Dr. Steven J Kass

Medical errors are currently the third leading cause of death in the United States, with more than a quarter of a million patients dying each year from these mistakes [1]. These errors are often attributed to human factors such as lapses in attention. However, medical personnel can be trained to detect errors early and take corrective action to prevent patient harm. Building training lessons and making them readily available as the need for additional training arises is a time-consuming task for trainers. In this paper, we present an algorithm to automatically build training lessons for nursing education using large language models (LLMs). Our algorithm is based on advanced prompt engineering techniques that incorporate storytelling techniques to create innovative training scenarios designed to enhance the situational awareness of nurses in dynamic clinical settings. The algorithm has been evaluated in a user study that requires nurse trainers to evaluate training scenarios built by subject-matter experts (SMEs) and our algorithm. The results showed that nurse trainers were not able to distinguish between human and AI-generated scenarios, rating the AI-generated scenarios as equally accurate and creative as human-generated scenarios.

This study examines how adding scents to a virtual reality (VR) environment impacts immersion, anxiety levels, and mood. Participants ( N = 40), aged 18–44 years with varying levels of VR technology experience, were randomly assigned to either the control (no scent), congruent (i.e., scents that match the visual scene), or incongruent scent conditions. Participants were assessed on changes to mood, anxiety levels, immersion, and performance on a memory recognition task. Results suggest that while the VR experience significantly reduced anxiety and lifted mood, scent did not significantly affect memory performance or immersion in the VR experience. Future directions for research include exploring the use of scent to engage participants more fully in exposure therapies.

This paper describes the development of a mobile phone application (MEDUSA) for training situation awareness. MEDUSA was developed as a cost efficient form of on demand training to supplement other more costly, traditional forms of training such as simulation labs and live patient rounds. Users care for a simulated patient and are tested using a version of the Situation Awareness Global Assessment Technique (SAGAT). Users receive feedback on how many SAGAT queries they answered correctly and are provided tips on how to improve SA. Initial usability testing demonstrated strong support for the program among a sample of nursing students. Continued efforts are on-going in the design and testing of a teaching portal that allows trainers to easily create their own scenarios.

As cell phones have adapted to becoming personalized computers, the functionality of these two different devices has converged. Conversely, the way people interact with these two platforms is vastly different. With much less space, different controls, and lower specifications, mobile interface design looks and feels significantly different than that of desktop devices. This paper aims to analyze what makes UI design for desktop applications different from mobile devices and how developers can ease this transition when porting from one device to another. The results of a survey found that computing power was a large factor in which device users chose for a particular task, while screen size was not quite as important. Through identifying differences in these applications and suggesting solutions, the awkward transition from bulky, powerful devices to the small screens of everyday mobile devices can be alleviated.

Psychomotor assessment tasks have been shown to be predictive of individual performance in aviation training. Predictive relationships have also been shown between performance in training and cognitive measures such as intelligence, processing speed, and spatial reasoning. The present study seeks to better understand the relationships between basic cognitive measures and a modern psychomotor tracking task. Participants completed a matrix reasoning task, a working memory capacity (WMC) span task, two reaction time tasks and a multiphase psychomotor tracking task. Only processing speed measures were correlated with psychomotor tracking tasks. Results support using psychomotor tracking tasks as a selection tool in addition to aptitude testing.

This study investigated the distracting effect that the mere presence of a mobile phone could have on simulated driving performance when drivers can hear a phone notification, but are unable to respond to it. A sample of 45 participants, categorized as either low or high mobile phone dependent, were randomly assigned to either a group that received a phone notification while driving in a simulator or to a control group. Both groups drove two scenarios, with the experimental group receiving, but not attending to, a phone call during the second scenario. Drivers who received the phone call veered off the road more often than drivers in the control group regardless of mobile phone dependency. Highly dependent drivers drove faster, were involved in more collisions, and drove over the centerline more frequently than less phone dependent drivers.

This paper describes a research program on the introduction of globally distributed teams into undergraduate software engineering courses. A pilot study, now completed, involved students at a single institution using two different virtual environments while cooperatively developing requirements artifacts in 3-person virtual teams. We describe the results of this pilot study and the plans for its extension to a three-year multi-institution, multi-culture, and multi-language setting.

Linguistic Geometry (LG) is a mathematical theory that captures the decision making processes similar to those of the great chess masters. In the current demonstration LG is applied to a military combat game that allows users to test their strategies against an intelligent simulated force while receiving feedback regarding the accomplishment of the task and amount of resources spent. Additionally, a usability study was conducted on LG-based software (LG-PROTECTOR) indicating that the software was easy to learn and remember and has an interface conducive to training. The next generation of LG software (LG-EXPERT) is currently under development to provide increased capabilities that will help train users to “Think Like A Commander” as they practice rules of engagement and learn effective courses of action. Additional training implications are discussed.

The purpose of the current study was to investigate the impact of circadian cycles, gender, and propensity for cognitive errors on driving performance. Participants (14 women, 6 men), classified as “Evening Types,” performed a low-fidelity driving task during either their peak (evening) or off-peak (morning) times. Although neither penalty nor damage scores varied significantly by time of day, a gender by time of day interaction was observed for damage scores. Specifically, women driving during the day (off-peak) had accrued significantly more damage to their vehicles than did women driving in the evening (peak). However, damage scores for men were not significantly impacted by the time of day in which they drove. Women also received significantly higher penalty scores regardless of time of day. Participants' overall propensity for committing cognitive failures was significantly related to penalty scores ( r = .50), but not to damage scores ( r = .39). The results help identify individual differences in risk factors for driving.

Validation is an essential step in developing a new measure. This research is concerned with gathering initial evidence of the construct validity of the Mild Motion Questionnaire (MMQ), a 39-item paper-and-pencil based tool for assessing sopite syndrome and other responses to non-sickening motion. Two groups of participants (49 US Navy trainee pilots versus 56 undergraduate students) completed the MMQ after exposure to one of four disparate stimuli: a familiarization flight, visuomotor sensory rearrangement, a control task, or progressive muscle relaxation. The various treatment conditions led to distinctly different symptoms of sopite in participants. The specific symptom profiles revealed through MMQ subscale scores differed greatly and predictably across conditions, suggesting that the MMQ effectively discriminated among the types of symptoms experienced. In addition, a convergence was observed between responses on an established motion sickness questionnaire and subscales of the MMQ pertaining to sickening responses to mild motion, supporting further the (convergent) construct validity of the MMQ.