One explanation for why we make a conjunction error in cases like Linda, the bank teller, is that we misuse what Tversky and Kahneman call a representative heuristic. The most cited example of this error came from Amos Tversky and Daniel Kahneman. Conjunction error (also known as Linda’s problem) is a formal error that occurs when it is assumed that certain conditions are more likely than one common one. The original report by Tversky and Kahneman [2] (later republished as a chapter in book [3]) describes four problems causing the conjunction error, including Linda’s problem. [Sources: 0, 5]
Amos Tversky and Daniel Kahneman are known for their work on a large number of cognitive errors that we all tend to make over and over again. Although Linda’s problem is the most famous example, researchers have developed dozens of problems that reliably detect conjunction error. This is known as conjunction error or Linda’s problem and is a source of behavioral bias in decision making. While representativeness bias occurs when we ignore low base rates, conjunction error occurs when we attribute a higher probability to an event with a higher specificity. [Sources: 0, 2, 3, 5]
Tversky and Kahneman (1983) asked participants to solve the following problems. Tversky and Kahneman have explored many variants of the Linda problem formula. The two found many logical errors, and we often make these errors when we are faced with information that seems a bit familiar. [Sources: 0, 2, 5]
As discussed in our article on storytelling errors, Kahneman and Tversky’s most famous and most controversial experiments involved a fictional woman named Linda. Assessing the connection of two events as more likely than just one of the events is an example of a connection error; the human tendency to do this is commonly known as cumulative error. Likewise, conjunction error also occurs when people are asked to bet with real money [16], as well as when solving intuitive physics problems in various projects. In his book Thinking Fast and Slow, which summarizes his life and Tversky’s work, Kahneman introduces biases that stem from misalignment — the false belief that a combination of two events is more likely than one event alone. [Sources: 0, 2]
The error of representativity and conjunction arises from the fact that we mentally shorten the path from the supposed plausibility of the scenario to its probability. But perhaps Tversky and Kahneman are wrong about the Linda case. Conjunction bias is a common reasoning error in which we believe that two events occurring together are more likely than one of those events to happen on their own. [Sources: 2, 5]
The so-called cognitive defect, which the study participants operated on in the Linda case, is also associated with the formulation of the problem. However, there are studies that have observed indistinguishable rates of conjunctive errors with formalized stimuli in terms of likelihood and frequency. The representativeness heuristic was invented by Daniel Kahneman and Amos Tversky, two of the most influential figures in behavioral economics. [Sources: 0, 1, 5]
With this way of stating the problem, if you attach great importance to the background information about Linda’s student days, we would expect study participants to attribute equal probability to 3 (a) and 3 (b). However, the probability that two events will occur together is always less than or equal to the probability that one will happen alone. [Sources: 5]
Daniel Kahneman and Amos Tversky have spent decades in psychological research to unravel the patterns in the mistakes of human thinking. So, if business is done right, the answer is that Linda is more likely to be a bank teller (rather than a bank teller and a feminist). [Sources: 2, 5]
The problem with the representativeness heuristic is that representativeness has nothing to do with probability, but we value it more than relevant information. Based on this description, subjects are asked which of the following statements is more likely. Prototypes guide our assumptions about probability, as in the example above about Steve and his profession. [Sources: 1, 4]
This term refers to the tendency to think that a combination of two events is more likely than either of these events to occur separately. When we try to make decisions about unknown things or people, we refer to this environment – the prototype – as a representative example of the entire category. According to a categorization theory known as prototype theory, people use unconscious mental statistics to understand what the “average” member of a category looks like. However, there is another main reason why the representativeness heuristic arises. [Sources: 1, 3]
However, the first option is a shorter letter than the other two and is more likely. This means that we often rely on labels to quickly judge the world. [Sources: 1, 3]
— Slimane Zouggari
##### Sources #####
[0]: https://en.wikipedia.org/wiki/Conjunction_fallacy
[1]: https://thedecisionlab.com/biases/representativeness-heuristic/
[2]: https://fs.blog/bias-conjunction-fallacy/
[3]: https://econowmics.com/the-conjunction-fallacy/
[4]: https://en.shortcogs.com/bias/conjunction-fallacy
[5]: https://www.psychologytoday.com/us/blog/the-superhuman-mind/201611/linda-the-bank-teller-case-revisited