Human Factors and Ergonomics Society
Houston Chapter

Session 1: Safety and Driving (Chair - Christy)

Developing a Holistic Model of Automated Vehicle Takeover Performance
Using Visual Looming-based Evidence Accumulation Models – Hananeh Alambeigi
Abstract: Driving crashes are a leading cause of death and injuries worldwide with over 1.2 million deaths and 20-50 million injuries each year. Automated vehicles (AVs) are expected to reduce these crashes and provide safety benefits. However, the safety of partially AVs is limited by the driver ability to takeover when automation fails. Understanding factors that influence takeover time and quality is a critical first step in designing safer systems. Additional steps are required to integrate these factors into a design process. One method of integration is through simulation frameworks that joins technology with driver models and produces safety-related predictions. Despite the considerable amount of work in the modeling of driver behavior during manual driving, models of driver AV takeover behavior are rare. Therefore, the goal of this study was to translate different components of a takeover process (e.g., action execution) into models of driver behavior (e.g., braking) that can accurately predict the real-life driver behaviors by capturing the impact of influential factors on takeover performance.

More than a Feeling: Sentiment and Emotions Can Reveal Information about
Trust in Self-Driving Vehicles – Ian Robertson
Trust has been identified as a key issue in the acceptance of self-driving vehicles. Trust in automated vehicles is commonly studied systematically via lab studies. While this approach allows investigators full control over how and what is studied, like all experimental studies there is a limit to the ecological validity of the results. That said, there are other potential approaches to studying users’ trust in self-driving vehicles that leverage their actual experiences. As users engage with new driving technology, they share their opinions and thoughts via comments on video platforms and social media and through reviews of current services and products (e.g., Tesla’s Autopilot). These comments are rich with information that may be relevant to understanding users’ trust in self-driving vehicles. One approach to capturing this information is sentiment analysis. Sentiment analysis refers to a broad class of methods that can be used to quantify and extract affectual information from texts. Theoretically, emotions are core to trust (users feel trust). Additionally, emotions influence trust formation via the impact that emotional states can have on cognitive process such as attention. Researchers and designers who work with automated vehicles may be able to extract information about trust in self-driving vehicles from texts generated by real users using sentiment analysis and related methods. To validate this approach, sentiment analysis was applied to passages written by participants who experienced a simulated ride in a self-driving vehicle. The valence extracted from these texts were found to positively associate with corresponding trust scores captured using a validated scale. Additionally, emotions were extracted and their association with trust in self-driving vehicles was explored. The practical and theoretical relevance of these findings will be discussed.

Drowsy Driving Near-Crashes Among Night-Shift Nurses:
A Naturalistic Driving Study – Anthony McDonald
Drowsy driving is a prevalent problem among shift workers that continues to cause injuries and deaths. Night-shift nurses in particular are susceptible to drowsy driving crashes due to their long working hours, disrupted circadian rhythm and reduced sleep hours. Simulator, closed-track and logbook studies have identified the negative impacts of shift-work on driving behavior including high number of near-crashes and falling asleep while driving. This study addressed the general gap in naturalistic driving studies to identify the work, sleep and drive related factors that contribute to near-crashes by conducting a longitudinal study involving ICU nurses in a large health system.

Is everything going right when nothing is going wrong? – Atif Ashraf
Standard Operating Procedures (SOPs) are considered vital in maintaining safe operations in high-risk industries such as the oil and gas and the petrochemical industry. Although a lot has improved in how safety is managed, majority of the incidents that occur in these industries often attribute deviation or non-compliance of procedural steps as a causal factor. Even though workers do not always follow the exact steps of the procedure they use when performing their tasks, very few studies currently investigate these differences between Work-As-Imagined (WAI) and Work-As-Done (WAD). This study proposes a novel coding framework to quantify and categorize the differences between Work-As-Imagined (WAI) and Work-As-Done (WAD) based on the Safety-II perspective. While the Safety-I approach tends to focus on controlling deviations from Standard Operating Procedures (SOPs) as the solution to prevent incidents, this framework aims to describe and code worker’s adaptive behavior while executing their tasks. The proposed framework checks three attributes for each step; Skip, Order, and Action to help decide if WAI aligns with WAD. The framework was tested by applying it to an actual work environment in a petrochemical facility. Initial findings successfully demonstrated the ability of the framework to describe the differences between WAI and WAD.

Session 2: Fatigue (Chair - Farzan)

Disaster Robotics - Camille Peres
Disaster robotics is a growing field. With that growth, more research needs to be done to examine and understand human robot interaction (HRI) in this domain to support effective and safe missions. Background: Disaster robotics is a term coined to describe the use of robots or unmanned systems at the “site of a disaster or extreme incident� (Murphy, 2014). These robots are used to survey the damage and aid in determining where to allocate resources more quickly and efficiently. Purpose: The purpose of this presentation is to show the current efforts underway to understand how fatigue can impact HRI through the following: establishing fatigue measures, understanding how fatigue affects humans-in-the-loop, and observing and understanding team dynamics of those who operate the robots. Method: Sixteen small unmanned aircraft systems (sUAS) pilots participated in the study, and participants were monitored for two days during training exercises conducted in real-world settings. The Human Robot Team Observation Tool App and GoPro cameras were used to record behavioral observations during training exercises, and multiple subjective questionnaires were used to capture participants’ fatigue level after exercises. Results: The fatigue measures indicated that participants’ fatigue level increased significantly as training went on from morning to evening. The behavioral observations suggested that participants were more distracted during mission execution when compared to pre-mission and post-mission. Conclusion: The data collected from this study shows important findings on fatigue as the day wears on as well as critical insights on human-robot interaction.

Correlations between fatigue measures from
various fatigue assessment methods and time on Offshore Platforms - John Kang
Fatigue is a physiological response during or after tasks that has a negative effect on worker safety and health. There has been an increase in workplace accidents, fatalities, and injuries as well as a decrease in work productivity due to fatigue. Fatigue has been implicated as one of the main causes of recent OGE (Oil and Gas Extraction) catastrophes such as the Exxon Valdez Oil Spill and the Texas City Refinery Explosion. The factors that contribute to fatigue include workload, sleep deprivation, and disruption of the circadian rhythm. Studies captured and compared various fatigue measures, mostly subjective, over the course of a single day or two, but no studies compared fatigue measures over time across subjective, physiological, and performance-based tasks in the offshore environment. A team of researchers conducted longitudinal studies of offshore OGE workers over the course of two 28-day hitches on two different drillships in order to capture workers’ daily fatigue levels. The objective of this study is to establish correlations between fatigue measures extracted from various fatigue assessment methods and the duration of hitch between three shifts (day, night, and swing shift). A total of 70 offshore workers were monitored each day using both objective and subjective fatigue assessment methods over the course of each 28-day hitch. After participants finished their 12-hour shift, they were asked to answer multiple questionnaires (Karolinska Sleepiness Scale, Borg's Rating of Perceived Exertion, Mental Fatigue) and to perform a 10-minute Psychomotor Vigilance Test (PVT). Participants wore actigraphy watches while sleeping and recorded their sleep information in sleep journals. Heart rate variability was measured before and after each 12-hour shift for the first two weeks using Actiheart. We used repeated measures correlation (RMCORR) and generalized linear mixed model analysis to establish correlations between fatigue measures and examine daily fatigue levels. We hypothesized that subjective and objective fatigue measures would be correlated, that workers' daily fatigue level would increase over time during a 28-day hitch, and that swing shifts would have greater fatigue levels than both day and night shifts. The analysis will provide important insight into how fatigue levels change over the course of a 28-day hitch as well as which fatigue assessment methods work best in an offshore environment.

Non-invasive Brain Stimulation as a
Fatigue Countermeasure: On Timing and Focality - Meredith Smoot
Cognitive fatigue has a significant impact on executive functions during emergency response operations. Typical solutions to mitigate fatigue, such as caffeine or user-interface alterations (e.g. multimodal feedback) do not address the root of the problem due to the complex, non-linear nature of the human fatigue response. We hypothesize that non-invasive high-definition transcranial direct current stimulation (HD-tDCS) could improve working memory (WM) and augment emergency responder performance under fatiguing conditions. Through this study, we investigate the efficacy of HD-tDCS as a fatigue countermeasure, explain the influence of stimulation on fatigue using concurrent neuroimaging, and discuss the impact of both task and stimulation time scales on fatigue states.

Session 3: Design and Usability (Chair - Melissa)

 
Complexity in Procedures: Initial Findings from an Experimental Study - Camile Peres
Procedures are vital to performing safe and effective tasks in high-risk, complex industries and their deficiencies contribute to many incidents every year. It is important that these procedures can accurately and simply present information to the users so that they have the best information and can put it into effect well. However, it is sometimes necessary to add information to procedures that may help the user, increasing the complexity. This study sought to evaluate the effect of complex procedures on performance and perception in a simulated chemical processing scenario.

Measuring Subjective Usability by Watching Others Using Video - Roslyn Shanklin
Safety precautions surrounding the COVID-19 pandemic limited usability practitioners and researchers ability to conduct in-person, contact-intensive usability testing. Thus, there was an increased need for usability testing that did not require co-location or even temporal synchronicity, yet still captured the nuances of product use. However, this need extends beyond COVID-19 and could have major implications for cost savings, sampling reach, and safety. To address this need, the current study explored Watching Others Using Video, a remote testing method where users rate usability after watching videos of others using a product. Watching Others Using Video has not been heavily used in the usability community due to previous findings of inflated usability ratings as compared to the traditional Use-Then-Measure method, measuring usability immediately after direct product use. The goal of this study was to mitigate the demonstrated score inflation by assessing how different video content affects perceptions of usability. Participants were asked to watch videos showing different levels of product use difficulty (i.e., errors and failures) for the following products: a website, a digital timer, and an electric can opener. Participants then rated each product with the System Usability Scale (SUS) and After-Scenario Questionnaire (ASQ). Results showed consistent usability score inflation across products despite video manipulations. This may be attributed to participants not reliably detecting portrayals of difficulty. Additionally, the error severities presented in the videos may have been negligible to perceptions of usability. Further research is needed to understand how to systematically improve the accuracy of Watching Others Using Video. Practitioners are cautioned against conducting Watching Others Using Video protocols until guidelines can be established for its accurate and effective use.

User-centered design of a mobile mental health application for college students: A phased usability testing approach - Elyssa Ramos
Introduction Mental health concerns are especially prevalent among students in higher education. According to the Center for Collegiate Mental Health (CCMH) [1], one of the most common mental health issues among college students is anxiety. Usage of university counseling services by college students has also increased, implying that a highly developed and efficient mental health infrastructure is needed [2]. For the past two decades, mobile technologies, specifically smartphones and smartwatches, have become an essential part of our daily lives [3]. This has opened up new opportunities to provide healthcare services using mobile platforms, an area called mobile health (mHealth) [4]. Given the popularity of mobile technology among younger populations, mHealth technologies may provide a unique opportunity to address college students mental health challenges. While some mental health apps have shown promise, our recent work [5] shows that usability issues are prevalent and there is a general gap in published research on the user-centered design and evaluation of mental health mHealth apps, specifically for college students. Methods This study applied mixed methods to evaluate the multiple user-centered design iterations of an mHealth app called Mental Health Evaluation and Lookout Program (mHELP). Initial usability testing relied on unstructured qualitative data with small pilot samples, while later iterations expanded the sample size and included quantitative measures. Overall, there were three phases of testing: Phase 1 was performed synchronously via Zoom with usability testing experts, Phase 2 was performed synchronously via Zoom with 33 participants consisting of college students from a large university, and Phase 3 was performed asynchronously via Maze (an online usability testing platform) with 75 participants consisting of college students from the same university. Results In this presentation, we showcase several key observations from this unique multiphase mixed methods usability evaluation as well as show how each phase contributed to a different type of iteration. Discussion Using this novel mixed-method usability approach to evaluate the mHELP app has proven to be beneficial because of the abundance of feedback provided by the target audience of the app as well as the refinement and validation of the design changes made between each phase of user-testing. Although new usability issues were identified with each iteration of testing, most concerns identified in previous phases were able to reach saturation. Therefore, this approach also allows for a continual improvement in both qualitative and quantitative methods of testing. References Heath CfCM. Annual Report. 2020. Beiter R, Nash R, McCrady M, Rhoades D, Linscomb M, Clarahan M, et al. The prevalence and correlates of depression, anxiety, and stress in a sample of college students. Journal of Affective Disorders. 2015;173:90-6. Rainie L, Poushter J. Emerging nations catching up to US on technology adoption, especially mobile and social media use. Pew Research Center; 2014. World Health Organization. mHealth: new horizons for health through mobile technologies. WHO; 2011. Wang X, Markert C, Sasangohar F. Investigating Popular Mental Health Mobile Application Downloads and Activity During the COVID-19 Pandemic. Human Factors. March 2021. doi:10.1177/0018720821998110

User-centered Design of a Safety Dashboard for Offshore Drilling - Brian Park
There have been many incidents and accidents in the offshore oil and gas industry that were linked to poor safety culture and/or fatigue. The safety of a worker is of the utmost importance, so it must be taken into account and monitored in a comprehensive manner. However, the oil and gas industry does not have a comprehensive monitoring system that includes both safety culture and fatigue measures. A variety of fatigue assessment methods have been used to capture workers' fatigue levels in both laboratory and applied settings to investigate which measures reflect workers' fatigue levels. Although safety culture and fatigue have been shown to be important, research still needs to be done on the best way to display employee safety measurements on a dashboard designed to promote quick supervisory decision-making. This presentation details our current efforts in the user-centered design of a dashboard monitoring the safety culture, fatigue, and readiness of offshore drilling workers. A total of 18 participants who served in supervisory or operational roles in the offshore drilling platforms were recruited to participate in interviews to understand their expectations and to collect user-centered requirements. In addition, as part of a workshop aimed at offshore fatigue measurement, a group of more than 20 stakeholders from oil and gas process industries participated in structured brainstorming to generate, discuss and polish various design requirements. Formal interface design requirements were documented using the functional information requirements analysis that specified necessary input and feedback elements within a safety dashboard. The interview questions also addressed how frequently safety culture and fatigue should be assessed to better inform the choice of visualization method for specific types of safety data. An initial prototype dashboard will be presented which was developed based on the developed functional information requirements. As our work is currently in progress, continual efforts include additional iterative design, formative and summative usability testing to create a better user-centered dashboard.

Session 4: Stress and Health (Chair - Camille)

Investigating Activity-based Detection of Stress Among College Students - Reza Jahromi
Stress is a major issue affecting more than 40% of college students throughout the United States, according to a recent study [1]. Increase in academic workload, homesickness, failing to network, and financial problems increase stress, which negatively affects the students' health [2]. Measuring human psychological dynamics, such as stress, is difficult because of the subjective nature of self-reporting and variability between and within individuals. Wrist-worn sensors have shown promise to facilitate objective measurement of behavior associated with psychological state and stress. In particular, heart rate, blood pressure, and electrodermal activity sensors, have been utilized to detect stress through physiological body responses [3, 4]. Although body acceleration has been proven to provide useful information about the behavioral patterns during stressful moments, less attention has been paid to the relationship between physical movement/activity and mental health. Therefore, this study investigated the possibility of detecting behavior correlated with stress using only built-in smartwatch accelerometer sensor data. The accelerometer was chosen because of its low-power consumption and relatively few privacy concerns. We collected naturalistic acceleration data over six weeks from 45 undergraduate and graduate students (36 females and 9 males) from a large university in Texas. Participants ages were from 18 to 34 years old (M = 21.65, SD = 3.5). Subjects also self-reported perceived stress on their phones. We preprocessed acceleration data using Python version 3.6.9 and performed feature extraction in time and frequency domains. Three model schemes, user-specific, general, and similar-users models, were proposed to train the learning algorithms for stress detection. We utilized several supervised learning algorithms, such as K-nearest Neighbors, Support Vector Machines, Random Forest, and XGBoost, to perform the classification task. We achieved a maximum accuracy of 88.91% for the user-specific model, 76.83% for the general model, and 84.41% for the similar-users model. Results indicate that the proposed methodology may inform potential tools for stress detection relying solely on data from a wrist-worn accelerometer. References [1] L. Acharya, L. Jin, and W. Collins, "College life is stressful today “Emerging stressors and depressive symptoms in college students," Journal of American college health, vol. 66, no. 7, pp. 655-664, 2018. [2] C. Son, S. Hegde, A. Smith, X. Wang, and F. Sasangohar, "Effects of COVID-19 on college students mental health in the United States: Interview survey study," Journal of medical internet research, vol. 22, no. 9, p. e21279, 2020. [3] T. G. Vrijkotte, L. J. Van Doornen, and E. J. De Geus, "Effects of work stress on ambulatory blood pressure, heart rate, and heart rate variability," Hypertension, vol. 35, no. 4, pp. 880-886, 2000. [4] F. Bousefsaf, C. Maaoui, and A. Pruski, "Remote assessment of the heart rate variability to detect mental stress," in 2013 7th International Conference on Pervasive Computing Technologies for Healthcare and Workshops, 2013, pp. 348-351: IEEE.

Using heart rate and body acceleration data acquired from wearable sensors to assess mental stress in college students - Moein Razavi
Stress is a common component in the development of mental illnesses such as depression and anxiety in college students [1,2]. It is possible to reduce the chance of acquiring mental disorders and, ultimately, to prevent stress-related pathologies if early stress diagnosis is followed by therapeutic intervention [3]. In our previous work, we evaluated a machine learning-based tool for continuous monitoring and detection of stress in a naturalistic context using a variety of metrics, with heart rate and body acceleration being two of the most descriptive characteristics. While the study showed promise, the population studied were combat veterans [4]. In this study, a total of 54 college students (45 females and 9 males) from a large institution in southern Texas participated in a follow-up naturalistic study that evaluated stress using a combination of wearable wrist-worn sensors and a mobile health (mHealth) application to improve the stress detection technology based on data from a college student population. The ages of the participants varied from 18 to 34 (M = 21.65, SD = 3.5). Data was collected using an mHealth app, called mental Health Evaluation and Lookout Program for college students (mHELP) installed on Apple smart watches and iPhones. This application collected physiological data at a frequency of one hertz, including heart rate and body acceleration. The application also had a feature that allowed users to self-report their stress by tapping on the watch face, which resulted in the creation of a time-stamped record of the self-reported stress incidents. For stress events, we chose 60-second intervals, 30 seconds before and 30 seconds after a stress event was recorded. Non-stress events were extracted from the remainder of the data. In the final sample, there were 3497 instances of stress and 29,475 instances of non-stress events. Non-stress events were categorized as non-stress windows, whereas stress occurrences were labeled as stress windows. To create and evaluate the algorithms, we randomly separated the data into training (34 participants - 82.3% of the total data) and testing (25 participants - 17.7%) sets. In the training set, non-stress events were detected in 93 percent of the data, whereas stress events were found in 7% of the data. Due to class imbalance, we upsampled the training data to decrease noise and increase resolution of the findings while minimizing the amount of information lost during the quantification process [5]. Based on a sensitivity analysis, a ratio of 10 (non-stress events) to 7 (stress events) windows was selected for upsampling. With the use of heart rate and accelerometer data recorded by off-the-shelf smart watches, we developed, assessed, and analyzed various machine learning algorithms for detecting stress events among college students. Results showed that among the algorithms that we trained in this study, the XGBoost was the most robust and the most accurate model, with an area under the curve (AUC) of 0.64 and an accuracy of 84.5 percent. We compiled a list of the most significant features that have an impact on the output of the detection algorithm. SHapley Additive exPlanations (SHAP) values and SHAP summary plots revealed that the standard deviation of body acceleration was the most significant body acceleration feature, and that the standard deviation of heart rate and the minimum heart rate were the most significant time-domain heart rate features for correct detection of stress. References: 1. Sano, A.; Picard, R.W. Stress recognition using wearable sensors and mobile phones. In Proceedings of the Proceedings - 2013 Humaine Association Conference on Affective Computing and Intelligent Interaction, ACII 2013; Geneva, Switzerland, 2013; pp. 671–676. 2. Melillo, P.; Bracale, M.; Pecchia, L. Nonlinear Heart Rate Variability features for real-life stress detection. Case study: Students under stress due to university examination. Biomed. Eng. Online 2011, 10, 113, doi:10.1186/1475-925X-10-96/FIGURES/1. 3. Sadeghi, M.; Sasangohar, F.; McDonald, A.D.; Hegde, S. Understanding Heart Rate Reactions to Post-Traumatic Stress Disorder (PTSD) Among Veterans: A Naturalistic Study. Hum. Factors 2021, 187208211034024, doi:10.1177/00187208211034024. 4. McDonald, A.D.A.D.; Sasangohar, F.; Jatav, A.; Rao, A.H.A.H. Continuous monitoring and detection of post-traumatic stress disorder (PTSD) triggers among veterans: A supervised machine learning approach; Taylor & Francis, 2019; Vol. 9; ISBN 0000000266826. 5. Visa, S.; Ralescu, A. Issues in Mining Imbalanced Data Sets-A Review Paper. In Proceedings of the Proceedings of the sixteen midwest artificial intelligence and cognitive science conference; 2005; pp. 67-73.

Combining Mobile Health with Health Coaching: A Home Study of the Effectiveness for Hypertension Patients - Farzan Sasangohar
Hypertension is a chronic disease that affects people of all ages and can lead to the development of other chronic health conditions including heart disease, stroke and kidney disease (U.S. Department of Health and Human Services, 2020). Total annual medical costs in the U.S. associated with hypertension are currently estimated at $131 to $198 billion annually with approximately $2,500 more per year per person with hypertension (U.S. Department of Health and Human Services, 2020). Hypertension can be attributed to several contributing factors including unhealthy lifestyle choices and chronic health conditions such as diabetes and obesity (Centers for Disease Control and Prevention, 2020). Mobile health (mHealth) is the use of mobile computing and communication technologies in healthcare and can facilitate data collection and enable self-management of chronic conditions such as hypertension (Owen et al., 2015). Health coaching is a patient-centered process based upon behavior change theory which includes goal-setting, education, encouragement, and feedback on health-related behaviors (Oliveira et al., 2017). Combining mHealth with health coaching has the potential to help patients adopt healthier behaviors and promote self-management of their health condition. An mHealth coaching app called HyperCoach was developed by researchers at Texas A&M University to assist hypertension patients with self-management of their disease. We partnered with the American Heart Association (AHA) to design a 30-day hypertension plan that enabled automated health coaching content via the HyperCoach mobile app. An at-home pilot study was conducted to assess the effects of the app-delivered plan on health outcomes, quality of life, compliance and engagement. This study recruited 35 hypertensive patients who were provided with blood pressure (BP) and weight scales that communicated via Bluetooth with the HyperCoach app. The app enabled automatic collection of BP and weight readings and transmission of the data to a cloud-based server. During the first 30 days of the study (health awareness phase), participants were provided with a limited version of the app that reminded them to take BP and weight measurements daily and provided participants feedback of these measurements in numerical and trend chart format. During the second 30 days of the study (health coaching phase), participants were provided with the full coaching version of the app. In addition to feedback on their daily BP and weight measurements, participants were provided a daily task to watch an educational video, read an educational pamphlet, or take a quiz/assessment from the AHA 30-day hypertension plan. This study found that awareness of hypertension health status alone may not be enough to change health-related behaviors except for those people in the most severe condition – hypertensive stage 2. However, the study found that providing health coaching information in conjunction with awareness of health status may encourage a person to change their health-related behaviors. This study demonstrated that using mHealth to support health coaching can assist in the self-management of hypertension. 

Designing mHealth Interventions for Underserved Communities - Samuel Bonet
Chronic diseases are on the rise in the US and globally, with disproportionate impact to underserved communities. Given issues related to access to care and the aging populations preference to receive care in place, there is a need for effective chronic disease self-management methods and technologies. However, self-management is difficult as it requires both knowledge about the disease state and the motivation to change one’s typical behavior. Remote patient monitoring and mobile health (mHealth) interventions are promising ways to support self-management by educating and engaging patients. Yet, limited design guidance exists for the development of mHealth technology in general and for underserved populations in particular. As underserved communities have different needs stemming from cultural differences, educational differences, and differing levels of access to healthcare, designing for underserved communities presents unique opportunities and challenges.
Work is in progress to develop mHealth design guidelines for underserved communities based on interviews conducted with patients and providers, literature reviews, and review of general design principles and heuristics. To be successful, remote patient monitoring and mHealth interventions must consider their intended user populations. This presentation will discuss our preliminary guidelines for doing so, in the context of underserved communities. The guidelines are focused around four themes: literacy, cultural tailoring, self-management, and ease of use. The theme of literacy considers the best way to present information to the user. This includes educational material and feedback on the users health status [1]. Cultural tailoring considers what needs to be customized to a communitys culture and how to do so. The values of the community and their traditions and cuisine should be considered when designing interventions [2]. This is commonly done through participatory design approaches [1,3]. The theme of self-management considers how interventions can sufficiently motivate behavioral changes and mediate access to healthcare. This is commonly done with text messages derived from health theory frameworks and by facilitating communication with medical providers [4,5]. Lastly, ease of use considers how intuitive and easy to learn the technologies are. This can be accomplished by designing technology that corresponds to users mental models. References [1]. Cole-Lewis, H. J. et al. Participatory approach to the development of a knowledge base for problem-solving in diabetes self-management. Int. J. Med. Inform. 85, 96-103 (2016). [2]. Resnicow, K., Baranowski, T., Ahluwalia, J. S. & Braithwaite, R. L. Cultural sensitivity in public health: defined and demystified. Ethn. Dis. 9, 10-21 (1999). [3]. Kline, K. N. et al. Incorporating Cultural Sensitivity into Interactive Entertainment-Education for Diabetes Self-Management Designed for Hispanic Audiences. J. Health Commun. 21, 658-668 (2016). [4]. Sittig, S., Wang, J., Iyengar, S., Myneni, S. & Franklin, A. Incorporating Behavioral Trigger Messages Into a Mobile Health App for Chronic Disease Management: Randomized Clinical Feasibility Trial in Diabetes. JMIR mHealth uHealth 8, e15927 (2020). [5]. Yin, Z. et al. Using Mobile Health Tools to Engage Rural Underserved Individuals in a Diabetes Education Program in South Texas: Feasibility Study. JMIR mHealth uHealth 8, e16683 (2020).

1 Participant Vocabulary Use and Selection Differences in Serial Versus Parallel Presentation of Product Reaction Cards - Michelle Spinelli
2 Fatigue Assessment Methods on Offshore Oil Rigs:
Feasibility and Compliance -John Kang
3 Interruption Recovery Methods and Their Impact on Decision Making in Supervisory-   level Command and Control Settings: A Systemic - Hyun Gee Jei
4 Momentary and Periodic Assessment of College Students' Stress and Anxiety: A   Mixed Methods Naturalistic Home Study – Samira Ziyadidegan
5 Investigating the Perceived Precision and Validity of a Field-Deployable Machine   Learning-based Tool to Detect Post-Traumatic Stress Disorder (PTSD)
 Hyperarousal Events – Moein Razavi
6 The effect of education and drowsiness detection on nurses beliefs and attitudes   towards drowsy driving – Karim Zahed
7 A View of Industry Standards Through a Verification
 and Validation Lens – Tom Shephard
8 Developing a Framework for Telehealth Integration
 into Clinical Workflow – Samuel Bonet
9 Hypoglycemia Detection Using Hand Tremors:
 A Home Study in Adults with Type 1 Diabetes – Reza Jahromi
10 The Rise, Fall, and Rebirth of a UX Team – Ron Vutpakdi
11 Validating a Complexity Scoring System for Procedures – Camille Peres
12 Investigating Attention Allocation While Interacting with an Interruption Recovery   Assistance Tool – Jacqueline Hernandez
13 High-Risk Industries and Safer Procedures – Emily Wilkinson
14 Counterfactuals, Task Performance, and Incidents – Kianna Arthur
15 Augmenting Remote Worker Vigilance:
 Effects of Virtual Supervision – Chidera Azubike
16 Expanding the HF/E Researchers Toolkit with
 Just About Right (JAR) Scales – Andrew Wismer
17 Impact of Social Activist Gaming on opinions of
 Border Security: Immigration – Vanessa Jones
18 Influence of cognitive stress on neuromuscular
 fatigue in older adults. – Halryn Nguyen
19 Influence of Social Views on Game Strategies and Physiological Measures
 of Stress During Social Activist Gaming – Yara Debs
20 Case Study: Risk Assessment of Interaction Among Workers and Their Tasks
 at Sub Sandwich Chain Restaurants during COVID-19 Pandemic – Ali Aljaroudi
21 Virtual Reality Equivalency in a Sorting Task – Kaylee Wilhite
22 Analysis of Cognitive Load in Personalized Emergency Medical Training Using   Augmented Reality - Isabella Pedron
23 Psychomotor Adaptation to Microgravity Using Virtual Reality – Zachary Schulte
24 The Effect of Persuasion Principles on Trust During
 Shoulder Surfing Attacks - Jinwoo Choi
25 Identifying To What People Attend When Perceiving Others
 Maximum Vertical Reach – Dana Wilder
26 Requirements for a Diabetes Self-Management Tool for Underserved Populations:
 A Multi-Stakeholder Approach - Kolby Killion