← All Findings
arxiv 80% relevant

Using Large Language Models to Assess Tutors' Performance in Reacting to Students Making Math Errors

Sanjit Kakarla, Danielle Thomas, Jionghao Lin, Shivang Gupta, Kenneth R. Koedinger

learningai-educationinstructional-design

Abstract

Research suggests that tutors should adopt a strategic approach when addressing math errors made by low-efficacy students. Rather than drawing direct attention to the error, tutors should guide the students to identify and correct their mistakes on their own.

Summary

Relevance to Cognitive Load Theory

This paper addresses error handling in tutoring, a critical moment for cognitive load management:

Error Response and Cognitive Load

  • Direct error correction can increase extraneous load by disrupting the student’s thought process
  • Guided self-correction maintains productive germane load

Student Efficacy and Load Tolerance

  • Low-efficacy students have less capacity to handle the additional load of error feedback
  • Strategic approaches that protect student confidence also protect cognitive resources

AI-Assisted Quality Assurance

  • LLMs can help identify when tutors may be creating unnecessary cognitive load
  • Real-time assessment enables immediate adjustment of tutoring strategies

Instructional Design Implications

  1. Error feedback should be calibrated to student efficacy levels
  2. Indirect guidance preserves working memory for productive learning
  3. Automated assessment can help scale effective tutoring practices

Read original