In 2020, the match-making website OkCupid asked 5 million hopeful daters around the world: “Could you date someone who has strong political opinions that are the opposite of yours?” Sixty per cent said no, up from 53% a year before.

Scholars used to worry that societies might not be polarised enough. Without clear differences between political parties, they thought, citizens lack choices, and important issues don’t get deeply debated. Now this notion seems rather quaint as countries have fractured along political lines, reflected in everything from dating preferences to where people choose to live.

Just how stark has political polarisation become? Well, it depends on where you live and how you look at it. When social psychologists study relations between groups, they often find that whereas people like their own groups a great deal, they have fairly neutral feelings towards out-groups: “They’re fine, but we’re great!” This pattern used to describe relations between Democrats and Republicans in the US. In 1980, partisans reported feeling warm towards members of their own party and neutral towards people on the other side. However, while levels of in-party warmth have remained stable since then, feelings towards the out-party have plummeted.

The dynamics are similar in the UK, where the Brexit vote was deeply divisive. A 2019 study revealed that while UK citizens were not particularly identified with political parties, they held strong identities as remainers or leavers. Their perceptions were sharply partisan, with each side regarding its supporters as intelligent and honest, while viewing the other as selfish and close-minded. The consequences of hating political out-groups are many and varied. It can lead people to support corrupt politicians, because losing to the other side seems unbearable. It can make compromise impossible even when you have common political ground. In a pandemic, it can even lead people to disregard advice from health experts if they are embraced by opposing partisans.

The negativity that people feel towards political opponents is known to scientists as affective polarisation. It is emotional and identity-driven – “us” versus “them”. Importantly, this is distinct from another form of division known as ideological polarisation, which refers to differences in policy preferences. So do we disagree about the actual issues as much as our feelings about each other suggest?

Despite large differences in opinion between politicians and activists from different parties, there is often less polarisation among regular voters on matters of policy. When pushed for their thoughts about specific ideas or initiatives, citizens with different political affiliations often turn out to agree more than they disagree (or at least the differences are not as stark as they imagine).

More in Common, a research consortiumthat explores the drivers of social fracturing and polarisation, reports on areas of agreement between groups in societies. In the UK, for example, they have found that majorities of people across the political spectrum view hate speech as a problem, are proud of the NHS, and are concerned about climate change and inequality.

As psychologist Anne Wilson and her colleagues put it in a recent paper: “Partisans often oppose one another vehemently even when there is little actual daylight between their policy preferences, which are often tenuously held and contextually malleable.”

This relative lack of divergence would, of course, come as a surprise to partisans themselves. This is the phenomenon of false polarisation, whereby there is widespread misperception of how much people on the left and the right are divided, not only on issues but also in their respective ways of life. When asked to estimate how many Republicans earn more than $250,000 a year, for example, Democrats guessed 38%. In reality it is 2%. Conversely, while about 6% of Democrats self-identify as members of the LGBT community, Republicans believed it was 32%. New research from Victoria Parker and her colleagues finds that partisans are especially likely to overestimate how many of their political opponents hold extreme opinions. Those overestimates, in turn, are associated with a disinclination to talk or socially engage with out-party members, avoidance that is likely to prevent people from forming more accurate impressions of the other side.

What drives these misperceptions? And why do citizens so dislike one another if they aren’t necessarily deeply divided on policy matters? Politicians certainly have incentives to sharpen differences in order to motivate and mobilise voters, rallying support by portraying themselves as bulwarks against the barbarians on the other side. Divisiveness also plays well on social media, where extreme voices are amplified. Moral outrage is particularly likely to go viral.

In a recent project led by Steve Rathje and Sander van der Linden at Cambridge University, we examined more than 2.5m posts on Twitter and Facebook. We found that posts were significantly more likely to be shared or retweeted if they referenced political opponents. Every word about the out-group increased the odds of a post being shared by 67% – and these posts were, in turn, met with anger and mockery.

In this increasingly toxic environment, reducing false polarisation and affective polarisation are major challenges. It is often suggested, for example, that if people were only to expose themselves to perspectives from the other side, it would breed greater understanding and cooperation. Yet this intuition turns out to be flawed.

Sociologist Christopher Bail and his colleagues offered sets of Democrats and Republicans money to follow a bot that would retweet messages from politicians, media companies and pundits every day for a month. Importantly, the messages always came from the other side of the political spectrum. Far from promoting harmony, it backfired. After a month of being exposed to conservative talking points, Democrats’ attitudes had become, if anything, marginally more liberal. And Republicans became more conservative following their diet of liberal views. When what you see from the other side strikes you as biased or obnoxious, it doesn’t endear you to their perspectives.

In this regard, the behaviour of elites matters. Political scientist Rasmus Skytte showed people messages from politicians that were either civil or rude. Interestingly, aggressive and unkind messages didn’t reduce trust in politicians or increase affective polarisation. It seems that incivility is what people have come to expect. But when they saw polite and respectful messages, they subsequently felt more trust towards politicians and became less affectively polarised.

These results suggest that we should expect better from our leaders and those with large platforms. Don’t reward divisive rhetoric with “likes”. Instead, follow politicians and pundits who embody norms of respect and civility, even when they disagree on policy matters. In fact, many of us might be better off if we took a break from social media altogether. When economists found that whenpeople who were encouraged people to disconnect from Facebook for a month spent less time online and were less politically polarised. They also experienced improved psychological wellbeing.

No one these days is worried that our societies are insufficiently polarised. But because so much of the polarisation is about emotions and identities rather than issues, it is still not clear that citizens are presented with good choices or that important issues are being deeply debated. Here again, we must expect better. Demand that politicians and pundits get into policy specifics. Let’s focus more on actual ideas for solving actual problems, where we, as citizens, may well turn out to agree on more than we realise.

Dominic Packer and Jay Van Bavel are psychologists and the authors of The Power of Us. To support the Guardian and Observer, order your copy at guardianbookshop.com. Delivery charges may apply.

Further reading

Uncivil Agreement: How Politics Became Our Identity by Lilliana Mason (Chicago, £19)

Breaking the Social Media Prism: How to Make Our Platforms Less Polarizing by Chris Bail (Princeton, £20)

The Wealth Paradox: Economic Prosperity and the Hardening of Attitudes by Frank Mols and Jolanda Jetten (Cambridge, £19.99)