People's lived experiences provide intuitions about health. Can they transform these personal intuitions into testable hypotheses that could inform both science and their lives? This paper introduces an online learning architecture and provides system principles for people to brainstorm causal scientific theories. We describe the Learn-Train-Ask workflow that guides participants through learning domain-specific content, process training to frame their intuitions as hypotheses, and collaborating with anonymous peers to brainstorm related questions. 344 voluntary online participants from 27 countries created 399 personally-relevant questions about the human microbiome over 4 months, 75 (19%) of which microbiome experts found potentially scientifically novel. Participants with access to process training generated hypotheses of better quality. Access to learning materials improved the questions' microbiome-specific knowledge. These results highlight the promise of performing personally-meaningful scientific work using massive online learning systems.

Learners worldwide collectively spend millions of hours per week testing their skills on assignments with known answers. Might some of this time fruitfully be spent posing and exploring novel questions? This paper investigates an approach for learners to contribute scientific ideas. The Gut Instinct system embodies this approach, hosting online learning materials and invites learners to collaboratively brainstorm potential influences on people's microbiome. A between-subjects experiment compared the performance of participants who engaged in just learning, just contributing, or a combination. Participants in the learning condition scored highest on a summative test. Participants in both the contribution and combined conditions generated novel, useful questions; there was not a significant difference between the two. Though participants in the combined condition both learned and contributed, this setting did not exhibit an additive benefit, such as better learning in the combined condition. These results highlight the promise and difficulty of double-bottom-line learning experiences.