Bias
Would You Wear Hitler's Sweater?
Errors of association lead to irrational behaviors.
Updated May 22, 2025 Reviewed by Kaja Perina
Key points
- We tend to read too much into chance correlations (association bias).
- We tend to see patterns even where there are none (beginner’s luck).
- We tend to think that the properties of one object can be transferred to another (contagion bias).
- Errors of association could be called remnants of an anterior stage in the evolution of the human mind.
This is the third post in a series on cognitive biases.
A cognitive bias is a mental shortcut or heuristic intended to spare us time, effort, or discomfort—often while reinforcing our self-image or worldview—but at the cost of accuracy and reliability.
For example, in explaining the behaviour of others, we tend to overestimate the role of character traits over situational factors—a bias called fundamental attribution error (FAE), which goes into reverse when it comes to explaining our own behaviour. Thus, if George fails to mow the lawn, I indict him with forgetfulness, laziness, or spite; but if I fail to mow the lawn, I am likely to absolve myself on the grounds of busyness, tiredness, or inclement weather.
Well over a hundred cognitive biases have been identified. After studying them, I found that most fitted into one of just eight groups, clusters, or themes:
- Errors of salience
- Errors of attention
- Errors of association
- Errors of framing
- Errors of attachment
- Errors of conformity
- Errors of control
- Errors of infallibility
In this post, the third in a series on cognitive biases, I’m going to look at the third group: what I have called errors of association. (If you missed my previous post on errors of attention, you can find it here.)
Remember that there can be a lot of overlap between named biases. Instead of trying to distinguish them, focus instead on the principles at play.
Errors of Association: Association Bias and Swimmer's Body Illusion
Correlation is not causation. I had a friend who loved my crab linguine. But one day, he became ill some time after eating it, and concluded that he was allergic to crab—what was likely an association bias. (I’m old enough to remember a time when it seemed that no one had any food intolerances, and I harbour a suspicion that many allergies have a similar origin story founded in association bias.)
In their prospective material, business schools like to highlight the income lift of doing an MBA with them. But this income lift is not likely to be solely the result of doing an MBA; rather, the kind of people who do an MBA are perhaps more likely to be highly motivated and career-driven, traits that already made them more likely to earn more.
Association bias also operates in looser ways. When receiving bad news, people are apt to transfer their negative feelings to the person delivering it, that is, to “shoot the messenger.” This can lead to leaders who are out of touch with reality, especially authoritarian leaders to whom no one dares speak the truth. (It’s actually quite hard to define a tyrant, but “one to whom no one dares give advice” might be as good a definition as any.)
Conversely, we often transfer our feelings about a person to their message, which is why ads often feature a celebrity darling. Ads for clothes and cosmetics invariably feature fetching people to suggest that we too will become attractive if only we buy these products (swimmer’s body illusion). Swimmer's body illusion refers to our tendency to confuse results with selection factors: we look at a swimming champion and think that we might get their body by taking up swimming, whereas in fact the swimming champion became a swimming champion because they already had a body that was fit for swimming.
Errors of Association: Contagion Bias, Beginner's Luck, and Hindsight Bias
Association bias can be even more primitive than that, harking back to a time when magical thinking dominated the human mind. For example, most people would balk at the idea of wearing Hitler’s sweater, to the extent of condemning and ostracising anyone who would be comfortable with the idea (contagion bias).
Years ago, on a wine research trip, I was invited to sleep in what transpired to be the room in which the previous winemaker had shot himself in the head. For all my good sense, it took me a long time to fall asleep that night.
We tend to see patterns even where there are none, failing to realize that many “patterns” owe to chance. Initial success in a game or sport can lead us to pursue that activity in the mistaken belief that we have a knack for it (beginner’s luck).
During the dot-com boom, many people ploughed their savings into stocks. On the back of their successes, some decided to ditch the day job and go into speculative trading, only to lose it all when the bubble burst. How could they not have seen it coming? Looking back upon it now, all the signs were there (hindsight bias).
In Conclusion
Errors of association such as association bias, contagion bis, and beginner's luck are rooted in pre-logical magical thinking. Historically, people believed that they could influence the world through ritual, which is imbued with so-called sympathetic, or imitative, or magic.
The anthropologist James George Frazer (d. 1941) argued that sympathetic magic operates by one of two laws, the Law of Similarity (that like produces like, or that an effect resembles its cause) and the Law of Contagion (that things which have once been in contact continue to act on each other at a distance). From the first principle, our ancestors inferred that they could produce any effect they desired merely by imitating it; from the second, that whatever they did to an object would affect equally the person with whom the object was once in contact.
In sum, errors of association might be thought of as a remnant of an anterior stage in the evolution of the human mind.
In my next post, I will be looking at errors of framing.
Neel Burton is author of the newly published How to Think Like Plato and Speak Like Cicero.