Book review: The Perils of Perception, by Bobby Duffy

After almost 25 years at Ipsos MORI, Bobby Duffy is an expert in social research. One of his particular interests is how we get things wrong, something that the agency has investigated through interviews with 100,000 people in 40 different countries. That work has now been brought together in the book The Perils of Perception: Why we’re wrong about nearly everything.

You might have come across the sorts of misperceptions we’re talking about. To take a few examples:

Teenage pregnancy: when asked what percentage of British teenage girls give birth every year, people guessed 19%. The correct answer is 1.4%.
Religion: people in France were asked what percentage of the population they thought was Muslim. They guessed 31%, when the reality is 8%.
Immigration: asked what the proportion of immigrants is in the America population, the average answer was 33%, over double the actual figure of 14%.

The book is packed with these sorts of delusions, showing how we don’t believe that crimes rates are falling, that we think everyone else is having more sex than they really are, and that we dramatically overestimate how many 30-somethings still live with their parents.

Books or media stories that focus on this sort of public ignorance can often be a little smug, delighting in wrong-footing their readers and highlighting how little we know. Duffy doesn’t do that, partly because he remembers how annoying it was when university lecturers did it, and mostly because he sees value in the misperceptions themselves: “they tell us about how we think, what we’re worrying about, how we see ourselves relative to others, where we think the norm is, and therefore how we’re likely to act ourselves. We can learn a lot by understanding why we’re so often so wrong.”

Importantly, having misperceptions doesn’t make you dumb. As the book repeatedly explains, it’s not just about knowing facts, but “how we think that causes us to misperceive the world”. There are a whole host of psychological reasons why we overestimate some things and underestimate others. These instinctive and subconscious mental shortcuts are called heuristics, and we all use them. If we’re guessing a number is going to be high, we hedge our estimate downwards. If we’re worried about something – immigration levels, for example – we’ll overestimate it. We are biased towards ourselves and people like us. These tendencies are there for a reason. We employ those thinking tools to avoid being overwhelmed by a world of information, to help us make faster decisions or even to keep us safe. Nobody is free from misperceptions, even scholars who have spent their whole lives studying them.

On the other hand, a lot of our prejudices are rooted in misperceptions, and prevailing myths can lead to discrimination and scapegoating. We can go badly wrong when policies are enacted based on perceived problems rather than real ones. Our misperceptions can lead us to ignore or deny progress, and then we risk throwing it away. Pervasive falsehoods can do a lot of damage, such as anti-vaccination campaigns and climate change denial. We need to be more aware of our misperceptions and open to correction, and Duffy concludes with a list of ten learning points. They include cultivating scepticism but not cynicism, being aware of extreme examples, teaching critical thinking in schools, and reading outside our ‘bubble’.

There is some overlap with Factfulness, which I read earlier this year. The main difference is that Factfulness covers international issues and The Perils of Perception looks at national questions, and they’re different in style. Duffy’s book is lower on anecdotes and isn’t as entertaining, but it’s stronger on the psychology and sociology. It makes direct connections to contemporary issues including Brexit, the changing debate around social media, and the rise of ‘fake news’. It’s clear, balanced, and cautious in its conclusions – and manages to be optimistic about the state of the world without being complacent about what still needs improvement.