Kristy Elizabeth Boyer
· Ph.D. ProfessorVerifiedUniversity of Florida · Computer & Information Science & Engineering
Active 2007–2026
About
Kristy Elizabeth Boyer, Ph.D., is a faculty member in the Department of Computer & Information Science & Engineering at the University of Florida. She holds a Ph.D. in Computer Science from North Carolina State University, obtained in 2010, and has academic backgrounds including a Master's degree in Industrial & Systems Engineering from Georgia Institute of Technology and a Bachelor's degree in Mathematics & Computer Science from Valdosta State University. Her research focuses on human-centered computing, with notable contributions recognized through various awards such as the Top Paper at the International Conference on SportsHCI in 2025, the UF Term Professorship in 2021, and multiple best paper awards at international conferences. Dr. Boyer has also received prestigious fellowships including the National Science Foundation CAREER Award in 2014 and the NSF Graduate Research Fellowship in 2007. She is actively involved in research groups such as the Learn Dialogue Group and the INTERNET Lab, and her work has been recognized for its impact in areas related to user modeling, adaptation, personalization, and artificial intelligence in education.
Research topics
- Computer Science
- Artificial Intelligence
- Psychology
- Social Science
- Mathematics education
- Mathematics
- Machine Learning
- Developmental psychology
- Human–computer interaction
- Multimedia
- Programming language
- Social psychology
- Communication
Selected publications
2026-04-13 · 1 citations
articleOpen accessCollegiate student-athletes train and compete in a dense data ecology where information about their bodies and performances circulates among coaches, staff, and fans. To understand how student-athletes themselves engage with this data, we conducted interviews with 20 student-athletes, identifying four modes of engagement: 1) performance-directive, executing training and targeting improvement; 2) reflective-monitoring, assessing the body’s reaction to training and daily load; 3) coach-mediated, receiving insights through staff expertise; and 4) selective-disengagement, intentionally stepping back to protect confidence or avoid overload. These findings fill a gap left open by three related areas of research: SportsHCI, collegiate athletics, and personal data engagement. Each mode entails reasons, practices, and trade-offs. Student-athletes draw on different combinations of these modes as they respond to training demands, coaching oversight, and their own well-being. Our findings highlight how an evolving data ecology creates opportunities and pressures, requiring student-athletes to balance performance with protecting their state of mind.
2025-11-03 · 1 citations
articleContinuous wearables collect physiological data throughout the day and are increasingly being used to help athletes and training communities achieve performance goals. While valuable, these devices collect extensive personal data with minimally studied sharing patterns. We investigate this issue in the context of elite United States collegiate sports by analyzing 15 semi-structured interviews with student-athletes. This paper reports on a qualitative analysis using the contextual integrity framework—an established privacy framework that identifies usage norms—to map the flow of information collected by continuous wearables. We highlight seven key descriptions of how data is being shared, identify privacy considerations, and provide recommendations for the research community to consider how data from continuous wearables is managed. Ultimately, this work presents a first step in aligning privacy protections with the rapid advancement of sport technology.
A qualitative examination of the evolving role of sports technology in collegiate coaching
Frontiers in Sports and Active Living · 2025-09-01 · 5 citations
articleOpen accessCoaches play a central role in shaping athlete performance and development. In collegiate sports, coaches must balance competitive goals with the broader needs of student-athletes. As technology becomes more available in sports, it is becoming increasingly embedded in the workflows and decision-making processes of coaching staff. While many recognize the growing presence of these tools in sports, there is limited understanding about how coaching staff select and integrate these tools into their professional practice. This study addresses this gap by investigating (1) the types of technologies that collegiate coaching staff use; (2) how coaches integrate those technologies into key coaching domains such as baseline testing, practice planning, and injury management; and (3) what motivates or hinders technology adoption in this environment. We conducted five semi-structured focus groups with 17 coaching staff members from National Collegiate Athletic Association (NCAA) Division I sports teams in the United States, representing men's American football, men's basketball, women's basketball, women's soccer, and women's volleyball. Participants included coaches, athletic trainers, strength and conditioning staff, dietitians, sports scientists, and administrative staff. We provide an inventory of technologies in active use to support key aspects of coaching. Our findings show that when aligned with coaching goals, technology offers valuable support for decision-making, individualized student-athlete management, and coach-athlete communication. These findings also point to the importance of supporting coaching staff in managing the growing demands of technology use. By highlighting how collegiate coaching staff apply technology, this study deepens understanding of what technology integration in coaching looks like in real-world practice. The insights may offer valuable direction for scholars, coaches, and organizations who aim to strengthen coaching practice and athlete outcomes through thoughtful integration of technology.
Proceedings of the AAAI Conference on Artificial Intelligence · 2025-04-11 · 1 citations
articleOpen accessSenior authorSimulating learner actions helps stress-test open-ended interactive learning environments and prototype new adaptations before deployment. While recent studies show the promise of using large language models (LLMs) for simulating human behavior, such approaches have not gone beyond rudimentary proof-of-concept stages due to key limitations. First, LLMs are highly sensitive to minor prompt variations, raising doubts about their ability to generalize to new scenarios without extensive prompt engineering. Moreover, apparently successful outcomes can often be unreliable, either because domain experts unintentionally guide LLMs to produce expected results, leading to self-fulfilling prophecies; or because the LLM has encountered highly similar scenarios in its training data, meaning that models may not be simulating behavior so much as regurgitating memorized content. To address these challenges, we propose Hyp-Mix, a simulation authoring framework that allows experts to develop and evaluate simulations by combining testable hypotheses about learner behavior. Testing this framework in a physics learning environment, we found that GPT-4 Turbo maintains calibrated behavior even as the underlying learner model changes, providing the first evidence that LLMs can be used to simulate realistic behaviors in open-ended interactive learning environments, a necessary prerequisite for useful LLM behavioral simulation.
ArXiv.org · 2025-06-12
preprintOpen accessProject-based learning plays a crucial role in computing education. However, its open-ended nature makes tracking project development and assessing success challenging. We investigate how dialogue and system interaction logs predict project quality during collaborative, project-based AI learning of 94 middle school students working in pairs. We used linguistic features from dialogue transcripts and behavioral features from system logs to predict three project quality outcomes: productivity (number of training phrases), content richness (word density), and lexical variation (word diversity) of chatbot training phrases. We compared the predictive accuracy of each modality and a fusion of the modalities. Results indicate log data better predicts productivity, while dialogue data is more effective for content richness. Both modalities modestly predict lexical variation. Multimodal fusion improved predictions for productivity and lexical variation of training phrases but not content richness. These findings suggest that the value of multimodal fusion depends on the specific learning outcome. The study contributes to multimodal learning analytics by demonstrating the nuanced interplay between behavioral and linguistic data in assessing student learning progress in open-ended AI learning environments.
2025-09-22
articleSenior authorHow Virtual Agents Can Shape Human-Human Collaboration: A Systematic Review
Lecture notes in computer science · 2025-01-01
book-chapterEngaging with Natural Language Processing: An Exploratory Study across Multiple Middle School Grades
Proceedings. · 2025-06-10
articleOpen accessSenior authorThis study examines how an NLP-focused AI curriculum was adapted for 5th and 8th graders with different prior experiences.Classroom video analysis revealed that the 5thgrade teacher used structured, hands-on activities, while the 8th-grade teacher fostered autonomous, peer-supported exploration.Both groups showed significant learning gains.Using activity system analysis, we explored how tools, guidance, and collaboration shaped curriculum enactment within students' Zone of Proximal Development (ZPD).Findings underscore the need for flexible AI curricula, showing that scaffolding and inquiry-driven methods effectively engage students across grade levels.
Coach, Data Analyst, and Protector: Exploring Data Practices of Collegiate Coaching Staff
2025-04-24 · 4 citations
articleOpen accessJMIR Human Factors · 2025-03-06
articleOpen accessSenior authorBACKGROUND: Given the global burden of insufficient physical activity (PA) in children, effective behavioral interventions are needed to increase PA levels. Novel technologies can help expand the reach and accessibility of these programs. Despite the potential to use heart rate (HR) to target moderate- to vigorous-intensity PA (MVPA), most HR research to date has focused on the accuracy of HR devices or used HR for PA surveillance rather than as an intervention tool. Furthermore, most commercial HR sensors are designed for adults, and their suitability for children is unknown. Further research about the feasibility and usability of commercial HR devices is required to understand how children may use HR during PA. OBJECTIVE: This study aimed to explore the use of a chest-worn HR sensor paired with a real-time HR display as an intervention tool among preadolescent children and the usability of a custom-designed app (Connexx) for viewing real-time HR. METHODS: We developed Connexx, an HR information display app with an HR analytics portal to view HR tracking. Children were recruited via flyers distributed at local public schools, word of mouth, and social media posts. Eligible participants were children aged 9 to 12 years who did not have any medical contraindications to MVPA. Participants took part in a single in-person study session where they monitored their own HR using a commercial HR sensor, learned about HR, and engaged in a series of PAs while using the Connexx app to view their real-time HR. We took field note observations about participant interactions with the HR devices. Participants engaged in a semistructured interview about their experience using Connexx and HR during PA and completed the System Usability Scale (SUS) about the Connexx app. Study sessions were audio and video recorded and transcribed verbatim. RESULTS: A total of 11 participants (n=6, 55% male; n=9, 82%, non-Hispanic White) with an average age of 10.4 (SD 1.0) years were recruited for the study. Data from observations, interviews, and SUS indicated that preadolescent children can use real-time HR information during MVPA. Observational and interview data indicated that the participants were able to understand their HR after a basic lesson and demonstrated the ability to make use of their HR information during PA. Interview and SUS responses demonstrated that the Connexx app was highly usable, despite some accessibility challenges (eg, small display font). Feedback about usability issues has been incorporated into a redesign of the Connexx app, including larger, color-coded fonts for HR information. CONCLUSIONS: The results of this study indicate that preadolescent children understood their HR data and were able to use it in real time during PA. The findings suggest that future interventions targeting MVPA in this population should test strategies to use HR and HR monitoring as direct program targets.
Recent grants
NSF · $500k · 2016–2021
NSF · $504k · 2015–2022
NSF · $497k · 2015–2016
NSF · $862k · 2018–2023
NSF · $1.6M · 2021–2026
Frequent coauthors
- 97 shared
James C. Lester
- 89 shared
Eric Wiebe
- 49 shared
Bradford Mott
- 32 shared
Joseph B. Wiggins
University of Florida
- 31 shared
Mehmet Celepkolu
University of Florida
- 27 shared
Robert Phillips
- 27 shared
Collin F. Lynch
Tufts Medical Center
- 24 shared
Michael D. Wallis
Awards & honors
- Top Paper, International Conference on SportsHCI, 2025
- UF Term Professorship, 2021
- Best Paper Award, International Conference on User Modelling…
- National Science Foundation CAREER Award, 2014
- Best Paper by a Student First Author, International Conferen…
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Kristy Elizabeth Boyer
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup