Cognitive Bias

15 July 2017
Aristotle defined humans as the rational animal. But he was wrong! The human mind is riddled with cognitive biases. At last count, there are something like 150 named cognitive biases—confirmation bias, in group bias, loss aversion, the Ikea effect, the halo effect, endowment effects. And every time you turn around, some clever psychologist is naming a new one. But whatever you name them, cognitive biases are a problem. They distort gender relations, racial relations, employment, education, politics, even science. And we’re mostly not even aware of them! That makes them very hard to correct. Indeed, it is not at all obvious that we can correct them. They may, in fact, be entirely beyond our control. 
Here’s an example of cognitive bias at work. Take a group of teachers. Randomly divide them into two groups. Give each group the exact same set of papers to grade. Put African American sounding names on the one set, Anglo sounding names on the other. The African American sounding names will elicit lower grades than the Anglo sounding names. Before you conclude that the teachers are just bigots, you should know that you get the same result for all different sorts of people, including the most open minded people in the world, the kind who insist that they would never intentionally discriminate against anybody. We ALL suffer from this sort of implicit cognitive bias: black, white, male, female. It doesn’t matter.  
Now think about how we might try to overcome our cognitive biases. Take our teachers. Maybe we should just instruct them to try harder, to focus like a laser on the on the quality of the papers they are given and to make concerted efforts to block out extraneous factors. If you give teachers an instruction like that, they will actually try hard to follow it. But ironically, for some cognitive biases, trying may actually make it worse. Trying can biases you toward thinking you’ve succeeded. Once you think you’ve succeeded in eliminating bias, you may think to yourself “Look, ma! No more bias.” And that can lead you to think, “This is what I think, and I’m not biased, so it must be true!” There is even a clever name for this effort induced cognitive bias. It’s called the “I think it, therefore it’s true” effect. The point is that we can’t just will our biases away. That’s because many of them are the result of unconscious mechanism that are beyond the reach of our conscious control.  
You might think that education and training would help. But there’s little to reason to take education and training as a cure all for our cognitive biases. Indeed, highly educated and trained scientists still suffer from cognitive biases. They even suffer from cognitive biases that are peculiar to scientific research. This shows that no domain of human cognition is immune to the effects of cognitive bias. And that includes us oh-so-rational philosophers. 
Before we throw up our hands in despair and conclude that the mind is just a junkyard littered with cognitive biases, it’s important to realize that the mind is also the product of evolution. And evolution doesn’t do junkyards. It does beneficial adaptations. And though cognitive biases may seem the very opposite of beneficial to us now, back in the day, on the Pleistocene savannah, they played a very adaptive role. In those circumstances, if you took time to consider things from all angles, you wouldn’t survive. As a consequence, natural selection designed our brains to make quick and dirty decisions on the fly. Sure, they didn’t always yield the truth. But they helped to keep our forebears alive. This means that perhaps we should regard our cognitive biases as features rather than bugs!
The problem with this line is that we’re not on the Savannah any more. Quick and dirty reasoning may have been good enough for hunter gatherers, but won’t necessarily cut it in our much more complex environment. To thrive in today’s world, we need to “debias” our minds—evolution be damned! Not that that would be an easy task. But it may not be as hard as one might think. Take those teachers. They can debias just by anonymizing student papers. That doesn’t take Herculean effort. And it doesn’t take a lot of education either. 
We shouldn’t let that fact make us think that there will necessarily be similar easy fixes in all cases. That would show display another kind of cognitive bias—what’s called optimism bias. But you don’t have to suffer from optimism bias, to think that if we take our biases one at a time and try to think of piecemeal strategies for overcoming them, we might just get somewhere, at least with some of them. We’d love to have you join us as we strategize together about ways of overcoming our many cognitive biases. 

Comments (4)

David99999's picture


Saturday, January 27, 2018 -- 11:45 PM

Some good-old-fashioned

Some good-old-fashioned mindfulness might help combat bias.

Tim Smith's picture

Tim Smith

Monday, December 16, 2019 -- 4:33 AM

That would be mindfulness

That would be mindfulness bias.
Funny ... but not so funny.

Harold G. Neuman's picture

Harold G. Neuman

Friday, December 13, 2019 -- 10:26 AM

Good point, David. Oh, that

Good point, David. Oh, that it could be true for more of us. My take on this is as follows:

It is rare to find mindfulness among those who suffer from a lack of mind fullness.

I wrote that somewhere---maybe in remarking on one post or another on the PT blog. Regarding this post and at least one previous to it, I opined that cognitive bias is learned behavior, and further perhaps, constitutes belief which would make it suspect from the beginning. There has been much cognitive bias exhibited on TV lately, particularly in connection with the goings on in government. It might be also called the denial of fact in the face of partisanship. One man's facts are another man's fiction. So, it was with sympathy that Simon wrote: ...a man hears what he wants to hear and disregards the rest...

Tim Smith's picture

Tim Smith

Wednesday, December 18, 2019 -- 7:12 AM

Bias is not what Ken thought

Bias is not what Ken thought it was.

David and Harold, alike, can use their reasoning until the cows come home they will still have the threat of bias in their thought, without any real resource to justify their stoicism.

Ruth Millikan is my guide at the moment on this subject. She has made inroads and arguments that advance the concepts of construction and natural kind.

Concepts are constructed, of what I don't exactly know. Unicepts are a good start.

Cognitive bias is a construct of concepts but most certainly a unified structure of two or more concepts.

The placebo effect, salience and recency are not cognitive bias - but that opens up the problem. I can't finish that here.

I am with Harold and David99999... there is a problem here.