THE REFERENCE LIBRARY
A growing collection of ideas for understanding the world — and your own thinking.
A confident story about a market movement can outrun the evidence. Taleb’s Bloomberg example separates two questions: whether a change is unusual and whether the proposed cause actually explains it.
Erikson describes Blues as expecting preparation and precise answers. His advice is to bring relevant documentation and admit when an answer needs checking.
Understanding the Way allows one to become adaptable, taking any form (light as a feather, fluid as water, stiff as a board) as needed, leading to invincibility in strategy.
The Boiling Frog is a metaphor for how holding grudges slowly consumes energy and can be passed down through generations, often long after the original cause is forgotten.
The Bowtie diagram is a visual tool for analyzing risks by considering threats, preventative measures, detective measures, and recovery measures. It helps think through what can go wrong and how to prevent or mitigate incidents, emphasizing the importance of a balanced system of protections.
The brain can be understood as a computer that processes information, an idea pioneered by Turing and others. This thinking tool helps explain how meaning and intelligence arise from mechanical parts, but warns against oversimplifying the brain's operations.
This study compared students’ forecasts of happiness in a future relationship with happiness reported by students already in relationships. The excerpt reports similar forecasts and group averages for falling in love; it describes the breakup groups without providing their results.
This concept, illustrated through a French trader's sayings and statistical evidence, argues that luck and randomness are the great equalizers in capitalism, causing even the largest corporations and most celebrated individuals to eventually fall and be replaced by newcomers.
Kahneman describes how demands on System 2 may weaken self-control. He reviews early ego-depletion experiments and a proposed glucose mechanism; these study reports do not establish that sugar reliably restores willpower.
Chaotic dynamics amplify small differences in initial conditions, limiting precise forecasts over long horizons even when the rules are deterministic. Useful short-term and statistical forecasts may still be possible.
The buyer’s model offers ways to stop a purchase decision from becoming an endless search: limit research, prioritize requirements, use the 10-10-10 exercise, and consider help in breaking a genuine tie.
The studies described here found overextreme forecasts among Dutch and American lawyers: low probabilities understated success and high probabilities overstated it. Discrimination was poorer in the American sample, especially for forecasts made far from trial.
A concept for distinguishing formal management systems (hierarchy) from informal, deeper interactions (camp) that build trust and reveal underlying values, based on the work of Bruce Gibbs and used by Roger Saillant.
The Cemetery of Letters refers to the bias of silent evidence, where we only see the successes (e.g., published books) and ignore the vast number of failures (e.g., unpublished manuscripts), leading to an overestimation of the uniqueness of successful individuals.
Basic logic uses quantifiers like 'for all' and 'there exists' to make precise statements about scope and truth, overcoming the ambiguity of everyday language.
Communication can fail through interacting causes. High stakes may increase a speaker’s motivation to be clear while also creating pressure that leaves less attention for monitoring ambiguous wording.
Erikson describes why people find giving and receiving feedback uncomfortable. He warns that avoiding feedback altogether leaves important issues unaddressed.
The passage describes technological failures that demand a rapid response. Its Guardian outage anecdote also shows the value of prevention: a more realistic test dataset could have exposed the error before release.
For some long-tailed duration models, an unfinished project’s expected remaining time increases as more time passes. Taleb contrasts this with a life-expectancy example in which remaining time falls with age. The effect depends on the distribution, rather than on delay alone.
Historical and institutional accounts favor formal academic medicine over empirical, trial-and-error practitioners (empirics). This bias rests on the fallacy that nonacademic practice must be nonrigorous or quackery.
Diffusion models describe how innovations spread from innovators and early adopters to a broader public. The chasm metaphor focuses on the difficulty of reaching people who want stronger evidence or different benefits before adopting.
A chemical bond is an attraction between atoms, ions, or molecules that enables the formation of chemical compounds, resulting from electrostatic forces in ionic bonds or electron sharing in covalent bonds.
A chemical reaction is a thermodynamic process where reactants transform into products through the breaking and forming of chemical bonds, often involving energy changes. It produces new substances with different molecular structures, unlike physical changes that only alter state or form.
John Searle's Chinese Room thought experiment argues that a computer program manipulating symbols cannot achieve true understanding or consciousness, challenging Strong AI. Daniel Dennett critiques it as a defective 'intuition pump' that misleads by obscuring crucial details.
The paradox of choice suggests that having too many options can lead to decision paralysis and dissatisfaction, as demonstrated by Sheena Iyengar's jam experiment where a smaller selection led to higher purchase rates than a larger one.
An acceptable option may still have a better alternative. Before settling, consider whether the choices offered leave out a way to obtain the same benefit with fewer drawbacks.
The same observation can fit explanations you have not considered. Ellenberg’s friend appeared to wear one dirty T-shirt every day but was using a fresh shirt from a box of identical stock. His discussion of design arguments extends the lesson: an inference depends on which hypotheses you include and what probabilities you assign them.
Taleb criticizes treating financial risk as Gaussian when extreme market moves are much more frequent than that model suggests. He argues that a familiar numerical measure can become an anchor despite evidence against its assumptions.
Cognitive dissonance is the mental discomfort experienced when actions contradict beliefs, often leading to self-justification rather than acknowledgment of error.
Endorsement by a respected or well-known person can accelerate acceptance of an idea or behavior. The catalyst analogy describes influence on adoption, without making endorsement necessary or sufficient for lasting change.
The Company of Man uses business equity as a metaphor for how choices affect dating prospects, then applies SWOT analysis to personal goals. The metaphor can organize questions about effort and context, but its gender rankings and value curves are assumptions rather than measures of a person’s worth.
The Complacent is a zone in the Cynefin framework located between the Clear and Chaotic domains, representing a state of overconfidence that can lead to catastrophic failures, as illustrated by Knight Capital's $440m loss in 2012.
Taleb distinguishes complex systems, whose interdependencies can produce cascading effects, from complicated systems with simpler responses. The distinction encourages attention to interactions that make isolated cause-and-effect explanations unreliable.
A sealed habitat must keep oxygen, food, water and many interacting species within conditions that support life. Biosphere 2 exposed dependencies that its planners had not fully anticipated. It illustrates why designing a viable ecosystem requires monitoring, feedback and adjustments as well as initial planning.
The source distinguishes software tasks with established methods from tasks that need experimentation. It uses coordinate calculations as a stable task and mentions tools for children learning programming.
In Cynefin’s Complicated domain, cause-and-effect relationships can be analyzed with appropriate expertise. The supplied account describes a leadership consultancy that values experts but often neglects the data and analysis some of its own decisions need.
This model distinguishes six responses to conflict: flight, fight, giving up, delegating the decision, compromise, and jointly developing a solution. It invites the parties to consider which response they are using and what alternatives remain.
A thought experiment to determine whether your beliefs, opinions, or preferences are genuinely your own or shaped by conformity. It involves imagining that the person you agree with no longer holds that view, or that the majority opinion is different, and checking if your stance remains.
The conjunction effect occurs when a combination of events is judged more probable than a broader event that contains it. A plausible added cause can make a narrower scenario feel more likely. Support theory relates this to how descriptions make supporting possibilities accessible.
The consequences model highlights a project tradeoff: early decisions can shape much of the work while information is still limited. It encourages timely decisions and explicit communication when a decision is deferred.
Self-serving manipulation can contribute to unrealistic projects, concealed risks, weakened oversight, and blame shifting. Paterson recommends independent challenge and scrutiny of information reaching senior leaders, alongside attention to the conditions that allow the behavior.
Survivorship bias and missing evidence can make a plausible causal story look more convincing than it is. Taleb urges caution with 'because': ask what evidence is absent and whether the explanation distinguishes causation from luck.
Humans are naturally drawn to the tangible, visible, and concrete while ignoring the abstract and unseen, leading to cognitive biases like confirmation bias and neglect of silent evidence. This results in a superficial understanding of uncertainty and randomness.
People operate in two distinct worlds: social norms (friendly, no immediate payback) and market norms (wages, prices). Mixing these norms can lead to negative outcomes, such as reduced motivation, damaged relationships, and loss of intrinsic goodwill.
The zero price effect is the extra appeal an offer can gain when its price falls to nothing. Ariely describes experiments in which people switched toward a free item even though the price gap between the available options stayed the same. A free offer can still cost time, require another purchase, or displace something you would value more.
Taleb argues that probability often resists intuition and that simplifying an argument for an audience can strip out essential uncertainty. His account contrasts persuasive presentation with the accuracy of the underlying reasoning.
The concept challenges the dismissive view of critics by arguing that some critics, like Abraham Wald, provide invaluable guidance that improves the actions of those in the arena, thus making a meaningful contribution despite not being directly in the fray.
A decision framework for finding direction in life by reflecting on your past, values, important people, hindrances, and fears, then imagining six different possible paths forward.
Munger argues that layers of investment fees, consultants, and trading costs can consume a foundation’s returns. He discusses low-cost indexing and concentrated long-term ownership as different alternatives, and uses a historical three-percent cost estimate to show how fees and donations can shrink assets.
Senge describes a pattern in which a quick fix relieves symptoms while weakening the capacity to handle the underlying problem. Dependence on the intervention then grows. He calls this shifting the burden to the intervenor.
Recurring reports can consume analytical capacity without helping anyone decide what to do. The problem arises when organizations keep adding summaries of outcomes while leaving the causes, discrepancies and decisions unexplored.
The Cycle of Distrust describes how repeated deceptive marketing and corporate behavior erode public trust, creating a self-reinforcing cycle of suspicion that damages even honest actors. It draws parallels to the tragedy of the commons, showing trust as a shared resource that is easily depleted by a few bad actors.
Practise critical thinking through repeated checks of sources, evidence, and assumptions. Familiarity with these checks can make them easier to use in everyday decisions.
The legend of Gun’s failed dams and Yu’s irrigation canals illustrates an approach to complexity: work with a system’s existing flows and context. The authors use the story to contrast rigid control with constraints that allow adaptation.
Leverage, when used unrestrainedly and systematically, can become a form of tyranny. The historical example of West Virginia coal company towns illustrates how economic leverage can be parlayed into total control over workers' lives and the state.
Our brains have inherent biases that distort perception and thinking, such as confirmation bias and inattentional blindness. Awareness of these biases can help us avoid irrational beliefs and make better decisions.
Parrish distinguishes a considered decision from a reactive choice. His process is to define the problem, explore solutions, evaluate options, and act, with more time devoted to choices whose consequences are substantial.
The default fund is a choice architecture tool where a pre-selected option is the automatic outcome if no active choice is made. The Swedish premium pension plan used a default fund (AP7) but nudged participants to actively choose their own portfolios, resulting in two-thirds making active choices.
We learn best from direct experience, but many critical decisions have consequences that extend beyond our 'learning horizon' in time and space, making it impossible to learn from experience directly.
When an event is rare relative to the available record, observations provide limited guidance about its frequency. Estimating its probability then depends more heavily on assumptions about outcomes beyond the observed data.