CUDOS Norms for Group Engagement
Run your decision group like a scientific community
Sociologist Robert K. Merton, who began his career as a teen magician, developed a set of norms for scientific communities that Duke argues are the ideal blueprint for any truthseeking group. The acronym CUDOS stands for Communism (data belongs to the group), Universalism (evaluate ideas on merit, not source), Disinterestedness (guard against conflicts of interest, including cognitive ones), and Organized Skepticism (encourage productive dissent).
Each element addresses a specific failure mode of group decision-making. Communism prevents strategic withholding of information. Universalism prevents shooting the messenger or uncritically accepting ideas from favored sources. Disinterestedness prevents outcome bias and motivated reasoning. Organized Skepticism creates safe space for dissent and ensures that groupthink does not go unchallenged.
These norms are not natural -- groups default to the opposite of each. Members withhold embarrassing details, judge ideas by source rather than merit, let known outcomes bias their assessments, and suppress dissent to maintain harmony. Implementing CUDOS requires explicit agreement, ongoing reinforcement, and a willingness to feel uncomfortable in the short term for long-term accuracy gains.
- Data should belong to the group; secrecy is the antithesis of learning.
- Evaluate ideas based on their merit, not the identity of who proposed them.
- Everyone has conflicts of interest, including cognitive ones, and vigilance is required.
- Productive dissent is the highest form of group participation.
- These norms must be explicitly adopted because they oppose our natural social defaults.
- Implement Communism (full data sharing)Agree that members will share all relevant details when presenting decisions for review, including uncomfortable ones. If you have an urge to leave out a detail because it makes you uncomfortable, that is exactly the detail you must share. Reward members for sharing more, not less.Pro tipExpert poker players share extraordinary detail when workshopping hands -- position, pot size, opponent history, chip stacks. Use their thoroughness as a template.WarningThe Rashomon Effect means any single account is incomplete. Query members to extract details they may have omitted.
- Implement Universalism (evaluate ideas on merit)Separate the message from the messenger. When presenting information to the group, initially omit who said it or where it came from. Evaluate the substance before considering the source. Practice finding value in ideas from sources you would normally dismiss.Pro tipWhen you hear an idea from someone you dislike, imagine your most respected mentor said it. When you hear one from someone you admire, imagine your least favorite person said it.WarningDuke spent a year dismissing strategies from players she considered 'bad' based on superficial assessment, missing profitable learning opportunities.
- Implement Disinterestedness (guard against bias)Be vigilant about how known outcomes, personal stakes, and cognitive biases influence assessment. Present decisions for group review before outcomes are known when possible. When reviewing after the fact, withhold the outcome initially to prevent resulting.Pro tipPhysicist Richard Feynman advocated outcome-blind analysis because even particle physics is vulnerable to confirmation bias when researchers know the hypothesis being tested.WarningConflicts of interest are not only financial -- our brains have built-in cognitive conflicts that bias every assessment.
- Implement Organized Skepticism (encourage dissent)Make productive disagreement a valued behavior. When members disagree, have them argue each other's positions. Assign devil's advocate roles. Frame the goal as finding the most accurate answer, not reaching consensus. Make 'being the best heckler' as valued as 'being the best cheerleader.'Pro tipThe CIA uses 'red teams' specifically tasked with finding flaws in plans. Adapt this practice for your group.WarningWithout organized skepticism, groups drift toward groupthink where dissent is suppressed and only confirming information is surfaced.
In 1967, three Harvard scientists published a review in the New England Journal of Medicine that blamed fat rather than sugar for heart disease. It was later discovered that the sugar industry had paid the scientists to write the paper, which followed a pro-sugar, anti-fat methodology.
Duke's poker group routinely analyzed hands without revealing how they turned out. When she began teaching poker seminars and used this approach, newer players were shocked and desperate to know the ending. Duke would tell them 'It doesn't matter.'
Robert K. Merton (born Meyer R. Schkolnick) was a teen magician who performed as 'Robert Merlin' before becoming one of the most influential sociologists of the 20th century. His 1942 paper on the normative structure of science, refined over 31 years, established the CUDOS norms. Duke recognized that these norms, designed for scientific communities, mapped perfectly onto the unwritten rules of her poker learning group and could be applied to any group seeking better decisions.