Scholarly Writings
Minecraft as a Learning System
Minecraft Education Edition for teaching?
This paper takes a close look at Minecraft Education Edition as a complete educational system. It uses Soft Systems Methodology, an approach for understanding complicated situations that involve many different people and perspectives, along with tools like rich pictures (visual maps of how everything connects) and CATWOE analysis (a checklist that identifies who a system serves, who runs it, and what it's meant to accomplish). The paper also includes a surface-level structural analysis, which examines the practical parts of the system such as devices, software, and setup. From there, it lays out a full plan for bringing Minecraft Education into K-12 classrooms, including training and support for students, teachers, and administrators, and finishes with a framework for measuring whether the technology was worth the investment.
Taylor Schlitz, I. (2024). Minecraft Education Edition as a Learning System: Analysis, Implementation, and Evaluation.
Perceptions of GenAI in HE
How do faculty perceive generative AI?
This study surveyed 150 faculty members across U.S. universities, all with established expertise in artificial intelligence, to understand how Generative AI (technology like ChatGPT that can produce text, images, and other content) is changing higher education. Nearly 70 percent of participants had adopted GenAI in their teaching, using it for tasks such as creating course materials, designing activities, and giving feedback. While most faculty held positive attitudes toward using GenAI for teaching, they were noticeably more hesitant about using it for assessment, meaning tests, assignments, and grading. The study also explores how faculty believe teaching should evolve, with many calling for a stronger emphasis on critical thinking and assessments that cannot be easily completed by AI alone.
An, Y., Oh, E., & Taylor Schlitz, I. (under review). Faculty Voices on Integrating Generative AI in Teaching and Assessment: Adoption, Perceived Impact, and Intentions to Change Practices. Journal of Further and Higher Education
GBL & AI integration
Game-Based Learning with AI integration: Creating a Reliable Instrument for Learner Perceptions
Taylor Schlitz and Carter (University of North Texas) built and validated a survey measuring how learners perceive AI-enhanced educational games, moving beyond vague "engagement" metrics. Based on the Cognitive Appraisal Questionnaire framework, it captures three dimensions (Difficulty, Enjoyment, and Creativity), tested through factor analysis, multidimensional scaling, and clustering in SPSS. The categories held up reliably, and perceptions proved best represented in two dimensions: Enjoyment to Difficulty, and structured to creative. Math and Science clustered with Difficulty while Creativity and Enjoyment grouped together, though Language Arts and Language Acquisition didn't fit neatly, hinting at a hidden factor or confusing wording to fix.
Carter, S., & Taylor Schlitz, I. (2025). Game-Based Learning with AI integration: Creating a Reliable Instrument for Learner Perceptions.
SITE AI Panel Proceedings
Assessing Perceptions of AI in Education
This SITE 2025 conference paper (Knezek et al., 2025) presents a panel discussion on measuring perceptions of AI in education, framed as early steps toward a unified taxonomy of attitudes toward AI. It showcases three complementary research approaches: a qualitative study finding that K-12 teachers are optimistic about using generative AI themselves but hesitant to let students use it; a discernment study showing people can't consciously tell AI-written from human-written text yet rate the two differently, with vocabulary choice the biggest tell; and a quantitative survey developing a validated instrument for learner perceptions of game-based learning. The paper closes by raising audience "dilemma" questions and pointing toward a follow-up taxonomy presentation at EdMedia 2025 Barcelona.
Knezek, G., Avalos, G., Cockerham, D., Carter, S., Schlitz, T. (2025) Assessing Perceptions of AI in Education. In R. Jake Cohen (Ed.), Proceedings of Society for Information Technology & Teacher Education International Conference (pp. 773-778).. https://www.learntechlib.org/primary/p/225596/.
Environmental Science Unit
The Plum Lake Environmental Science Unit
"The Plum Lake Project" is a six-week environmental science unit for middle school students (grades 6–8), built on the ADDIE framework and a NASA-adapted 5E model. It teaches scientific inquiry, mathematical modeling, health science, and civic stewardship through a story-driven investigation of lead contamination in a fictional lake, combining narrative videos and graphic novels with hands-on labs, real-world math, and discussions of health risks and community responsibility. Each of the six chapters includes structured activities, discussion questions, and assessment rubrics, culminating in a summative portfolio and a student-created community action plan. Because the unit has not yet been implemented, its effectiveness is presented as a projection, with anticipated challenges including instructor preparation time, uneven technology access, and district pacing constraints.
Taylor Schlitz, I. (2026). The Plum Lake Project: Teaching Scientific Inquiry, Modeling, and Civic Stewardship Through Environmental Contamination.
Developing Generative AI Videos for Environmental Science Education
The Plum Lake Project
A design case study documenting the development of the Plum Lake Environmental Science Educational Video Series, a six-chapter generative AI video curriculum teaching middle school students scientific inquiry, mathematical modeling, and environmental contamination. Built with MagicLight.AI and grounded in Mayer's Cognitive Theory of Multimedia Learning, the project aligned content across NGSS, Common Core Mathematics, and TEKS standards with a three-level differentiation framework. The study documents a clear production learning curve, reporting a 60–70% reduction in development time across chapters while critically examining character-consistency challenges, quality control, ethics, and equity, including a comparison against parallel AI graphic-novel work.
Warren, S. J., Holian, A., Burg, R., Taylor Schlitz, I., Holian, G., & Jones, P. (In development). Developing Generative AI Videos for Environmental Science Education: The Plum Lake Project.
Dynamic Systems Engineering: A Holistic Approach to Supporting Complex Education and Media Design
Dynamic Systems Engineering?
A theoretical and methodological paper introducing Dynamic Systems Engineering (DSE), a stage-based methodology that blends soft systems methodology with system dynamics to bring engineering rigor to educational media development. Traditional models like ADDIE and ASSURE offer broad guidance but lack the structure needed to manage the many interacting systems (pedagogical, technological, psychological, and social) involved in complex digital learning products. DSE addresses this through a ten-stage process emphasizing stakeholder engagement, iterative visual modeling, CATWOE analysis, and systematic documentation, illustrated through an NIH-funded project developing generative AI graphic novels that teach middle school students about environmental toxins while aligning to NGSS, Common Core, and TEKS standards.
Generative AI as an Unsteady Tool for Instructional Development
Production Failures, Rebuilds, and Hard Lessons from AI-Assisted K-12 Science Media Development.
An NIH-funded SEPA project I worked on developing a graphic novel and five-chapter documentary video series for 7th through 9th grade environmental science. Building production around two commercial generative AI tools (DALL-E and MagicLight.AI) exposed systematic visual consistency failures: character ages and skin tones drifted between panels, a recurring dog changed breed, and lab equipment embedded in characters' bodies. The graphic novel required a full rebuild after 18 months. I trace these failures to architectural causes, mainly session-stateless generation, and apply Failure Mode and Effects Analysis to characterize eight failure modes. The contribution is showing how the K-12 grant context amplifies every platform instability into project-level risk, arguing that teams should evaluate cross-session consistency rather than single-image quality before committing federally funded workflows to vendor-controlled AI platforms.
Warren, S., & Taylor Schlitz, I. (In development). Dynamic Systems Engineering: A Holistic Approach to Supporting Complex Education and Media Design
