Facilitator-Mediated Versus AI-Mediated Problem-Based Learning in Pharmacology Education: A Quasi-Experimental Study Among Second Year Medical Students

In the classrooms where future doctors learn their craft, the method of teaching has long been a subject of careful debate. For decades, medical educators have moved away from simple memorization of facts toward a style of learning called problem-based learning. In this approach, students do not sit passively while a lecturer speaks; instead, they gather in small groups to tackle complex, real-world clinical scenarios. They must work together to identify what they do not know, find the answers, and apply that new knowledge to solve the problem. This process is designed to build critical thinking and teamwork, skills just as vital as knowing the names of drugs or diseases. Traditionally, these sessions are guided by a human teacher, known as a facilitator, who asks questions, keeps the group on track, and ensures every student participates. However, the rapid rise of artificial intelligence has introduced a new variable into this equation. Students now have access to powerful digital tools that can instantly summarize information, generate ideas, and answer questions. The question facing modern medical education is whether these digital assistants can replace the human guide, or if they serve a different purpose entirely.

A recent study conducted at two medical colleges in India set out to compare these two paths directly. Researchers worked with 150 second-year medical students who were learning about pharmacology, the science of how drugs work in the body. The students were divided into two groups to tackle the same set of clinical problems over a four-week period. One group followed the traditional path, working through their case studies with a human facilitator present to guide the discussion. The other group worked with the same clinical problems but used artificial intelligence tools to help them find information and structure their thinking, without a human teacher leading the session. The goal was to see which method produced better results in terms of test scores, the ability to think critically, and how well the students worked together as a team.

The results revealed a nuanced picture of how learning happens. When the students took tests before and after the four weeks, both groups showed significant improvement. The students who worked with a human teacher improved their scores, and the students who worked with artificial intelligence improved their scores by a very similar amount. In terms of raw knowledge gained, the two methods were equally effective. However, the story changed when the researchers looked deeper at how the students behaved and thought during the process. The groups led by human teachers performed significantly better in specific areas of teamwork and critical thinking. They were more effective at collaborating, responding to feedback, and solving problems in a structured way. The students using artificial intelligence, while successful in gathering information, showed lower scores in these collaborative and analytical skills. The human-guided groups demonstrated a stronger ability to explain their reasoning and analyze the information they collected, suggesting that the presence of a teacher helps students refine their thinking in ways that a digital tool alone does not.

Despite the differences in how the students performed, their satisfaction with the experience was high in both groups. The students who used artificial intelligence reported that they had better access to learning materials and found the discussions more productive in terms of the volume of information they could explore. They appreciated the speed and breadth of resources the technology provided. Yet, the human-guided groups felt that the teacher's presence was crucial for their learning. This tension was also visible in the views of the faculty members who observed the study. While many teachers recognized that students were using these tools to generate answers and create presentations, they expressed deep concern about the risks. The primary worries were that students might rely too heavily on the technology, potentially bypassing the hard work of thinking for themselves, and that the use of these tools could lead to issues with academic honesty. The faculty noted that while the technology could enhance the quality of a discussion, it also risked undermining the core goal of problem-based learning, which is to develop the student's own ability to reason through a problem.

The study concludes that neither method is a perfect replacement for the other. Artificial intelligence is a powerful resource that can make learning more accessible and efficient, allowing students to gather vast amounts of information quickly. However, it does not yet fully replicate the benefits of human guidance in fostering deep critical thinking and effective teamwork. The most promising path forward appears to be a hybrid model. In this approach, the efficiency and resourcefulness of artificial intelligence are combined with the mentorship and guidance of a human teacher. This blend would allow students to use technology to handle the heavy lifting of information gathering while relying on a facilitator to ensure they are thinking critically, working well together, and truly understanding the material. Such a strategy could offer the best of both worlds, preparing medical students for a future where technology is ubiquitous, while ensuring they retain the essential human skills required to care for patients.