In Japan, many people believe that blood type shows your personality. For example, type A people are said to be careful.

But in 2014, a Japanese psychologist studied survey data from more than 10,000 people in Japan and the United States. He found almost no difference in personality between blood types. So why do the readings feel true? Their words are often vague and fit almost anyone.

Blood-type personality readings are very popular in Japan, but psychologists have long doubted them. In 2014, social psychologist Kengo Nawata analyzed large social surveys from Japan and the United States, covering more than 10,000 people in total. When he compared answers to many questions about life and attitudes by blood type, he found almost no meaningful differences. A 2015 study using a standard personality test found only a very weak link.

So why do the readings seem accurate? One reason is the Barnum effect: we accept vague descriptions that could fit anyone as true for ourselves. Another is confirmation bias, which makes us remember examples that match our beliefs and forget those that do not.

In 1949, the psychologist Bertram Forer handed each of his students what looked like a personal description, and most rated it as highly accurate. Every student had received exactly the same text. This trick of the mind, now called the Barnum effect, helps explain why blood-type personality readings feel so convincing in Japan. Add confirmation bias, the habit of noticing examples that fit our beliefs, and a vague stereotype can start to look like a law of nature.

Psychologists have long doubted the readings, and in 2014 Kengo Nawata put them to a large test. He analyzed social surveys from Japan and the United States with more than 10,000 respondents, comparing answers to dozens of questions about lifestyle and attitudes across blood types. In almost every case the differences were negligible. Where small gaps did appear, they were too tiny to predict how any individual would answer. A 2015 study using a standardized personality inventory found only a weak, uncertain association.

Researchers also note that people who know the stereotypes may describe themselves slightly differently because of them. The belief is common in Japan and some neighboring regions but rare in Western countries, which points to culture as its source.