Do LLMs Make Mistakes Like Students? Exploring Natural Alignment between Language Models and Human Error Patterns

2025-02-21

AI Summary

This study investigates whether large language models (LLMs) exhibit cognitive alignment with students in multiple-choice question (MCQ) distractor selection—specifically, whether LLMs’ predicted probabilities of generating incorrect options (distractors) correlate with empirically observed student misselection frequencies. Method: Leveraging real-world educational datasets, we conduct probability distribution correlation analysis (Spearman’s ρ), cross-model comparison, and distractor-level alignment evaluation. Contribution/Results: We provide the first empirical evidence that, when LLMs err, they significantly prefer the most frequently selected student distractors—a robust cognitive bias alignment observed even in smaller-scale models. LLM-generated distractor probabilities show moderate positive correlation with student misselection distributions (ρ ≈ 0.4–0.6), confirming cross-scale consistency between model-generated and human misconceptions. These findings establish a novel, empirically grounded paradigm for automated, cognitively plausible distractor generation in educational assessment.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Analogy
• Natural Language Processing: (Large) Language Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large Language Models (LLMs) have demonstrated remarkable capabilities in various educational tasks, yet their alignment with human learning patterns, particularly in predicting which incorrect options students are most likely to select in multiple-choice questions (MCQs), remains underexplored. Our work investigates the relationship between LLM generation likelihood and student response distributions in MCQs with a specific focus on distractor selections. We collect a comprehensive dataset of MCQs with real-world student response distributions to explore two fundamental research questions: (1). RQ1 - Do the distractors that students more frequently select correspond to those that LLMs assign higher generation likelihood to? (2). RQ2 - When an LLM selects a incorrect choice, does it choose the same distractor that most students pick? Our experiments reveals moderate correlations between LLM-assigned probabilities and student selection patterns for distractors in MCQs. Additionally, when LLMs make mistakes, they are more likley to select the same incorrect answers that commonly mislead students, which is a pattern consistent across both small and large language models. Our work provides empirical evidence that despite LLMs' strong performance on generating educational content, there remains a gap between LLM's underlying reasoning process and human cognitive processes in identifying confusing distractors. Our findings also have significant implications for educational assessment development. The smaller language models could be efficiently utilized for automated distractor generation as they demonstrate similar patterns in identifying confusing answer choices as larger language models. This observed alignment between LLMs and student misconception patterns opens new opportunities for generating high-quality distractors that complement traditional human-designed distractors.

Problem

• Explores LLM alignment with human error patterns
• Investigates LLM and student distractor selection correlation
• Assesses LLM capability in educational assessment tasks

Innovation

• LLMs analyze student error patterns
• Correlate LLM mistakes with student misconceptions
• Use small models for distractor generation