Enhancing medical education with AI-powered simulations: Leveraging scaffolded learning and adaptive technology

How does AI intersect with medical training?

Medical education is continually evolving due to rapid technological advancements, transformations in healthcare environments, and the growing expectations placed upon medical trainees. Traditionally, the cornerstone of medical teaching was lectures and textbooks. In 2025, however, learners require dynamic, interactive experiences to effectively bridge the gap between foundational knowledge and real-world clinical practice. Artificial intelligence (AI) has emerged as a tool for reshaping and bolstering medical education. Specifically, AI-powered healthcare simulations (HCS) offer promising opportunities to significantly enhance individualized learning, accelerate development of clinical reasoning, and foster students' sense of autonomy and preparedness (1).

Research shows that technology-based teaching methods have become increasingly important as medical curricula grow more complex. AI-powered learning tools have gained popularity for various applications in medical education, such as personalized feedback and assessment, case-based learning, and test-preparation (2,3). One such platform that embodies these principles is DDx by Sketchy, which provides adaptive guidance through immersive, AI-driven clinical scenarios designed explicitly to scaffold the learner's experience.

What is scaffolded or mastery learning in medical education?

To address the tension between covering vast amounts of information and effectively developing learners' critical reasoning skills, many health sciences programs have adopted educational frameworks that actively support learners through graduated, responsive interventions, such as scaffolded learning and mastery learning (4).

Historically rooted in the educational theories of Vygotsky, scaffolded learning is an instructional framework in which students are provided tailored support and guidance as they progress in their learning journey. The idea is to offer structured assistance at the beginning of the learning process and gradually reduce support as learners gain competence and confidence (5, 6). A related concept is mastery learning, in which students must achieve a set level of proficiency before advancing to more complex tasks or concepts (7). By integrating both scaffolded learning and mastery learning, educators can ensure that learners build a robust foundation of knowledge and skills at each stage, ultimately fostering deeper clinical reasoning and better performance in demanding healthcare environments.

How is scaffolded learning implemented in medical education?

1. Early Exposure to Foundational Knowledge: At the start of medical education, students are introduced to core concepts through lectures, textbooks, and guided discussions. Instructors deliberately sequence learning activities to gradually build in complexity. By breaking content into manageable steps, providing timely feedback, and offering tools such as concept mapping, interactive visual aids, and guided reading assignments, educators ensure that each new skill builds on previously mastered ones (5).
2. Simulation-Based Learning: As students advance, they engage in simulation-based learning, where they apply theoretical knowledge to practical, real-world scenarios. AI-powered simulations enhance this experience by offering dynamic feedback, adaptive difficulty levels, and personalized case progression (8).
3. Progressive Reduction of Guidance: As learners become more proficient, instructors reduce direct guidance, allowing students to take the lead in clinical reasoning and decision-making. This stage is crucial for transitioning from knowledge acquisition to real-world application (4).
4. Peer Learning and Collaboration: Scaffolded learning is also reinforced through peer discussions, mentorship programs, and case-based learning sessions. Senior students or educators provide initial guidance, but as learners progress, they engage in collaborative problem-solving (9, 10).
5. Assessment and Reflective Practice: Regular assessments, including self-assessment, formative quizzes, and instructor feedback, help students track their progress. AI-powered platforms use data-driven insights to identify knowledge gaps and suggest personalized learning pathways (11).

How does DDx by Sketchy implement AI-powered scaffolded medical education?

Interactive case guidance

A key element of scaffolded learning is providing learners with personalized support at the right times. DDx by Sketchy exemplifies this principle through its case guidance feature, which offers incremental hints and prompts to guide clinical reasoning without simply presenting the final answer. Hints are personalized based on the learner's responses, ensuring targeted assistance rather than generic explanations. Rather than providing solutions, the platform nudges students toward the correct thought process, reinforcing diagnostic skills.

Real-time feedback and performance analytics

A key feature of DDx is its ability to engage students in dynamic conversations that challenge them to think critically, justify their clinical decisions, and identify key findings. The student receives real-time feedback, highlighting errors and offering explanations for incorrect responses. This immediate feedback loop is crucial for self-directed learning, as it allows learners to understand their mistakes, adjust their reasoning, and improve their decision-making skills. DDx utilizes this feedback to generate performance analytics, allowing educators to track students' progress over time and identify areas needing improvement.

Scenario-based learning and clinical decision-making

DDx presents learners with diverse clinical cases, each requiring unique decision-making approaches. By engaging with a wide array of patient scenarios, students develop diagnostic acumen, refine treatment planning skills, and learn to adapt their approach based on evolving clinical presentations. Instructors have the flexibility to select cases from the DDx library according to complexity, subject matter, or curricular focus.

On-demand learning and flexibility

Traditional classroom settings have limitations in terms of scheduling and availability of instructors. DDx overcomes these barriers by allowing students to engage in learning at their own pace, anytime and anywhere. This flexibility supports self-directed learning by enabling learners to revisit difficult cases, practice repeatedly, and deepen their understanding without time constraints.

The future of medical education is adaptive, scaffolded, and AI-powered

AI-powered simulations are revolutionizing medical education by providing dynamic, interactive, and personalized learning experiences. DDx by Sketchy exemplifies how AI-driven tools can incorporate scaffolded learning principles to strengthen learners' knowledge and encourage critical thinking. Through strategic implementation of AI-powered tools like DDx, medical educators can strengthen students' self-directed learning, cultivate advanced clinical reasoning skills, and ultimately prepare them for the nuanced complexities of patient care.

Frequently asked questions

What is scaffolded learning and how does it apply to medical education?

‍Scaffolded learning is an instructional framework rooted in Vygotsky's educational theory, in which students receive structured support at the start of learning and gradually become more independent as competence develops. In medical education, this translates to sequenced case complexity, adaptive hints, guided reflection, and feedback that targets specific reasoning gaps rather than simply confirming correct answers. The goal is to build learners from dependent knowledge-receivers to independent clinical thinkers.

What is mastery learning in clinical training and why does it matter?

‍Mastery learning requires students to demonstrate a defined level of proficiency before advancing to more complex content or responsibilities. In clinical training, this approach ensures that foundational reasoning skills — history-taking, differential generation, diagnostic prioritization — are genuinely consolidated before learners face higher-stakes environments. Without mastery checkpoints, students may advance with reasoning gaps that compound under the pressure of real patient care.

How does AI simulation support self-directed learning in medical students?

‍AI simulation platforms like DDx enable self-directed learning by allowing students to engage with cases asynchronously at their own pace, receive immediate structured feedback without faculty presence, and revisit difficult cases as many times as needed. The immediate feedback loop — tied to specific reasoning decisions rather than final answers — helps students identify exactly where their thinking diverged and self-correct before the gap compounds.

How do AI-driven performance analytics help medical faculty identify struggling students?

‍DDx generates longitudinal performance data at the level of individual reasoning steps — not just final diagnostic accuracy. Faculty can see where students consistently struggle: whether in generating a differential, ordering hypothesis-driven tests, or interpreting results in context. Cohort-level analytics allow programs to distinguish individual learners who need targeted support from systemic patterns that suggest a curriculum gap, enabling more precise and earlier intervention than episodic faculty observation allows.

References

1. Hamilton A. Artificial Intelligence and Healthcare Simulation: The Shifting Landscape of Medical Education. Cureus. 2024;16(5):e59747.
2. Chan KS, Zary N. Applications and Challenges of Implementing Artificial Intelligence in Medical Education. JMIR Med Educ. 2019;5(1):e13930.
3. Preiksaitis C, Rose C. Opportunities, Challenges, and Future Directions of Generative Artificial Intelligence in Medical Education. JMIR Med Educ. 2023;9:e48785.
4. Masava B, Nyoni CN, Botma Y. Standards for Scaffolding in Health Sciences Programmes. J Med Educ Curric Dev. 2023;10.
5. Masava B, Nyoni CN, Botma Y. Scaffolding in Health Sciences Education Programmes: An Integrative Review. Med Sci Educ. 2023;33(1):255-273.
6. Vygotsky LS, Cole M. Mind in Society. Harvard University Press; 1978.
7. Bloom B. Human Characteristics and School Learning. McGraw-Hill; 1968.
8. Bauer E, Heitzmann N, Fischer F. Simulation-based learning in higher education. Studies in Educational Evaluation. 2022;75:101213.
9. Visser CL, et al. Scaffolding Clinical Reasoning of Health Care Students. J Med Educ Curric Dev. 2020;7.
10. Burtson KM, et al. Academic Coaching to Promote Self-Directed Learning in Graduate Medical Education. J Gen Intern Med. 2025.
11. Wang HS, Chen S, Yen MH. Effects of metacognitive scaffolding on students' performance. Phys Rev Phys Educ Res. 2021;17(2).