This study aims to analyze elementary science-gifted students’ completed artifacts from a three-dimensional (3D) design software mentorship program: scientific concept application, digital-based problem solving, and selfunderstanding and reflection. Using 69 artifacts produced by sixth-grade elementary students who performed an identical task (“Designing My Own Creative Air Fire-extinguisher”) as the object of analysis, we inductively typified examples of the three dimensions and calculated their frequencies. The results showed that students selected, adjusted, and redefined science concepts, centered on fluid mechanics, within the design context; discovered different problems for the same task; solved them using 3D modeling tools; and reflected on their abilities, limitations, and growth in forms grounded in their task performance experience. In particular, an orientation toward extending into further learning and inquiry appeared most frequently (46 cases), confirming that self-understanding and reflection at this stage were realized as experience-embedded self-perception and learning orientations, rather than as concrete career aspirations. This suggests that the competency changes identified through quantitative tests and interviews are embodied in the artifacts, whereas artifact analysis can also reveal manifestation patterns that are not captured by the item structure of the assessment instruments. This study demonstrates that student artifacts, when analyzed together with tests and interviews, can serve as a third data source, affording a multidimensional understanding of students’ actual competencies. It also proposes group-based mentorship education and design history analysis as future tasks.

Abstract
1. Introduction
2. Theoretical Background
    2.1. Mentorship Education and Artifact-BasedLearning
    2.2. Exploring Competencies through ArtifactAnalysis
    2.3. Three Competency Dimensions Manifestedin Artifacts
3. Research Method
    3.1. Participants and the Mentorship EducationProgram
    3.2. Data Collection and Object of Analysis
    3.3. Analytic Framework and AnalyticProcedure
4. Results
    4.1. Science Concept Application
    4.2. Digital-Based Problem Solving
    3.3. Self-Understanding and Reflection
5. Discussion
    5.1. Artifact Manifestation of Self-Understandingand Reflection: Experience-Based Concreteness
    5.2. Problem Finding and Variation for theIdentical Task
    5.3. The Data Value and Limitations of DigitalArtifacts
    5.4. The Scarcity of Collaboration and FutureTasks
    5.5. Conclusion and Suggestions
Acknowledgments
References

• Jinju Hwang(Pohang Heunghae Technical High School, Pohang 37544, Korea)
• Hyonyong Lee(Department of Earth Science Education and Science Education Research Institute, Kyungpook National University, Daegu 41566, Korea) Corresponding author