Persily wrote, however, that his analysis falls into a second school of thought:
I don’t think most of the emotional polarization is motivated by a misunderstanding of the facts. In fact, I think many in this field make the mistake of thinking that the line to watch is the line between truth and lies. Rather, I think the critical question is usually whether the truth is relevant or not.
In this context, according to Persily, “partisan polarization resembles religious polarization. Trying to ‘disprove’ someone’s long-standing religion will rarely do much to convince them that your god is the right one.”
Seen this way, party affiliation is an identity, Persily wrote, “and it shows familiar dynamics in identity politics”:
People root for their team and find facts or other narratives that justify doing so. Remember that most people don’t spend a lot of time thinking about politics. When they do, their attitudes grow out of other affinities they have developed over time from signals sent by trusted elites or friendship networks.
Jay Van Bavel, professor of psychology and neuroscience at NYU, shares Iyengar’s view of the key role of the changing media environment. In an email, he wrote:
Much of the affective polarization is delusional or based on misperceptions. For example, people have exaggerated stereotypes about the other party (and what members of the other party think of them), and when you correct these false perceptions, they quickly become less hostile.
People are motivated, he continued,
to assert evidence that confirms their beliefs and affirms their identities. For committed supporters, they are often more motivated by these social goals than by a desire to be accurate. People also share misinformation for social reasons: it can signal loyalty and help people gain status in some partisan communities.
An important component, Van Bavel said, “is based on the misperceptions they’ve absorbed from their social network in (social) media stories. It suggests that if we could just provide accurate and diverse portrayals of other groups, it could reduce the growing trend towards affective polarization”.
But, he warned, “correcting misinformation is extremely difficult; the impact tends to be quite small in the political realm and the effects don’t last long.”
In a 2021 paper, “Identity concerns drive belief: The impact of partisan identity on the belief and dissemination of true and false news. Andrea Pereira, Elizabeth Harris and Van Bavel surveyed 1,420 Americans to see which of the following three alternatives best explained the rise and spread of political misinformation:
The ideological values hypothesis (people prefer news that reinforces their values and worldviews), the confirmation bias hypothesis (people prefer news that conforms to their preexisting stereotypical knowledge), and the political identity hypothesis ( people prefer news that allows them to believe positive things about political in-group members and negative things about political out-group members).
His conclusion:
Consistent with the political identity hypothesis, Democrats and Republicans were more likely to believe news stories about in-group value-defending behavior or out-group value-undermining behavior. Belief was positively correlated with willingness to share on social media across all conditions, but Republicans were more likely to believe and want to share political fake news.
Several studies have been published in recent years describing the success or failure of various approaches to reducing levels of misperception and affective polarization. The difficulties facing these efforts are reflected, in part, in an October 2022 paper, “Interventions That Reduce Affective Polarization Do Not Necessarily Improve Antidemocratic Attitudes,” by Jan G. VoelkelStanford sociologist and eight colleagues.
The authors found that even when “three depolarization interventions reliably reduced self-reported affective polarization,” the interventions “did not reliably reduce any of the three measures of antidemocratic attitudes: support for nondemocratic candidates, support for partisan violence and prioritizing partisan ends over democratic means.” .”
