UCare generously supports MinnPost’s Second Opinion coverage; learn why.

Overconfident people tend to fool others about their abilities, sometimes with disastrous results

REUTERS/Brendan McDermid
Self-deception is not always beneficial. It can also be costly and even dangerous — and not just for the overconfident individual.

People who are overconfident about their own abilities tend to fool others into believing they are overtalented as well, according to a study published online Wednesday in the journal PLOS ONE.

The study also found the reverse to be true: People who under-rate their own abilities tend to be mistakenly viewed as less able by others.

“We think this supports an evolutionary theory of self-deception,” said study co-author Vivek Nityananda, a neuroscience research associate at Newcastle University in Great Britain, in a released statement. “It can be beneficial to have others believe you are better than you are, and the best way to do this is to deceive yourself — which might be what we have evolved to do.”

Yet, self-deception is not always beneficial. It can also be costly and even dangerous — and not just for the overconfident individual. As Nityananda and his co-author, Shakti Lamba, a behavioral ecologist at Exeter University, note in background information in their study, biased beliefs about people’s abilities have led to “errors with disastrous consequences, including airplane crashes, financial meltdowns and war.”

This is because, as other research has reported, overconfident people are more likely to take risks. So if the rest of us are fooled about such people’s abilities and promote them to positions of responsibilities and power, “we may be creating institutions, including banks, trading floors, emergency services and armies, that are also more vulnerable to risk,” write Nityananda and Lamba.

A real-life setting

The new study involved people in a real-life setting, thus making it the first one to offer direct evidence suggesting that fooling oneself helps fool others, according to Nityananda and Lamba.

The study’s participants were 72 undergraduate students at two British universities who were taking a small, first-year seminar that involved reviewing, debating and discussing course material with a teacher and about seven other students. After the first class, the students were asked to privately predict the grade that they believed they and each of the other students would receive when the course was completed. Seventy-one of them did not know any of the other students.

Six weeks into the course, after the students had gotten to know each other — and the course — better, they were asked to repeat the predictions.

At the end of the semester, the researchers compared both sets of predictions with the students’ actual grades. Thirty-two students (about 45 percent) had underrated how they would do compared to the others, while 29 students (40 percent) had been overconfident and 11 students (15 percent) had assessed their own ability accurately.

The comparison of the predictions also revealed a strong correlation between the students’ ratings of themselves and the ratings of their peers. The students who predicted high marks for themselves received high marks from their peers, no matter how they eventually performed in the course. The same was true for the students who underrated themselves.

Interestingly, the predictions made at the six-week point in the course were the same as those made at the beginning of the course. The overconfident students were still overrated by their peers.

Individual benefits, societal risks 

This study, like all studies, has its limitations. It involved a small number of homogenous participants (undergraduate psychology and anthropology students, mostly women, at two large British universities). The findings may not hold up among other groups of people.

Still, say Nityananda and Lamba, the results “support the idea that self-deception facilitates the deception of others.”

“While the benefits of overconfidence are apparent, it is less clear whether underconfidence can also be advantageous,” they add. “There may, however, be situations in everyday life where individuals underplay their abilities to their competitors in order to either avoid immediate conflict or to steal an advantage at the right movement, the ‘underdog’ effect. ‘Dummying up’ or appearing less knowledgeable than you are may also be a way to avoid working as hard as others.”

Yet whether the self-deception involves overconfidence or underconfidence, it likely plays “a profound role in shaping the world we inhabit,” conclude Nityananda and Lamba, “from our smallest interactions to the institutions we build.”

PLOS ONE is an open-access journal, so you can read the entire study at the journal’s website.

Comments (12)

  1. Submitted by Paul Brandon on 08/28/2014 - 09:34 am.

    The usual limitations

    While the setting may have been “real-life”, the consequences were trivial. Since the predictions were private, there would have been no social consequences. No significant costs for the subjects making Type I or Type 2 errors. So generalization to the real world is still limited.

    • Submitted by Alex Seymour on 08/28/2014 - 01:03 pm.

      It is real

      Then I would suggest that you check out the research done in behavior economics and finance. There is a ton of research done with stock selection with ordinary people, portfolio managers, stock analyst and economic forecasters. It is obviously a less controlled situation, but stock analyst and economic forecasters must publicly declare their views ahead of time. There is a strong biased towards those who offer narrow confidence intervals, that those who are confident.

      This bias does not show up in weather forecasting which shows that this cognitive defect can be overcome with diligence. Check out Nate Silver’s Signal and the Noise as a starting point.

      • Submitted by Paul Brandon on 08/28/2014 - 03:37 pm.

        Been there, read it

        A lot of the BE research is also done on college sophomores.
        And of course stock analysts can’t beat the chimp with a dartboard (or a random number generator, if you wish).

        • Submitted by Alex Seymour on 08/28/2014 - 07:20 pm.

          But it is not random

          You point on stock analyst is kind of true but also misses the point. Chimps throw darts at random, stock analyst don’t. Most stock analyst have average performance, but it you consider the large number of highly competent motivated people, who all have access to the same data, this makes sense. The market is efficient and the playing field is level. There is no secret sauce.

          Which takes us to the article. How does one stand out? One way is to make loud confident bold predictions. They tend to gain a lot of followers but have poorer predictions than average – as the article suggests. There are analyst and portfolio managers who consistently do better. However, they know their worth and charge accordingly.

          You point on BE on college students is true, but is not one of the groups that I mentioned.

          • Submitted by Steve Titterud on 08/29/2014 - 10:12 am.

            There IS a secret sauce.

            It’s called High Frequency Trading.

            This legal scam allows an enormous volume of trades to be CANCELLED – in milliseconds !! Try that with your own broker, or even if you are a trader yourself, YOU try it. Aside from that, try executing buy trades that are allowed to be fulfilled in milliseconds, then sell the same property some milliseconds later. Maybe you’ve primed the pump in the meantime with a few cancelled trades.

            See how far you get without the secret sauce of high frequency trading – short hops to the SEC’s data center (or even IN it), lots of computational horsepower, and a small staff of those who develop the algorithms which execute the scam.

            Your claim of a level playing field is quite untrue.

            • Submitted by Alex Seymour on 08/29/2014 - 02:45 pm.

              It is a secret sauce, but for a different dish.

              What point are you trying to make? HFT are not stock analyst and don’t make predictions. I would point to Warren Buffet as a counter example, who definitely trades for the long term.

              If you are trying to point out that the stock market is flawed than yes, there is a class of HFT that exploits flaws to game the system. They are parasites and this should be fixed. However, HFT is very ill defined term. I can point to other HFT which bring efficiencies to the market by adding depth to the market, compressing spreads.

              Check out this short video. It is promotional but I think it speaks to the point. Or check out Michel Lewis’s Flash Boys.
              http://www.iextrading.com/about/

  2. Submitted by Thomas Swift on 08/28/2014 - 10:17 am.

    We’d be remiss not to observe how timely this piece is, given the present occupant of the White House.

    • Submitted by Paul Brandon on 08/28/2014 - 10:51 am.

      Let’s see….

      Who was it who described himself as “the decider”?

      • Submitted by Chris Farmer-Lies on 08/28/2014 - 11:13 am.

        Not to mention “Mr. Swift’s” repeated comments confidently asserting that the occupant of that very same house would be displaced by Mitt Romney in 2012.

  3. Submitted by jody rooney on 08/28/2014 - 11:27 am.

    Sounds about right to me

    I don’t know that if they did or did not do hypothesis testing but I certainly wouldn’t have reported in an article for general consumption. But given the descriptive data not much on fire here.

    But we have already seen the overconfidence problem identified in the book “The Best and the Brightest” and I believe in the film which I haven’t seen on the Enron scandal.

    Based on the games I had to learn when I became an manager and supervisor I would say that this is a pretty good assessment.

  4. Submitted by John Lindell on 08/28/2014 - 12:33 pm.

    Isn’t this the Dunning-Kreuger Effect

    Not an expert, but the behavior described in the article seems to be consistent with this phenomena.

  5. Submitted by Rachel Kahler on 08/28/2014 - 02:41 pm.

    Overconfidence or narcissism?

    I’m not sure whether this study could distinguish the two. Narcissism could very well masquerade as simple overconfidence, since narcissists hold a view that they’re simply better. Of course they’ll do better than average (in their own minds). While there almost certainly are individuals that are overconfident but not narcissistic, I would guess that there’s a high correlation between overconfidence and narcissism. And, since there’s also a high correlation between narcissism and sociopathy (or at least reduced empathy, if not a total lack thereof), it’s not terribly surprising that overconfidence can have disastrous results since narcissists and sociopaths have a hard time caring whether or not the outcome affects anyone else negatively. And, since research has connected sociopathy (and sometimes psychopathy) with high corporate positioning, including in banks, it’s not terribly surprising that disasters involving overconfidence have been linked to financial meltdowns.

Leave a Reply