Research map / Large language models / AI and society
Supporting Upper Elementary Students in Learning AI Concepts with Story-Driven Game-Based Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Presents PRIMARYAI, a story-driven game-based AI education environment for upper elementary students with a Use-Modify-Create scaffolded AI planning quest, and reports a classroom study with 16 fourth graders.
Problem
Introducing AI concepts to upper elementary students (ages 8-11) in engaging, effective ways, and determining whether scaffolding in a story-driven game supports learning of AI planning.
Approach
A story-driven science-inquiry game (penguin population decline) with a block-based AI planning interface, whose quests are scaffolded by the Use-Modify-Create framework, with the planning quest extended from three to five missions to ease the transition to Create.
Key results
- Post-test scores (M=5.43, SD=1.87) significantly exceeded pre-test scores (M=4.36, SD=1.98), t(13)=3.51, p<.01, for 14 students with both tests.
- 93.75% of students completed the Use phase, 66.67% of those completed the Modify phase, and 60% of those completed the Create phase.
- 93.33% of students who completed Mission 1 also completed Mission 2, and all students who completed Mission 2 completed Mission 3.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use PRIMARYAI's AI Planning quest | Fourth-grade students · AI Planning content knowledge assessment | score: 4.36 -> 5.43 | Helped |
| Use the Use-Modify-Create progression across the AI Planning quest | PRIMARYAI · AI Planning quest mission completion | completion rate: 93.75%; 66.67%; 60% | Helped |
| Require adding or deleting blocks in Mission 3 | PRIMARYAI Mission 3 · AI Planning quest | time to complete: 124.09 -> 542.53 | Hurt |
| Add two missions to the Modify phase before the Create phase | PRIMARYAI · AI Planning quest | student success transitioning between phases | Helped |
Surprising findings
- Students moved around 10 blocks on average in Mission 1 even though no changes were required, suggesting exploration or misunderstanding of the task.
- Students who completed Mission 4 attempted it at least twice on average, while those who failed attempted it only once, suggesting persistence rather than premature abandonment.
Limitations
- Small sample of 16 students (14 in learning gain analysis) from a single fourth-grade classroom in one school.
- Technical networking issues forced restarts and interrupted gameplay.
- Qualitative analysis covered only about one hour of video; generalizability not established.
Experimental setup
Datasets: pre-test and post-test on AI planning (11 multiple-choice questions) gameplay trace logs classroom video recordings
Metrics: paired t-test on pre/post test scores mission completion rates time spent per mission blocks moved per mission Pearson correlation of mission times
Builds on: Use-Modify-Create framework
Closest papers by meaning
- “From Unseen Needs to Classroom Solutions”: Exploring AI Literacy Challenges & Opportunities with Project-Based Learning Toolkit in K-12 Education (AAAI 2025 · similarity 0.87)
- An Effectiveness Study of Teacher-Led AI Literacy Curriculum in K-12 Classrooms (AAAI 2024 · similarity 0.86)
- Shaping AI Interest in Rural Middle Schools with Unplugged Learning: Gender Differences and Teacher Insights (AAAI 2025 · similarity 0.86)
- Understanding K-12 Teachers’ Needs for AI Education: A Survey-Based Study (AAAI 2025 · similarity 0.86)
- Fostering Epistemic Insights into AI Ethics through a Constructionist Pedagogy: An Interdisciplinary Approach to AI Literacy (AAAI 2025 · similarity 0.86)
- Word2Vec4Kids: Interactive Challenges to Introduce Middle School Students to Word Embeddings (AAAI 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.