Think for yourself: how to judge expertise in a time of conflicting opinions

Originally published on June 14, 2020.

Do you regularly outsource your thinking?

Vikram Mansharamani, is a Harvard University lecturer and the author of the book Think For Yourself: Restoring Common Sense in an Age of Experts and Artificial Intelligence. 

Mansharamani believes that because we're overloaded with information, we frequently "turn to experts and technologies that can help us filter the information, decide what's important, and help us optimize our decision making process. However, in the process we find we give up our autonomy and there are some costs to doing so and that's really what I'm concerned about."

He spoke with Spark host Nora Young about why he believes we've mindlessly turned over our thinking to technology, experts and rules. 

Here is part of their conversation.

Isn't deep expertise a good thing? The last time I checked no one was asking me to perform open heart surgery, for example.

I have no problem with deep expertise, and in fact we do want to rely on experts. My argument really is not one that we should dismiss experts but more that we need to, as I say in the book, keep experts on tap, not on top. It's really about harnessing the power of expertise without giving up control or management of that expertise. So I find it equally problematic to blindly defer to experts as I do to those who completely dismiss experts. So it's not about dismissing or deferring to experts, it's really about the nuance of staying in control and extracting the best value we can from experts.

As you point out, behavioural economics suggests that humans are far from these hyper-rational maximizers of benefits. We can be irrational. We can be not very good, for example, at weighing probability. So should we really be that ready to trust ourselves?

Sure. So I guess what I'm getting at is not only a trust factor, in terms of trusting ourselves, but more importantly just being mindful of that. The fact that you yourself, in this question, asked whether you should trust yourself indicates a mindfulness about whether you should or shouldn't. That is far better and more appropriate than those who blindly, without thinking about it, go forth. That is my main point here, which is we need to think about how we think. We need to think about the inputs we put into our thinking process. We need to mindfully manage where we focus.

At one point in the book you point out, and this is quite astonishing, that experts in general are not much better than you or I in predicting outcomes, and that people with deep specialization can be even worse than non-experts at prediction. So why does that happen? What's going on there?

So in terms of navigating uncertainty and thinking through complex, dynamic, forward-looking scenarios where we don't have a known probability set, etc. — environments like that are particularly dangerous for expert predictions. 

And there's a lot of reasons for that, but generally it comes from the belief that was formed around some significant research conducted by Philip Tetlock, who was a professor who studied expert decisions and expert predictions over a long period of time, and he found that the more focused the expert, the less likely their prediction would be successful. And in fact those outside the area of expertise were more successful at predicting. 

So I guess that's really a little bit of what I'm getting at when I say that experts are perhaps not best to rely upon in times where there's a lot of uncertainty and a lot of dynamism.

Given that we can't thoroughly research and coordinate all of our experts for everything, how do we decide when you say, 'okay, in this case I'm just going to listen to the expert and I'm not going to research it, whereas in this case this is important enough to me that I actually am going to take charge of it'?

Sure. I think it has to do with the stakes. A very simple answer is look for a high-stakes decision: you probably want to invest the time in connecting dots and asking multiple people multiple questions. So a simple way, you might think it's intimidating, but I would suggest it should feel empowering. Why not? 

There are interesting cross connections that will emerge when you take a beginner's mind and ask questions that might be naive asking people who you think may not be the relevant person to ask. Ask everyone. What do you think? What do you think? You might come off as naive and perhaps even less informed, but I would argue in that process you will come to understand your decision a lot better.

So, Vikram, let's say you convinced me, I want to cultivate my self-reliance. Are there things that I can start to put into practice this week to kind of develop those muscles?

I would ask when you get handed a piece of advice or an expert asks you to do something, ask a very simple question. You can even say, in a non-threatening way, that's fascinating, I really appreciate the advice. Now just so I can get myself more informed. Why do you believe that? How did you get your information? And just ask that simple question. I think that simple question will enable you to have a different form of conversation with those giving you advice. 

This interview has been edited for length and clarity. To hear the full conversation with Vikram Mansharamani, click the 'listen' button at the top of the page.