I’m reading Charles Duhigg’s book “The Power of Habit: Why We Do What We Do in Life and Business”, and it’s leaving me with an uneasy feeling. Duhigg is a great reporter, whose work in the New York Times I enjoy and that is why I bought the book. But in the age of the replication crisis in the social sciences, it’s hard to know what to trust.
The book is about how habits form, what habits do, and how to change them. It’s full of useful and interesting stories about business that I enjoy reading in their own right. But then there’s is the psychology part of it. References to studies with small samples raise my skeptical alarms. Other studies just sound like the kind that have failed to replicate. Again, I know Duhigg as careful and thoughtful and I tell myself this when he assures the reader that many studies show the same thing.
For example, Duhigg recounts in detail a famous experiment that is alleged to show how willpower is a limited source that can be depleted. The study basically had people resist eating fresh backed chocolate chip cookies, and showed that this made them give up on an impossible puzzle faster than a control group. Willpower is like a muscle that gets depleted is the lesson to be learned.
This feels like exactly the kind of study that would die under the scrutiny of replication that has knocked down so many social science claims. And indeed, as Daniel Engber writes in Slate, this theory is coming under a lot of scrutiny as willpower depletion studies fail to replicate.
This isn’t to say that I know the claims in Habit are wrong, or even a criticism of that book. I’m not an expert on willpower depletion, and the conclusions seem up for debate rather than fully debunked at this point. This is more a meta commentary on how bad science has cast so much doubt that it is harder to know what to trust when it comes to popular social science books. So here is my suggestion: publishers should publish the books along with peer reviews from experts that can be found free online. I’d get a lot more out of these books if I knew what experts saw as the most important caveats, and what they saw as the firmest ground.
Until then, it’s harder these days to confidently think it’s worth spending time reading many kinds of popular science books.