Irwin Hudson’s research while affiliated with Army Research Laboratory and other places

Head mounted displays (HMD) are becoming ubiquitous. Simulator sickness has been an issue since the first simulators and HMDs were created. As computational power and display capabilities increase, so does their utilization in technologies such as HMDs. However, this does not mean that the issues that once plagued these systems are now obsolete. In fact, evidence suggests that these issues have become more prevalent. Whether the system is Augmented Reality (AR), Virtual Reality (VR), or Mixed Reality (MR) the issues associated with simulator sickness or cybersickness have become more widespread. The reasons are uncertain, but probably multiple. One possible reason is the concept of vection, which is the illusion of movement to the participant where there is none physically. Vection plays a vital role in immersion and presence, however; it is also integral in simulator sickness. Another potential reason is the availability of HMDs. Traditionally a tool used in military training or laboratory settings, HMDs have now become a consumer item. This work reviews the current state of HMD issues such as simulator sickness or cybersickness. It reviews the similarities and differences of the sickness states that are commonly found with HMDs. Also, terms such as presence and immersion are delineated so they are used appropriately. The current theories on simulator sickness and cybersickness are reviewed. Further, the measurement and mitigation strategies currently being employed to reduce sickness are reviewed. Lastly, suggestions for more accurate measurement are recommended.

The study of performance, workload, and stress have become a mainstay in the field of Human Factors. These constructs are multi-faceted and are assessed by a variety of measures. In seeking to enhance performance by managing mental workload and stress, it is important for measures to be anchored to meaningful criteria. Workload and stress must be considered with respect to the performance measures that address the most central objectives. While workload and stress research has progressed over the years and includes research across different levels and domains, there has been less effort to link measures to specific performance outcomes. The present study examined four performance metrics from the same task in terms of the workload and stress measures that are most closely associated with, and predictive of them. Results indicated that different sets of workload and stress measures predicted different performance measures, suggesting that measures should also be selected based on the performance criteria of interest.

Personnel Selection has been a long standing focus in the fields of Organizational Psychology, Human Factors Psychology, Business Management, Human Resources, and Industrial Engineering. Assessment methods in personnel selection can be categorized into subjective and objective methods. Selection assessments are often broad in attempting to capture the essence of person for success in a role or organization. However, this type of approach often yields inconclusive and biased subjective results. Therefore, focusing on key skills seems to be more beneficial. The skill of focus for this effort is decision making. Since those who make more good decisions are often influential and rise to leadership positions, it is imperative that better ways to uncover, assess, predict, or enhance DM skills, be developed. To do so, a review and firm understanding of personnel selection and decision making is necessary.

Soldiers benefit from the ability to detect threats conveyed via human kinesic cues, or non-verbal body movements. Simulation-Based Training (SBT) supplies an avenue to improve kinesic cue detection performance. Instructional strategies in SBT are designed to enhance performance outcomes, and are tied to workload in terms of training effectiveness. Three instructional strategies were analyzed in a between-subjects design for their degree of perceived workload: Highlighting, Massed Exposure, and Kim’s game. Workload included subjective mental and global demand subscales from the NASA-Task Load Index (NASA-TLX). A multivariate analysis of variance showed that Kim’s game contained the highest mental and global demands; Highlighting produced the lowest mental and global demands. The differences in demands suggest rationales for strategy placement, in terms of combinations (e.g., layers or progressions) and other applications (e.g., air traffic control, medical diagnosis, and After-Action Reviews). The strategies’ workload differences are also traced to differences in attention and working memory.

In order to survive in combat zones, an individual soldier must be proficient in the following skills: Pattern recognition and Behavior cue detection. Although, current military training requirements are inadequate for developing pattern recognition and behavior detection, research shows that Simulation-based Training, via virtual environments (VE’s) can improve pattern recognition and behavior detection skills. However, the use of VEs for visually dependent tasks may also increase simulator sickness in some individuals. This experiment compared a virtual version of Kim’s game (i.e., an observational game to increase memory and performance) to a control group to assess the role of simulator sickness on performance. Participants were randomly assigned to either the Kim’s game or control condition and completed a pre-test, training vignette, and post-test. During the experiment, participants recorded their level of simulator sickness using a questionnaire developed by Kennedy et al. (International Journal of Aviation Psychology 3:203–220, 1993 [1]). The data analysis revealed that the Kim’s game group reported higher levels of simulator sickness symptoms which had a negative effect on performance (i.e., detection accuracy and false positive detection). The results also indicated that there was a positive correlation in the control group between disorientation and detection accuracy. This implies that the control group may have become familiar with the experimental task, suggesting that simulator sickness did not negatively impact their performance. The following paper discusses the influence of simulator sickness on performance and offers new ways to reduce simulator sickness for behavior cue detection training.

The use of avatars as learning agents is becoming increasingly popular in the sports, education and military domains due to the rapid advancement in distributive technologies (e.g., internet, virtual worlds, etc.). When it comes to military and sports, Simulation-Based Training has proven to be cost-effective, due largely to restrictions on time, costs and safety [1]. As virtual reality and virtual worlds have become cheaper and more powerful in computer terms, the subject of how an avatar relates to an avateer (the avatar’s controller) is becoming increasingly popular. More precisely, interest rests on how an avatar’s appearance may promote or disrupt training objectives, by affecting the behavior or the psychology of a user, and thus subsequently raising or degrading learning. Virtual simulations for training have often shared the aspect of avatars found in Virtual Reality, video games, and Virtual Worlds. This paper examines how avatar representation can provide insight into manipulating avatar appearance for training demands. Existing literature suggests avatars act as drivers for affective changes in attitude and motivation, and can be integrated into an instructional strategy.

Variable selection for predictive modeling has traditionally relied on theory in the psychological domain. Given the recent advancements in computing technology and availability, researchers are able to utilize more sophisticated mathematical modeling techniques with greater ease. The challenge becomes evaluating whether theory or mathematics should be relied upon for model development. The presented analyses compared the use of hierarchical and stepwise variable selection methods during a predictive modeling task using linear regression. The results show that the stepwise variable selection method is able to obtain a more efficient model than the hierarchical variable selection method. Implications and recommendations for researchers are further discussed.

The evolution of robots from tools to teammates requires a paradigm shift. Robot teammates need to interpret naturalistic forms of human communication and sense implicit, but important cues that reflect the human teammate’s psychological state. A closed-loop system where the robot teammate detects the human teammate’s workload state would enable the robot to select appropriate aiding behaviors to support its human teammate. Physiological measures are suitable for assessment of workload in adaptive systems because they allow continuous assessment and do not require overt responses which disrupt tasks. Given the large variability in physiological workload responses across individuals, an algorithm that accommodates variability in workload responses would be more robust. This study outlines the development and validation of algorithms for workload classification. It discusses (i) a workload manipulation paradigm, (ii) the evaluation of the algorithms for deriving a workload index that is individualized, and (iii) parameter selection for optimal classification.

Psychological constructs, such as engagement and flow, can be used to determine an individual’s involvement in a task and predict levels of performance during Simulation-Based Training (SBT) in military operations. This experiment used a virtual form of Kim’s game (an observational training game that includes memorization of objects and later recall), to improve pattern recognition and behavior cue detection, during SBT. The purpose of this experiment was to assess participant engagement and flow between two conditions, Kim’s game vs. the control. Seventy-five participants were randomly assigned to either condition, and completed a pre-test, training vignette, post-test, and multiple questionnaires which assessed the individuals’ levels of engagement and flow. Experimental results show the control group reported higher levels of both engagement and flow in all subscales, and flow as a higher predictor of performance than engagement. This paper examines plausible explanations why the engagement questionnaire did not assess differences in performance. The lack of statistically significant findings may be a result of the engagement survey questions and the type of task (i.e., discrete or continuous). Finally, this paper provides future recommendations for examining the role of engagement and flow for simulation-based behavior cue detection training.