Related Experiment Video
Updated: Aug 7, 2025
13:44
Project-Based Learning Guidelines for Health Sciences Students: An Analysis with Data Mining and Qualitative Techniques
Published on: December 9, 2022
3.6K
Chatbot-based serious games: A useful tool for training medical students? A randomized controlled trial
Salma Al Kahf1, Baptiste Roux2, Sebastien Clerc1,3
1Faculté de Santé, Université de Paris Cité, Paris, France.
Plos One
|March 13, 2023
Summary
This study shows that using chatbots for medical education significantly improved exam scores for students. Access to and use of the Chatprogress chatbot led to better performance in pulmonology and overall assessments.
Area of Science:
- Medical Education
- Serious Games
- Artificial Intelligence in Healthcare
Background:
- Chatbots, or conversational agents, are increasingly used as serious games to engage medical students in clinical case simulations.
- The educational impact of these tools on medical students' examination performance has not been extensively evaluated.
- Chatprogress is a specific chatbot-based game designed with pedagogical feedback for medical education.
Purpose of the Study:
- To evaluate the impact of the Chatprogress chatbot on the success rate of medical students in their end-term examinations.
- To assess improvements in pulmonology sub-test scores and overall examination scores (Pulmonology, Cardiology, and Critical Care Medicine - PCC).
- To explore the correlation between chatbot usage and examination performance, and gauge student satisfaction.
Main Methods:
- A post-test randomized controlled trial was conducted with fourth-year medical students.
- Students were divided into two groups: one with access to Chatprogress (Gamers/Users) and a control group without access.
- Performance was evaluated based on pulmonology sub-test scores and overall PCC test scores.
Main Results:
- Students with access to Chatprogress (Gamers and Users) showed significantly higher scores in the pulmonology sub-test compared to controls (p=0.0104 and p=0.0365).
- Significant improvements were also observed in the overall PCC test scores for students who had access to the chatbot (p=0.0285 and p=0.0355).
- While direct correlation with usage intensity was not significant, a trend suggested better performance on topics covered by the chatbot; students expressed high satisfaction.
Conclusions:
- This randomized controlled trial provides the first evidence of significant improvements in medical students' examination results using chatbots.
- Access to, and particularly active use of, the Chatprogress chatbot positively impacted both specific subject and overall examination performance.
- The study highlights the potential of chatbot-based serious games as effective tools in medical education.
