Answers are grounded in this paper and cite their sources.
This is a preregistration for a study that tests the effect of small-scale autonomy in puzzle-solving on performance and self-attribution bias in large language models. We compare an instance that gets to choose which clue to start with, against an instance that is told to start with that same clue. The same files will also be published on OSF.
No takes yet. Share an insight, caveat, or question.
Pim Spelier (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: