How “Noise” Affects Our Judgment

From a Bookshelf interview by William Tipper and Taylor Cromwell  in the Wall Street Journal headlined “How ‘Noise’ Affects Our Judgment”:

In expertise we trust: Throughout business, science, medicine and the law, we count on the experience and training of those in carefully vetted roles to deliver consistent information and judgments. But what if we are, collectively, unable to spot the way human variability undermines that consistency?…What if, in our quest to make institutions free of unfair bias, we are overlooking an even wider cause of misinformation and injustice?

In “Noise: A Flaw in Human Judgment,” Daniel Kahneman, Olivier Sibony and Cass R. Sunstein explain how noise undermines the workings of professionals in fields such as medicine, law, and economics, while it interferes with tasks like hiring and evaluating personnel in every industry. Wherever there is judgment, they argue, there is noise—and we are largely unaware of its power….

Nobel Prize-winning psychologist and economist Daniel Kahneman and his co-authors, Olivier Sibony and Cass R. Sunstein, sat down to discuss the findings in their book. Here are a few excerpts from our conversation:

What is it like to perform a “noise audit” for an institution and how do people respond?

Daniel Kahneman: People are shocked because they don’t expect it. There is really a deep reason for the surprise that we experience when we are confronted with the actual amount of noise that exists. . .When I look at a problem, I see it and I see the solution to it and I have the sense that I see the world as it is, because that’s the way it is. . .we each have the conviction that we see the world correctly, that the world is as we see it, and therefore we expect the person next to us if we respect that person, we expect that person to see the world the way it is, which is the way I see it.

It’s sometimes called “naive realism”. . .each of us feels that we see the world as it really is, and therefore others must see it in the same way.

What was the role of noise in how institutions, governments, and public health experts responded to the pandemic?

Olivier Sibony: One thing that struck me is that for perhaps the first time we had a chance to see how a lot of different countries and sometimes a lot of different states or regions within those countries responded to essentially the same problem. That’s an opportunity we don’t very often get.

Sometimes there is an epidemic that strikes one country and we see how that country responds. And we can speculate that another country would have dealt with it differently. But it’s only speculation. Here, of course, there are local differences and circumstances, but by and large it was basically the same problem—to [compare] countries I know well—in Italy and in France, or in Germany and in France.

We could see that in both, on the overall handling of the crisis and on very specific topics like “How do we deal with a particular vaccine. . .when we discover. . .side effects,” the judgments that are made by different health authorities. . .in different countries can be radically different. That’s a very striking example of noise, and we can then speculate about all the reasons why there is this noise. But the key thing here is there is a lot of variability in those judgments that ideally should have been identical.

What’s an example where variability in judgment is useful and desirable in the context of business decision-making process?

Cass Sunstein: You’re deciding what to do with your company in the next month. You want a large number of independent views. People can have different perspectives and different bits of information to use as inputs [to] inform and improve decisions.

Olivier Sibony: Another example would be any kind of creative process or any kind of technological search. If you have three teams of engineers looking for the solution to a technology problem or to a research problem, you very much want them to take different avenues to try to solve that problem. The reason you want that is because at the end of that process there is going to be some sort of selection. You’re going to figure out what works and what doesn’t and because there is going to be a selection, variation is welcome.

Speak Your Mind