Participants Prefer Low-Noise Partners Over High-Noise Explorers
In a preregistered experiment published Dec 8, 2025 in PLoS Computational Biology, Morishita et al. tested whom people choose as observational reinforcement-learning partners and found that most participants preferred low-noise (consistent, high-performing) individuals. Exploratory analyses showed participants favoring low-noise partners tended to rely on imitation of observed actions, suggesting individual learning styles shape partner selection.
Key Points
- 1Finds that participants prefer low-noise (consistent, high-performing) partners when selecting observational learning partners.
- 2Indicates consistent partners appear more reliable and facilitate imitation, reducing uncertainty in observers' decision-making.
- 3Suggests designers should match learners with low-noise demonstrators to leverage imitation and improve learning outcomes.
Scoring Rationale
Empirical preregistered evidence for partner preference, but limited generalizability and no large-scale field validation in diverse populations.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
