Characteristics of generated immediate feedback of teachers' improvisational decisions within agentic simulations to support mathematics instruction

• In Event: AERA Poster Session Wednesday 3:45pm
  In Poster Session: Modeling, Measurement, and Design in Technology-Enhanced Learning Environments

Abstract

This paper characterizes feedback generated for teachers rehearsing improvisational decision-making during a student-centered mathematics discussion in a GenAI-powered simulation. The analysis sought to describe the nature of the feedback concerning high and low-quality pre-designed teacher questions. Across 12 iterations of the same simulation using a sample representation of teacher responses, a majority of the feedback to high-quality responses placed a strong emphasis on the value of incremental time-expansive clarifications of students’ mathematical ideas. In contrast, the feedback to low-quality responses emphasized changing time-efficient corrections to time-expansive clarifications. With these results, the paper recommends future development of design-analysis pairings of generated feedback within simulations that offer incremental feedback to teachers’ improvisational decision making, rooted in expertise in mathematics teacher education.

Author

• Erin Barno, ETS