The oil slick effect, or why we systematically overgeneralise

9th June, 2022

If you have colleagues, what do you think of them? Are they smart? Competent? Motivated? Open to new ideas? Good communicators? Do they work well as a team? The answer may not depend on what you think. And that fact suggests a reason why the modern world now seems so poisonously polarised.

In the 1970s, the psychologist Barry Staw gave a collaborative task to groups of strangers, inviting them to analyse some corporate data and make predictions about the company’s future earnings and sales. When the task was complete, he told each participant how well their group’s forecasts had worked out. Then he asked these individuals to evaluate the group they’d been working with.

But Staw was telling a white lie: he gave each group’s forecast a good or bad rating purely at random. There was no connection between how well the group did and how well Staw told them they’d done. Nevertheless, Staw found that when people believed their group had made an accurate forecast, they told him that they’d been working with open-minded, motivated, clear, intelligent and collegiate people.

But when they were falsely told that their group had made poor predictions, they explained to Staw that this was no surprise, as the group was narrow-minded, lazy, abstruse, foolish and mutually antagonistic.

Subsequent researchers found the same pattern, even when they repeated the experiment with well-established teams. As Phil Rosenzweig explains in his book The Halo Effect, this behaviour is not confined to colleagues. We have a systematic tendency to overgeneralise both praise and blame. Profitable companies are presumed to have superior policies and procedures across the board. This halo effect operates in reverse, too: scandal-struck politicians see their opinion poll ratings fall on every issue, from economic competence to foreign policy. Apparently we struggle to acknowledge that something can be good in some ways and bad in others, whether that thing is a president, a corporation or our own teammates.

The reverse halo effect is sometimes called the “devil effect” or the “horn effect”. Neither term has quite caught on. So let me offer another: the oil slick effect. Disagreements, like oil slicks, seem to spread much further and more ruinously than we would think. It’s not possible for somebody simply to be wrong about something; they must be wrong about everything, and wicked, too. The oil slick covers everything and ruins everything.

I can’t help but wonder if this oil slick effect is worse than it used to be. Consider the following data, reported in Ezra Klein’s book Why We’re Polarized: in 1960, when Americans who supported the Republicans or the Democrats were asked whether they would object to their son or daughter marrying across party political lines, very few were perturbed: 5 per cent of Republicans and 4 per cent of Democrats.

When the same question was asked 50 years later, opposition to inter-party marriage had risen almost tenfold, to 49 per cent of Republicans and 33 per cent of Democrats. Politics moved from the kind of thing sensible people could agree to ignore, to an all-consuming Sharks and Jets-style vendetta in which to cross the political divide is an unforgivable betrayal. The oil slick spread from the political to the personal.

This might be understandable if the policy stakes had risen, but the evidence suggests that policy itself is almost irrelevant. Republicans in the US used to be free-traders; in the UK, the Conservatives used to be pro-business. Most of their voters do not seem to object to large shifts in their policy platforms – their loyalty is to something else.

The halo effect is not new. It was first named and identified by the psychologist Edward Thorndike over a century ago. Why might it have become more acute? One clue comes from a study conducted a decade ago by three social psychologists, Angela Bahns, Kate Pickett and Christian Crandall. They studied friendship groups on small and large university campuses. Large campuses seemed more diverse on the surface but, with a wider choice of possible friends, students clustered in like-minded cliques. On smaller campuses, with less choice, they were forced to forge friendships across potentially awkward differences in attitudes to politics, religion, sex and lifestyle choices such as exercise and smoking.

Perhaps the modern world is more and more like the big campus – full of a huge diversity of views, and yet offering us every option to associate with people just like us. This is most obvious in social media, where by design we self-silo, but we can also select our own podcasts and politically sympathetic TV channels.

As Bill Bishop argued in his book The Big Sort, we are even clustering into socially homogeneous neighbourhoods. The world is a wider, more diverse place, and that means our choices of who we deign to read, watch or even have a drink with may be narrowing. The halo effect has long been a feature of our psychology, and there has always been a temptation to let the oil slick poison our thinking. That toxic temptation used to leave a person isolated, with nobody available to live up to their standards of purity. Today, the oil slick can spread freely.

Written for and first published in the Financial Times on 13 May 2022.

The paperback of The Data Detective was published on 1 February in the US and Canada. Title elsewhere: How To Make The World Add Up.

I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.

Pin It on Pinterest

Share This