Learning AI from your own movement — sensors, learning and social impact taught after school to upper elementary students

1Checking an abstract idea against your own body

Teaching AI to elementary students tends toward the abstract. Data, learning, models — none visible, none touchable. The approach here is to make the teaching material out of the students own movement. Running, jumping, stopping. That movement becomes numbers through a sensor, and from those numbers a computer judges something. When what you did turns into data in front of you, the question of what data is becomes concrete.

2Three concepts in order

1. 1First conceptHow computers perceive the world through sensors
2. 2Second conceptHow AI systems learn from data
3. 3Third conceptHow AI can impact society
4. 4Research questionIn what ways can upper elementary AI learning be supported through learner-driven investigation of data generated from their own embodied movement in an after-school program

That social impact comes third says something about the character of the program. How a thing works and how it acts in society are not separated into different lessons. The stance is that technical understanding and the judgment of a citizen living with the technology are learned together from the start rather than added on later.

3Three years of refinement

Implementation runs across three years serving approximately 360 upper elementary students, gathering evidence on learning, self-efficacy and program effectiveness while continuously refining the experiences. It is designed on the assumption of being fixed while in use rather than built once and distributed.

The interdisciplinary team combines computer science, educational technology and biomechanics — the last following from working with bodily movement.

4What gets published

Outcomes are to include new knowledge about effective approaches to elementary AI education, a freely available AI-PLAY curriculum and instructional materials, and a publicly available educational movement dataset supporting future research and broader adoption.

Of the NSF awards this site holds as of 2026-09-02, four fall under AISL, and this site also covers another award in that program, on a learning environment where young people work with an AI coding agent. Amounts are the obligated amount as of the check date and may change.