Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 29, 2026International Journal of STEM EducationOpen Access

The impact of an LLM-based educational agent on learning achievement, cognitive dynamics, and student perceptions in computer science education

View Full Paper
Ask AI
Bookmark
Share

Authors

XLXu LiSouthwest Petroleum UniversityLZLiu ZhHunan UniversitySJShiyan JiangFujian Normal University

Discussion

Loading...

Member takes

Overview

Randomized trial examines the role of LLM-based agents on learning outcomes and student engagement in computer science courses, indicating positive effects.

Key Points

  • This study aims to explore the effects of an LLM-based educational agent on learning outcomes and cognitive dynamics in a computer science course.
  • Conducted a large-scale quasi-experiment with 313 sophomore students across four classes (three experimental, one control) over four weeks.
  • Employed statistical analyses for learning achievement and cognitive engagement levels.
  • Used lag sequential analysis (LSA) for cognitive behavior and structural equation modeling (SEM) for interpreting survey data.
  • The agent-enriched environment significantly improved learning achievement compared to traditional instruction.
  • LSA identified distinct cognitive trajectories, showing student-agent interactions had high-frequency engagement with a 'Query-Evaluation-Query' verification loop.
  • SEM confirmed that positive perceptions of the agent enhanced learner engagement through increased satisfaction.

Cite This Study

Li et al. (2026) studied this question.

synapsesocial.com/papers/6a69a2f1c8da07d9defa704ahttps://doi.org/10.1186/s40594-026-00641-y
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1LPITutor: an LLM based personalized intelligent tutoring system using RAG and prompt engineering2025 · 28 citations
  2. 2Cultivating independent thinkers: The triad of artificial intelligence, Bloom’s taxonomy and critical thinking in assessment pedagogy2025 · 60 citations
  3. 3A survey on large language model based autonomous agents2024 · 1,642 citations
  4. 4Distributing Accountability, Not Capability: Phase Separation and the LLM Workflow Quadrant in Autonomous AI Agent Architectures2022 · 596 citations
  5. 5The impact of AI-assisted pair programming on student motivation, programming anxiety, collaborative learning, and programming performance: a comparative study with traditional pair programming and individual approaches2025 · 102 citations