Human cognition is shaped by systematic tendencies called cognitive biases, which are predictable deviations from rational judgment. Confirmation bias leads us to favor information that confirms our existing beliefs and discount evidence that contradicts them. Availability bias makes us judge the probability of events by how easily examples come to mind, often overestimating vivid or recent occurrences. Anchoring bias causes us to rely too heavily on the first piece of information presented, even when it is irrelevant. Hindsight bias, the "I knew it all along" effect, distorts memory of past predictions after the outcome is known. These biases shape not just what we believe but how confident we feel about those beliefs.
Biases also distort how we assess ourselves and others. The Dunning-Kruger effect describes how those with limited skill tend to overestimate their ability, while experts often underestimate theirs. Overconfidence bias reflects excessive trust in one's own judgment. In-group bias favors members of our own group, and the fundamental attribution error overemphasizes personal traits while underestimating situational factors in explaining others' behavior. Self-serving bias attributes our successes to ourselves and our failures to external circumstances. Naive realism is the belief that we see the world objectively and that those who disagree are simply biased. Motivated reasoning is the broader pattern of reasoning toward a preferred conclusion rather than the most accurate one. Cognitive dissonance—the discomfort of holding contradictory beliefs—often triggers rationalization, in which we produce plausible but false reasons to ease the tension, and in extreme cases, denial, in which we refuse to accept reality despite evidence.
Decision-making is also shaped by how information is framed and how options feel. Loss aversion makes us prefer avoiding losses over acquiring equivalent gains, while the framing effect shows that the presentation of information influences our choices independently of the underlying facts. Status quo bias produces a preference for the current state of affairs, and the halo and horn effects cause one positive or negative trait to color our overall judgment of a person. The planning fallacy leads us to underestimate the time and resources tasks require, while recency bias weighs recent events more heavily than distant ones. Survivorship bias directs attention to those who "survived" a process while ignoring those who did not. The bandwagon effect, the IKEA effect (overvaluing what we helped create), and the endowment effect (valuing what we own) all reveal how context and ownership distort valuation. The just-world hypothesis leads to victim-blaming under the assumption that the world is fair, and moral licensing permits unethical behavior after a virtuous act. The bystander effect shows that helping behavior is reduced when others are present.
Beyond these motivational and social biases, several biases distort our statistical and probabilistic reasoning. The conjunction fallacy leads us to judge the conjunction of two events as more probable than one of them alone. Base rate neglect causes us to ignore underlying statistical frequencies in favor of specific information. Regression to the mean describes the tendency for extreme observations to be closer to average on subsequent measurements, often misinterpreted as a real effect; the law of large numbers, by contrast, states that as sample size grows, the sample mean approaches the population mean, while the cognitive law of small numbers reflects the mistaken belief that small samples reliably reflect population properties. The illusion of control makes us overestimate our influence over events, and optimism bias leads us to underestimate the likelihood of negative events affecting us personally. Many of these errors arise from heuristics—mental shortcuts that are often useful but systematically mislead in particular conditions—and represent a bias-variance tradeoff in belief formation, balancing simplification against capturing variability.