The Gap Between Calculus and Cognition
Probability theory provides a rigorous mathematical framework for quantifying uncertainty. Human psychology, however, often operates on a set of intuitive heuristics—mental shortcuts—that routinely violate this framework. The field of behavioral probability, a core discipline at the Las Vegas Institute of Probability Theory, investigates this gap. We conduct observational and experimental research to catalog, understand, and model the systematic errors people make when judging likelihoods, assessing risks, and making decisions under uncertainty. Our location provides a fertile, high-stakes environment to observe these biases not just in controlled labs, but in situ, where real money and emotions are on the line.
Catalog of Cognitive Illusions
Several key biases form the cornerstone of our research. The gambler's fallacy is the mistaken belief that past independent events influence future ones. A roulette player seeing five reds in a row may wrongly believe black is 'due,' assigning it a probability greater than the true 18/38 (on a double-zero wheel). We study the strength of this fallacy across different games, cultures, and levels of education, measuring how it affects betting patterns.
Conversely, the hot-hand fallacy is the belief that a person experiencing success with a random process has a higher chance of continued success. In basketball, fans believe a player who has made several shots is 'hot.' In gambling, a craps player on a winning streak may be perceived (by themselves and others) as having a 'lucky roll.' Our research involves tracking perceived versus actual conditional probabilities to dissect the neural and social mechanisms behind this illusion of predictability in random sequences.
Overconfidence in personal probability assessments is another major area. People consistently overestimate the precision of their knowledge, assigning overly narrow confidence intervals to estimates. A poker player might be 90% sure they have the best hand when the actual Bayesian probability, given the cards and betting action, might be 70%. We study how this bias impacts bluffing, bankroll management, and risk of ruin.
Misunderstanding of compound events is also common. People tend to overestimate the probability of conjunctive events (e.g., hitting a specific multi-stage slot bonus round) and underestimate the probability of disjunctive events (e.g., experiencing at least one loss in a series of bets). This has profound implications for how games are marketed and how players perceive their chances of winning big versus losing steadily.
The Role of Emotion and Narrative
Our research extends beyond cold cognition into the realm of affect. The affect heuristic dictates that people judge the probability of an event by how they feel about it. A desired outcome (winning a jackpot) may feel more likely, while a feared outcome (losing a mortgage payment) may feel less likely, irrespective of the objective odds. We use biometrics and self-reporting in controlled gaming environments to correlate emotional arousal with shifting probability judgments.
Humans are also storytellers. We impose narratives on random sequences to make them meaningful. A string of losses becomes a 'lesson' or 'bad luck,' while a win is attributed to skill or a lucky charm. This 'narrative fallacy' helps individuals cope with randomness but distorts their mental model of the underlying stochastic process. LVIPT ethnographers and psychologists collaborate to document these narratives, understanding them as a fundamental part of how people interact with chance.
Applications: Designing for Better Decisions
The goal of this research is not merely to document human error but to mitigate its harmful effects. One application is in the design of decision-support tools. For instance, we collaborate on software that provides real-time, objective probability assessments to poker players, helping to counter overconfidence and the gambler's fallacy. We also work with regulators and operators on responsible gambling initiatives. This includes designing clear, non-misleading information about game odds and the nature of randomness, and creating 'reality check' tools that interrupt play with accurate summaries of time spent and probabilistic expectations.
Furthermore, our findings have broad societal relevance. Understanding how people misjudge the risks of medical procedures, climate events, or financial investments is crucial for effective communication in public health, policy, and education. By studying these biases where they are most starkly revealed, the Las Vegas Institute of Probability Theory contributes to a general science of better decision-making, helping individuals and institutions align their subjective probabilities more closely with objective reality, and navigate a risky world with clearer eyes.