Base rate neglect fallacy
“I misjudge the probability of an event by ignoring important background information.”
In probability theory, the base rate, or initial probability, relates to the probability that an event will occur, that an outcome will come about, or that individuals will have a certain characteristic. For example, about 90% of humans are right-handed. The base rate of right-handed people is thus around 90%, and if you pick a person at random, you have approximately a 90% chance that that person will be right-handed. The base rate neglect fallacy, which can be seen as a cognitive bias, is in fact a group of phenomena whereby base rates are not sufficiently taken into account in reasoning processes. In its most typical manifestations, it takes the form of a conflict between a base rate probability and information about a specific case, such as an individual or a part of the population. Often the fallacy will manifest itself when the specific case involves a stereotype.
The following example illustrates a case where the base rate neglect fallacy is implicit:
Ireland is one of the places in the world with the most red-haired people per capita. A person is selected at random from the Irish population. What are the odds of selecting a person with red hair? If you rate the odds as high (say, over 30%), you are ignoring the fact that red-haired people make up only about 10% of the population of Ireland. You neglected, or forgot, to take this base rate into account, probably due to the stereotype that Irish people are often red-haired.
The dominant explanation for this bias involves the dual process model, according to which our judgements and decisions are based on two types of processes, some fast and intuitive (called “type 1 processes”), others slow and deliberate (referred to as “type 2 processes”). One version of this explanation is that our judgements are often based on a representativeness heuristic (a type 1 process): we judge a given situation not according to all the information at our disposal, but according to what we believe to be representative of said situation, such as a stereotype . In the example of Ireland, there is a conflict between the base rate that 10% of Irish people are red-haired, and a stereotype that many Irish people are red-haired.
This bias can manifest itself in the form of rather harmless prejudices but can also have more harmful consequences. For example, when Donald Trump says that "so are white people [getting killed by the police]" he is telling the truth, but he fails to mention that African Americans are disproportionately killed: African Americans make up less than 15 % of the U.S. population but account for approximately 25% of police deaths. Trump's claim, which seems to suggest that whites are being killed by the police as much as African Americans, can thus be understood as an oversight, or at the very least a neglect of the base rate of African Americans in the population.
More generally, base rates can play an important role in various decision-making contexts, including in the medical world, in courts, or in the world of science and technology. To give just one example from the medical world, when assessing the reliability of a coronavirus test, it can be important to take into account the rate of infection within a given population. This would allow verification that the test is reliable, that it detects all cases. Neglecting the base rate could therefore have serious consequences.
Thoughts on how to act in light of this bias
Challenge stereotypes by learning about relevant base rates.
Identify situations or contexts conducive to misleading us, where there is a conflict between stereotypes and the base rate information.
How is this bias measured?
The base rate neglect fallacy has been observed on numerous occasions and in various forms in experimental settings . A typical task is to give participants certain information (for example, about the composition of a population and the effectiveness of a test) and then ask them to make judgements about probabilities (for example, the probability that a person has a virus, considering the information provided). A discrepancy between a produced judgement and a judgement in accordance with the laws of probability suggests that the base rate has been neglected.
This bias is discussed in the scientific literature:
This bias has social or individual repercussions:
This bias is empirically demonstrated:
 Kahneman, Daniel & Amos Tversky (1973). On the psychology of prediction. Psychological review, 80(4):237–251.
 Barbey, Aron & Steven Sloman (2007). Base-rate respect: From ecological rationality to dual processes. Behavioral and Brain Sciences, 30(3):241–254.
Cosmides, Leda and John Tooby (1996). Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under uncertainty. Cognition, 58(1):1–73.
Gigerenzer, Gerd, Wolfgang Hell & Hartmut Blank (1988). Presentation and content: The use of base rates as a continuous variable. Journal of Experimental Psychology: Human Perception and Performance, 14(3):513–525.
Individual level, Representativeness heuristic, Need for cognitive closure
Julien Ouellette-Michaud, PhD candidate in philosophy, McGill University; philosophy teacher, Collège de Maisonneuve.
Translated from French to English by Susan D. Renaud.