Name
Improving the Assessment Process During Matchy-Matchy, a Team-Based Activity to Reinforce Connections, with the Help of AI
Date & Time
Tuesday, June 9, 2026, 10:57 AM - 11:12 AM
Location Name
Oglethorpe F
Authors
Mark Hernandez, ETSU Quillen College of Medicine Suman Dalal, ETSU Quillen College of Medicine Tom Ecay, ETSU Quillen College of Medicine
Presentation Topic(s)
Technology and Innovation
Description
PURPOSE
“Matchy-Matchy” (MM) is a team-based activity where learners can match
facts, drug mechanisms, and patient case features. After completing MM tasks,
teams submit results (e.g., photos), and a selected team presents their
findings. In prior MM implementations, allowing learners to use chatbots for
immediate feedback did not enhance the overall learning experience. This
study explored an alternative application: using generative AI chatbots to
aid faculty in evaluation of team submissions (photos or images).
METHODS
MM activities were integrated into selected TBL sessions within the
2nd-year medical curriculum (Brain, Body and Behavior and
Endocrine/Reproduction courses). Preliminary findings suggested that AI
chatbots (e.g., ChatGPT, Gemini) could be resourceful in the assessment
process, with varying levels of effectiveness depending on the type of
Chatbot used.
RESULTS
From an evaluator’s perspective, faculty had to manually assess all MM
results to identify unique team errors. AI chatbots were tested against the
manual process and accurately identified the errors. The chatbot consistently
flagged specific mismatch categories, such as incorrect mechanism of drug
action and adverse effects. By identifying these errors, the AI provided
faculty with insight into student misconceptions. Analysis of these findings
and student feedback prompted several updates to the activity process.
CONCLUSION
The application of generative AI tools to support educators in evaluating
team decisions during MM exercises is a valuable strategy to improve the
quality of the learning experience by increasing the efficiency of session
time and reducing faculty assessment burden. This process also shows
potential to provide immediate, targeted feedback to learners and has led to
important modifications, including spacing the cases and introducing tailored
board-style questions between cases. As generative AI becomes integral to
clinical practice and serves as a diagnostic safety net, developing these
critical thinking skills through activities like MM will be increasingly
meaningful for health sciences learners.