology limitations, it is important to determine the level of speech fidelity required to support effective team training. Such requirements can guide developers, who face a series of fidelity-related issues and tradeoffs. Examples of engineering decisions from recent empirical engineering research are presented in the context of a general analysis of the speech fidelity issue for synthetic team members.

This paper is available on the 2002 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

MOBILE-COMPUTING DISPLAY

TYPE–SITUATION AWARENESS AND

PERFORMANCE ISSUES

Mary Mc. Wesler, MS

International Truck and Engine Corporation

Development and Technology Center

Ft. Wayne, Indiana

William P. Marshak, Ph.D.

SYTRONICS INC.

Dayton, Ohio

Jennie Gallimore, Ph.D.

Wright State University

Dayton, Ohio

With the advent of wearable/mobile computing in the military domain, Human Factors questions arise as to how to implement the systems on dismounted soldiers in a battlefield setting. To design effective navigation systems we must consider the hardware used. While laboratory tests are useful and help provide initial design information, field studies are critical for determining the successfulness of the system. Critical issues to training and operation include the choice of display and the ease of cognitive understanding.

This study answered, to some extent, questions about Display Type (monocular occluding Head Mounted Display or a small area Head Down Display) on participant situation awareness (SA) and navigation performance in a fielded domain. The results of the study demonstrated that the Display Type did not play a factor on participants’ navigation performance and SA. It was expected that the demands of the task would be the driving factor in the choice of display and VPF to be utilized. This was not the case. Display Type was not an issue and the Egocentric - Bird’s Eye View Visual Presentation Format (VPF) performed best in most tasks.