THE REFERENCE LIBRARY
A growing collection of ideas for understanding the world — and your own thinking.
Paterson argues that calling a failure a culture problem does not finish the causal analysis. Investigate how goals, incentives, processes, resources, and external pressures produced the behavior being described.
The cum hoc ergo propter hoc fallacy occurs when a causal relationship is assumed between two events simply because they occur simultaneously, without sufficient evidence. It warns that correlation does not imply causation.
Curation is the process of selecting and presenting a focused set of options to help people make choices. It is a key strategy for brick-and-mortar retailers to compete with online giants by offering quality and serendipity rather than endless variety.
Keep a short list of questions you care about and use it to guide your reading, conversations, and experiments. Refine broad interests into specific questions, then revise them as evidence changes what you understand.
Keep a written set of questions that matter to you, then use new reading, conversations, and observations to refine them. The list can direct attention and suggest connections. Those connections become useful hypotheses when you ask what evidence would support or challenge them.
Curiosity encourages questions and exploration beyond immediate practical goals. The passage argues that education and continued inquiry can support learning throughout life.
The author calls detailed, questioning preparation a curious mind. The approach breaks a plan into tasks, considers what could go wrong, and checks practical dependencies before an event takes place.
The cognitive bias where knowledgeable individuals struggle to imagine what it's like to lack that knowledge, causing them to overestimate others' understanding and communicate ineffectively.
The Home Run Derby curse attributes a later batting slump to the contest itself. To assess that claim, distinguish a causal effect from selection of unusually strong hitters, regression toward typical performance, and selective recall.
Repeated processes can accumulate changes too small to notice in a single round. Dennett connects this pattern in making tools, biochemical cycles, and evolution, and suggests looking for cycles behind apparently mysterious complexity.
An organizational account describes Cynefin becoming a shared way of thinking, supporting safe-to-fail experiments and reflection on how people act and interact.
Cynefin is a sense-making decision framework that categorizes systems into Ordered, Complex, and Chaotic domains based on constraint types, and provides guidance on appropriate actions and transitions between domains.
The author connects Cynefin with Daoist fluidity: respond to the situation, acknowledge uncertainty, and avoid forcing a rigid plan where inquiry and adaptation are needed.
The author introduces group exercises for exploring Cynefin’s Clear, Complicated, and Complex domains. The supplied passage gives facilitation and debrief directions, but not the exercises themselves.
The Cynefin Framework is a decision-making tool that categorizes problems into five domains: Clear, Complicated, Complex, Chaotic, and Confused. It helps identify the appropriate approach—best practices, good practices, sense-making, or rapid response—based on the level of uncertainty and cause-effect relationships.
Cynefin uses stories to examine patterns that shape decisions. It asks groups to notice when past experience traps them in one interpretation and to create opportunities for other perspectives.
The Cynefin framework categorizes problems into five domains; in the Complex domain, organizations can use experiments like advice calls to probe and sense, enabling emergent learning and adaptation.
The Cynefin Framework can be used to design a balanced innovation portfolio that spans different domains (Clear, Complicated, Complex, Chaotic) and timeframes (tactical, strategic, horizon), incorporating safe-to-fail experimentation and a barbell strategy to manage risk and reward.
Cynefin is a sense-making framework that categorizes problems into domains (simple, complicated, complex, chaotic, disorder) to guide decision-making. Applied to software, it helps teams embrace uncertainty and choose appropriate approaches—from agile experimentation in complex domains to plan-driven methods in complicated ones.
The authors use Cynefin to examine workplace well-being through relationships, working practices, policies, and structures. Surveys can describe reported well-being, but a score alone does not explain what needs to change. Their approach combines those measurements with attention to the conditions employees experience.
Our days are composed of multiple overlapping rhythms—biological, environmental, and social. Recognizing and harmonizing with these rhythms can improve anticipation, navigation, and emotional well-being.
Cheng describes arguments in which someone replaces a stated position with a harsher supposedly equivalent claim, then attacks the speaker for it. Checking the alleged equivalence can expose where the accusation goes beyond what was said.
Emphasizing common ground and using the steel man approach—considering the strongest version of an opposing argument—can make others more open to changing their minds, as demonstrated by expert debaters and negotiators.
Data aggregation makes information easier to obtain and analyze, but group averages conceal differences within groups. An area’s average income, for example, does not describe the income of every resident.
Data can create economic value when an organization uses it to improve decisions or operations. The author’s potential-asset framing asks what the data will be used for, who can analyze it and whether the resulting information will justify its cost.
Data preparation includes checking missing entries, correcting errors, and investigating unusual values. Imputation fills gaps under explicit assumptions; outlier screening flags observations for review. Neither filling every gap nor removing every extreme value guarantees a sound analysis.
Businesses capture data through everyday operations and obtain specialty data through tracking or research. Mapping the organizations that contribute this information helps identify useful sources for a decision.
Data mining bias occurs when patterns or causal relationships are falsely identified in random or coincidental data, leading to superstitious beliefs and conspiracy theories. It exploits human tendency to see structure where none exists.
Sampling determines which observations support a decision and which population the findings can describe. Define that population, examine who is reachable and who is missing, and choose a sample size and time window suited to the question.
The tendency to find patterns in historical data by testing many trading rules, leading to false confidence in strategies that are merely overfitted to past data and suffer from survivorship bias.
Routine decisions may be suitable for automation when outcomes are defined and relevant data support reliable evaluation. Strategic decisions often require more judgment about goals and uncertainty. In both cases, decision quality depends on the data, the method, and checks on its limitations.
Dawes’ Rule gives cues equal weight: it counts favorable binary cues and subtracts unfavorable ones. The calculation is simple, but it still requires examining every cue. Its usefulness depends on the accuracy gained for that effort.
The dead salmon fMRI study illustrates how failing to correct for multiple comparisons in large datasets can produce statistically significant but meaningless results, highlighting the danger of false positives in scientific research.
Human beings are inherently egocentric, favoring themselves subconsciously. To counteract, ask what a rational person would do in a similar situation.
People often try to correct perceived biases in their judgments, but their debiasing attempts frequently fail, leading to insufficient correction, unnecessary correction, or overcorrection—sometimes resulting in judgments that are more biased than the original.
The planning fallacy can persist even when people estimate aggregate completion rates or imagine plausible delays. In the studies described here, linking a specific upcoming assignment to the timing of similar past assignments eliminated the usual optimistic bias for that task.
Debiasing programs aim to improve judgment by changing how people approach a task. The source describes military, educational and social-skills programs while emphasizing how difficult it is to isolate the effect of judgment training on real-world behavior.
People rely on naive theories to detect and correct bias in their judgments, but they are more likely to attribute bias to others than to themselves due to motivational and cognitive factors, making self-debiasing difficult even among experts.
In many Bronze Age societies, rulers systematically forgave agrarian debts to preserve social cohesion, prevent creditor control, and maintain a free population for taxes and military service. This practice acknowledged that not all debt is due to personal failing, but often results from systemic pressures like drought or war.
Checking the sequence of events can challenge a causal story, but earlier occurrence alone does not establish causation. Granger analysis asks whether past values of one variable improve forecasts of another, given the information included in the model.
Deceptive cropping removes context from a video or quotation so that it appears to support a different claim. Checking the longer recording can reveal the omitted qualification, question, or population being discussed.
Taleb uses Roman decimation to discuss an agency problem: someone may benefit from abandoning a group while others bear the loss. He presents the threat of shared punishment as a deterrent; the example raises questions about incentives, responsibility, and collective punishment.
Decision support supplies information for a person to use; decision automation also selects or carries out an action. Automation can use fixed rules, statistical models, or learning systems. Repeated decisions with clear outcomes and feedback are useful candidates for evaluation.
Evaluate a decision using the information, alternatives, and reasoning available at the time. Record that reasoning so later outcomes can help you improve the process without letting hindsight turn every success into proof of skill or every failure into proof of a bad choice.
Decision evasion (buck-passing, procrastination, obfuscation) occurs when decision makers are accountable to conflicting, mutually contemptuous audiences of equal power, especially when institutional norms permit avoidance. Experimental evidence shows that people will avoid decisions even when alternative options are unattractive.
A matrix lays out combinations of selected conditions or choices so each can be examined. Its coverage depends on which dimensions and possibilities you include.
A decision swear jar is a precommitment contract where you identify language and thinking patterns that signal irrationality or bias, and use those as triggers for immediate reflection to maintain truthseeking.
Etheredge and Levy weigh the costs, risks, and possible regret involved in suing over a competing product. Their dilemma could be mapped in a decision tree, although no tree or probability estimates are supplied.
Decision uncertainty concerns what we do not know about a choice and its possible outcomes. Banasiewicz’s idea of a degree of knowability asks how much of that uncertainty useful facts and probability estimates could reduce.
Taleb distinguishes doubts we cannot act on from uncertainties we can prepare for. His practical question is how to limit vulnerability and turn imperfect knowledge into action without needing a precise forecast.
Decision weights describe how probabilities influence choices, rather than what probability people believe an event has. People may give small chances disproportionate weight and respond especially strongly to a change from near certainty to certainty.
WRAP organizes decision making around widening options, reality-testing assumptions, attaining distance before deciding and preparing to be wrong. It offers practical checks against narrow framing, confirmation bias, short-term emotion and overconfidence.
Match the response to the decision obstacle. Compare a few serious options, test important assumptions, bring the organization’s priorities into debate, and agree in advance on conditions that would trigger reconsideration.
Neuroscientific research by Antonio Damasio reveals that emotions are essential for decision-making; individuals with impaired emotional processing struggle to make even simple choices. This principle challenges the assumption that logic alone drives decisions, especially in negotiation, where understanding emotional factors is key.
Treating decisions as bets involves explicitly recognizing that every choice is a bet on alternative futures, each with risks and opportunity costs. This framework encourages evaluating probabilities, payoffs, and opportunity costs to make better decisions.
Decisions based on global impressions from experience tend to underweight rare events because they are not salient unless they attract separate attention, in contrast to choices from explicit description which often lead to overweighting of rare events.
A technique that stops group discussions from getting stuck in a narrow frame by challenging status-quo bias and setting tripwires. It involves seeking out solutions from others and creating triggers for action.
A knowledge panel can provide leads about an unfamiliar source’s identity and background. Check those details and the article’s evidence before deciding whether to trust a particular claim.
Deductible aversion is a preference for lower insurance deductibles that can lead people to overpay for coverage. In some plans, the extra premium exceeds the largest possible reduction in out-of-pocket costs, making the lower-deductible option more expensive regardless of claims.
Deductible aversion is a preference for low insurance deductibles even when a household could bear a larger loss and save on premiums. Thaler and Sunstein argue for concentrating coverage on losses that would cause financial hardship.