Hypothesis
On a complex survival-ranking task, deliberating groups would outperform a statistical 'wisdom of crowds' aggregate, and weighting individual answers by confidence would improve aggregation, as recent studies on simpler tasks found.
What failed
Both crowds and groups beat individuals, but collective decision-making did not outperform the wisdom of crowds, and confidence-weighted aggregation showed no advantage over unweighted aggregation.
Broken assumption
Deliberation and confidence weighting improve group judgments regardless of task complexity and group size.
Why
Simulations varying group size and weighting sensitivity indicated the confidence-weighting null came from the small group size, not from the complexity of the survival task.
From the published paper: Hamada et al. Cognitive Research Principles and Implications, 2020. ↗CC BY