How do we come to be sure of our beliefs

Published: 10-11-2021

Imagine having to guess what is governing the sequence of numbers: “2,4,6”. To guess the rule, you can tell other strings of numbers and ask if the strings match the rule. The task is apparently very simple … “10,12,14?” – yes, it complies with this rule. “106,108,110?” – yes, this sequence also complies with this rule. So there is no need to wonder – the correct answer is “even numbers increasing by two”. And here’s the surprise… the answer is wrong.

Such a study was conducted by Wason in 1960 (as cited in: Lewicka, 2001). Most of the participants in his study did not guess the rule because right away, before asking the first question, they thought that the sequence was governed by the rule of “even numbers increasing by two” and then asked questions that could only confirm their belief. They used confirmatory strategy for testing hypothesis. So they didn’t actually verify it, they only confirmed it. It would be better if they asked for a sequence of numbers that, according to their original hypothesis, does not conform to the rule, such as “1,3,5”. Then they would also get the answer that the sequence fits the rule and after the next questions they could come up with the correct answer, that is “every sequence of ascending numbers”. The rule was much broader than they assumed, and such sequences as “20,21,100”, “32.33.34” or “½, 8, 9 ¾” also fit it … but they had no way of reaching such a conclusion by asking the questions they were asking.

In everyday life, this is also how we come to be sure of our beliefs very often. Instead of verifying our previous assumptions, we only look for data that can confirm them and we do not notice other data or we think that they are not worth your interest for some reason. This can be seen, for example, in people who believe in horoscopes. They do not notice situations in their environment that could contradict what they read in the horoscope. Like the participants in the Wason study, they look only for confirmation. It is very easy to find confirmation of such general statements as: “private matters will become important today” or “the day brings with it many favorable changes and situations”. A lot of situations can be applied to the slogan “favorable changes and situations”: the bus has changed its route and you will now commute to school or work faster; the boy you like invited you for coffee; a local shop lowered the price of your favorite cookies … But since there were also unfavorable changes on the same day, for example your favorite yogurt went up, you might not notice it anymore … Moreover, people who read horoscopes do not take into account that the statements about personality traits they contain are very general and may apply to many people, not only to themselves. For example, “you are characterized by extremely strong will and great stubbornness in achieving your goals” or “you do not care about fame or promotions, but you like to be appreciated by your boss, even in words”. Who doesn’t like to be appreciated by the boss? And a strong will? It’s nothing unusual, many people have it. In psychology, this phenomenon of assigning very general statements to oneself is called the “Barnum Effect“, from the circus performer who claimed that he owes his successes to the slogan “something that suits everyone.”

The Barnum effect is not only visible with horoscopes. There are studies that show that the Adult Children of Alcoholics syndrome is actually just the Barnum Effect. People who grew up in a family where at least one of their parents was an alcoholic, according to the information on the ACoA syndrome (Lilienfeld et al. 2011), are often characterized by low self-esteem, a sense of shame or guilt, and a tendency to take too much responsibility for others, and in turn, sometimes they do not feel responsible, they have too high a need for approval, an excessively developed sense of loyalty, a feeling of helplessness and difficulty controlling reflexes. Logue, Sher, and Frensch (1992) conducted a study in which people who had at least one alcoholic parent and those who did not have one were to respond to different statements. Some of these statements were among those allegedly referring to the ACoA (e.g. “In times of crisis you tend to take care of others”, “You are sensitive to the difficulties of others”) and some were generalities not related to ACoA (e.g. “You have a superb sense of humor”,”You are not disturbed by variety and change and see them as challenges.”). Of course, the subjects were not told which features describe ACoA and which do not. It turned out that in fact many people who grew up in a family with alcoholism identified themselves with the statements describing the Adult Children of Alcoholics syndrome. However, they also identified themselves with statements that do not describe this syndrome. Likewise, people who did not grow up in a home with alcoholism identified themselves with both the statements describing the ACoA and those that did not describe the syndrome. There were no differences between the two groups of people. The prevalence of the ACoA syndrome is probably due to the fact that the features that supposedly describe a typical adult child of an alcoholic are generalities that can be easily attributed to each other. A person who has some problems in his/her life and looks for the cause of these problems, and at the same time had an alcoholic parentage, when reads such characteristics, often finds that most of them fit him/her. He/she is then relieved because finally has “found out” why he/she is not doing well – all because he/she is an Adult Child of the Alcoholic. In reality, however, his/her problems may arise for a completely different reason, not related to childhood at all.

The use of a confirmatory strategy of testing hypotheses was also noticed in boxing judges. An analysis of 126 boxing fights from the Polish Championships in 1986 and a study in which the judges were shown a video of the fight and the order of the rounds shown was mixed, showed that the judges in their ratings of the entire fight to a large extent relied on who won the first round. On the basis of who performed better in the first round, they hypothesized that this player was better, and in the following rounds they assessed both players biased. The fighter who won the first round had a better chance of winning the entire fight, and likewise, which of them won the second and third rounds depended on which of them won the previous round (Tyszka, Wielochowski, 1991).

Once we come to a belief, it is difficult to change it. We reject information that does not match our views because we dislike changing our opinions. In one study (Kofta, Doliński, 1997 as cited in: Tyszka, 1999) a story was presented about a doctor who misdiagnosed a patient, as a result of which the patient died. Then the subjects were informed about the circumstances of the story. They found out, among other things, that the doctor on duty was very tired because he had been working all day the previous day, and had to work because the hospital did not have enough doctors. One group of respondents read about the circumstances justifying the doctor immediately after reading about his mistake, while the other group after a five-minute break. People in the group who read about the circumstances that caused the fatal error only after the break, assigned the doctor greater responsibility for his act, gave a higher mark of the moral blameworthiness of his mistake, and postulated a higher penalty than those in the first group. They assessed the doctor very negatively already before the break and the mention of extenuating circumstances only later did not change their opinion.

Information that is contrary to our views may, in certain situations, even confirm our beliefs. Such a phenomenon was observed in a study of supporters and opponents of the death penalty (Lord, Ross, Lepper, 1979). All participants in this study were given two texts to read. One of them argued that the death penalty was effective in deterring potential criminals, while the other argued against it. Both texts were based on reliable sources and written in a scientific style. If people were rational in forming their opinions, they would conclude from both texts that the issue of the death penalty is very complex and that it is difficult to say unequivocally whether it deters potential murderers or not. However, it happened otherwise. Both supporters and opponents of the death penalty paid attention only to the text that was consistent with their beliefs. Supporters believed that only the text that argued that it discourages committing serious crimes is credible, and in the second they found some minor errors, methodological flaws and believed that it was not credible, and so were the opponents of the death penalty – only the text saying that it was not a disincentive to committing serious crimes was worth taking into account for them. As a result, supporters of the death penalty became convinced that it was right, and opponents that it was wrong.

In 2004, before the presidential elections in the United States, a neurobiological study was conducted in supporters of George W. Bush and John F. Kerry (Westen et al., 2006). The Republicans and Democrats interviewed listened to different information about both candidates, and functional magnetic resonance imaging (fMRI) measured the activity of different areas of their brains. They found that when they listened to information that was inconsistent with their beliefs, their brain areas responsible for cool rational reasoning were almost inactive. So it’s no wonder that people often stick to their beliefs so hard and it’s so hard to change their opinion. It is difficult to rationally process conflicting information if the brain regions responsible for rational thinking are inactive …

References

Lewicka, M. (2001). Czy jesteśmy racjonalni? w: M. Kofta i T. Szustrowa (red.) Złudzenia, które pozwalają żyć. Warszawa: PWN.
Lilienfeld, S. O., Lynn, S. J., Ruscio, J. I Beyerstein, B. L. (2011). Istnieje syndrom DDA (Dorosłego Dziecka Alkoholika). W: 50 wielkich mitów psychologii popularnej (s. 310-313). Warszawa – Stare Groszki: Wydawnictwo CiS.
Logue, M. B., Sher, K. J., Frensch, P. A. (1992). Purported characteristics of adult children of alcoholics: A possible „Barnum effect”. Professional Psychology: Research and Practice, 23, 226-232.
Lord, C. G., Ross, L., Lepper, M. R. (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal od Personality and Social Psychology, 37(11), 2098-2109.
Tyszka, T. (1999). Psychologiczne pułapki oceniania i podejmowania decyzji. Gdańsk: Gdańskie Wydawnictwo Psychologiczne.
Tyszka, T., Wielochowski, M. (1991). Must boxing verdicts be biased? Journal of Behavioral Decision Making, 4, 283-295.
Westen, D., Blagov, P. S., Harenski, K., Kilts, C. i Hamann, S. (2006). Neural bases of motivated reasoning: An fMRI study of emotional constraints on partisan political judgment in the 2004 U.S. presidential election. Journal of Cognitive Neuroscience 18(11), 1947-1958.