This cognitive bias can push people into more extreme ideological positions

Numerous psychological studies have found that we seek out information that supports our pre-existing views – and avoid information that might contradict them. This is particularly true for politicised issues: for instance, people will choose to forgo cash in order to avoid reading opposing views on topics like same-sex marriage and gun control.

The implication of these studies is that this so-called "selective exposure bias" may be pushing us into more polarised positions. After all, if we ignore evidence that could contradict our beliefs, we may end up feeling even more strongly that our own view is the correct one. Yet, as the authors of a new study in Psychonomic Bulletin & Review point out, although plenty of research has shown that this bias exists, there hasn't actually been much work on how it affects our beliefs and behaviours. Now the researchers find that the bias can indeed shape people's beliefs in at least one area: their attitudes towards diversity.

Jonas De keersmaecker and Katharina Schmid from Universitat Ramon Llull recruited more than 2,000 Spanish citizens to participate in the study. First, participants responded to three questions measuring pro-diversity beliefs – essentially, how much they believed that cultural or ethnic diversity is good for a society. They also indicated whether they were for or against increasing Spain's support for refugees from North Africa.

Participants then had to choose between two options for a task that would take place a few months later. One option was to read eight arguments about refugees from the opposite perspective to their own (i.e. people in favour of more refugee support would read arguments against helping refugees, and vice versa); if they picked this option they would enter a lottery for a €10 payment. The other option was to read eight arguments consistent with their own beliefs, and to enter a lottery for a smaller payment of €7. People who settled for the lower payment to read arguments consistent with their own beliefs were classified as showing selective exposure bias.

Five months later, the participants completed the second part of the study – but they didn't actually have to read the arguments at all. Instead, they simply completed the scale measuring pro-diversity beliefs once again.

The team found that 58.6% of people showed selective exposure bias, choosing to read arguments consistent with their own beliefs for less money. And this bias seemed to influence participants' beliefs about diversity – at least among those who were against increasing support for refugees. Within this group, those who had chosen to read arguments against helping refugees had less favourable attitudes towards diversity five months down the line, compared to those who were willing to read pro-refugee arguments. (The researchers took into account participants' baseline attitudes towards diversity in this analysis).

In other words, people who were against helping refugees and who were also not receptive to information that might challenge their beliefs had more negative opinions about diversity over time, compared to others who were similarly against helping refugees but who were open to hearing other arguments. And remember that participants never in fact saw any arguments, so this effect wasn't caused by actually reading information in the study that contradicted or supported their beliefs. Instead, the researchers write, "[t]he present study suggests that negative opinions about diversity might partly originate from one's bias to avoid positive information about diversity over time".

Interestingly, people who were in favour of increasing support to refugees didn't show this effect: they had similar attitudes towards diversity whether or not they were receptive to reading opposing arguments. Why is this the case? Only future research will tell. But, as the authors point out, this shows why it's important to study how the selective exposure bias might affect the behaviours and attitudes of different groups.

The results aren't great news for attempts to increase support for diversity. If people who are against diversity in the first place also avoid any initiatives that might challenge their beliefs – such as opportunities to have positive interactions with people from another culture or ethnicity – they may become more entrenched in their position. The researchers propose one solution: "prodiversity initiatives should be tailored in such a way to avoid positive intergroup experiences being optional, but to make them inevitable."