Anyone encountered this 'inverted' version of Dunning-Kruger?

In My Humble Opinion
1

The Dunning-Kruger effect observes that people with low ability tend to overestimate it.

But there are times I’ve observed the opposite: people who are really smart, who are really hard to convince when they’re wrong about something. Often, the thing they’re wrong about is in an area outside their expertise.

One example: a lifelong friend of mine (“J”) is indeed really smart. He is literally a practicing brain surgeon, and used to be a rocket scientist (sorta: he developed artificial intelligence for NASA). He is also an accomplished musician and composer–we have collaborated off and on over the years. A long time ago, I was at his home watching MTV with him, when the video for the song “Cryin’” by Aerosmith came on. He listened for a moment, and asked if I remembered who had done the song originally. When I answered that it was not a cover, but a new, original song by Aerosmith, he disagreed. At the end of the song, when the credits came on, stating that it was written by three members of Aerosmith, and was on their recently-released album, he still insisted he had heard the song before. I suggested that he was responding to the fact that the song was catchy, and the melody contained a lot of pop cliches, which would make it seem more familiar the first time you heard it. But he was insistent he had heard the entire song, years before. We dropped the matter for lack of importance.

Another example: a childhood friend of mine (“B”) is a brilliant, very successful attorney. She is a very outspoken capital-L Libertarian and incessant proponent of the flat tax, neither of which I will comment on. What I will comment on is the fact that she is an equally vocal Truther, frequently showing me links to analyses by purported experts claiming that, e.g., WTC7 was destroyed by explosives planted in the building, not by parts of the towers falling on it as the government claims. I confess that I haven’t really delved into these analyses, having concluded long ago that the events of 9/11 were caused by terrorists crashing hijacked jetliners into the Twin Towers. Another reason I haven’t delved into them is that most of the people I’ve known who have made such claims did not impress me as being very credible or smart. But “B” is certainly an exception to this.

Anyone else observed this?

Also, has this phenomenon–that smart people may be harder to convince that they’re wrong than are less-smart people–been compared and reconciled with the Dunning-Kruger effect by any psychologists?

2

Backfire effect? or more like belief perseverance, both of which are forms of confirmation bias.

I don’t know if either have another name.

3

I think what you’re describing is still the Dunning-Kruger effect. Nothing about the Dunning-Kruger effect says it only applies to unintelligent people. It applies when the person lacks expertise in the particular subject matter at hand, regardless of how intelligent they are.

In fact in an interview I heard on NPR David Dunning said literally everyone, even him, experiences the Dunning-Kruger effect from time to time.

4

Linus Pauling is a famous example. Nobel Prize winner (twice, actually, but once for science) and yet he had this thing about the wonder cure Vitamin C that he just wouldn’t let go of.

j

5

It’s a new word to me, but I ran across Ultracrepidarianism which is maybe what you’re talking about?

6

I disagree. Putting aside what Dunning said in a radio interview, I feel the Dunning-Kruger effect is fairly specific.

It doesn’t say that people in general sometimes misjudge their own abilities. It says that intelligence is the ability we use to make reasoned judgements and therefore people with low intelligence have a poor ability to judge things. That seems pretty obvious. What makes it interesting is when you apply the general principle back on to their own intelligence and see how people with low intelligence are unable to judge that they have low intelligence.

To make the point clearer, put aside judging intelligence and imagine you were asking two people - one intelligent and one non-intelligent - who never water-skied to judge how good they would be at it if they tried.

The intelligent person would reason, “Water-skiing is a skill that people have to learn with a lot of practice. I have never water-skied so I have no practice in it. Therefore, I would probably do very poorly at water-skiing.”

The non-intelligent person would reason, “I don’t know. I guess I’d do okay.”

Now imagine the same two people are taking a multiple choice test. But it’s in French, a language which neither one speaks. And they’re asked to guess their scores.

The intelligent person would reason, “I didn’t understand the questions or the answers, so I just made random guesses. There were four choices to each question so around a quarter of my random guesses will be right. I think I got a 25.”

The non-intelligent person would reason, “I don’t know. I guess I got around a 90.”

Finally, imagine the same two people are taking a multiple choice test in English. And they’re asked to guess their scores.

The intelligent person would reason, “I’ve taken multiple choice tests like this in the past and I generally get around a 90. This test seemed the same to me. I think I got a 90.”

The non-intelligent person would reason, “I don’t know. I guess I got around a 90.”

You can see the difference in how intelligent people and non-intelligent people think. The intelligent person is able to use evidence and reason to assess his own abilities. He’s able to make good judgements even about situations where he determines he would do poorly. He understand both his abilities and his limitations.

The non-intelligent person doesn’t do that. He doesn’t even try applying evidence or reason to determine the correct answer to a question. He just guesses an answer. And if you asked him why the other guy made more accurate judgements than he did, he probably wouldn’t say “He used reason and I didn’t. So his methods produced better judgements than mine did.” He would be more likely to say “He got lucky and guessed better than I did.” The non-intelligent person doesn’t understand the process the intelligent person is using and why it works.

All of which, I realize, is something of a sidetrack away from the OP. I’ve explained why I don’t feel it’s the Dunning-Kruger effect. What I feel it is is simple arrogance. Somebody who is used to being right in one area in which they have knowledge can start assuming they are right in all areas, even in areas where they do not have any specific knowledge.

You can be intelligent and capable of learning the information needing to be an expert in a subject. But you still need to learn the information. To use the OP’s example, his friend must be intelligent in order to have learned enough about the law to have become a brilliant lawyer. So we can assume that if she chose, she could learn enough about economics or engineering to be a good or even brilliant economist or engineer. But having the capability to learn these subjects is not the same as having actually learned them. A brilliant lawyer should not assume they understand economics or engineering as well as they understand the law.

7

As I learned in a social psychology class Dunning and Kruger studied intelligent people too. And yes, the opposite is also true, many smart people think they’re less competent.

cite

8

Based on the 1999 paper by Kruger and Dunning, the “effect” is largely described as inept/ignorant people not having the cognitive ability to recognize their deficiencies, or as it’s been generalized, “dumbass and too stupid to know it”. Secondarily, bright people tend to underestimate their performance in comparison to others. In followup, other sources have interpreted the Dunning-Kruger effect as occurring in otherwise intelligent people.

“The reality is that everyone is susceptible to this phenomenon, and in fact, most of us probably experience it with surprising regularity. People who are genuine experts in one area may mistakenly believe that their intelligence and knowledge carry over into other areas in which they are less familiar.”

The phenomenon of brain surgeons and Nobel Prize winners embracing spectacularly stupid and illogical beliefs, including bogus conspiracy theories probably falls more under the category of selective deficits in critical thinking capacity. They have the intellectual capacity to accumulate and evaluate evidence in an objective fashion when it comes to their area of expertise, but lack the discipline to arrive at logical conclusions when it involves something outside their field, particularly when they are emotionally invested in a particular belief. Someday one hopes that researchers will discover how this compartmentalization of brain function/dysfunction occurs on an anatomic and physiologic basis.

9

Am I mistaken, or is the Dunner-Kruger effect less about intelligence and more about people with a low ability to perform something specific having an inability to recognize how bad they are at it? I know Mensans who are convinced that essential oils are good for your health and believe the environment, and not germs, are what causes sickness. None of them are biologist though.

10

I have no opinion on whether it’s also the Dunning-Kruger effect, but it’s definitely Ultracrepidarianism.

11

Are you also putting aside what the seminal paper says? Because it doesn’t say anything about intelligence. It’s titled “unskilled and unaware of it” and says “overestimation occurs, in part, because people who are unskilled in these domains suffer a dual burden: Not only do these people reach erroneous conclusions and make unfortunate choices, but their incompetence robs them of the metacognitive ability to realize it. … In essence, we argue that the skills that engender competence in a particular domain are often the very same skills necessary to evaluate competence in that domain.”

It says quite directly that, indeed, people in general sometimes misjudge their own abilities. It’s clearly stated as domain-specific. None of the experiments suggest the participants are generally unintelligent or intelligent people, unless you think unintelligent people are lacking even in humor and grammar.

12

By the way, one thing that really bothers me about the pop culture, Internet version of DK is that you’ll hear people suggest that the skilled people are humble and rate themselves low. If you look at the graph in the original paper, it’s the opposite. While everyone rates themselves above average, the actual top performers rate themselves higher than anyone else rates themselves. The lowest 20% of participants think they’re in the 50th or 60th percentile, but the top 20% think they’re in the 80th percentile. It just turns out they’re actually even higher than that.

So it’s not that skilled people think they’re unskilled. They rate themselves much higher than anyone else does. They just totally overestimate how much everyone else sucks. The gold medalist doesn’t think he’s middle of the pack; he thinks he’s getting bronze.

13

I don’t think those are necessarily contradictory. The top performers are “humble” in the sense that they assume that they rank lower within their own elite cohort than they actually do.

However, they are more accurate than the lower cohorts when it comes to the broad estimation of their achievement level. Which we would expect, given that the top-rankers are more knowledgeable about the subject.

So the top performers are indeed “humbly” under-rating themselves, but in the context of a far more specific and better-informed comparison.

14

But you’re changing the claim. Nobody runs around the Internet saying that skilled people are full of themselves, but not as full as they should be. They say smart people think they’re dumb. They say things like “A wise man knows he knows nothing.” In reality, they don’t. According to the study, a wise man knows he knows a lot more than most everyone else. He just thinks surely some of these people are as wise as him. But nope, there’s only like one other master there and a couple of experts. The rest of the people are light years behind him.

It annoys me because saying otherwise completely undermines the point of the study. The big takeaway is that when you know something really well, chances are nobody else does. When you know something, you tend to think (erroneously) that everyone else knows it too, and when you don’t, you think nobody else knows much more about it than you do.

If we start saying the popular version, that “dumb people think they’re smart, smart people think they’re dumb,” we get the exact opposite of the real findings. Dumb people know they’re dumb, the study says. They just think everyone else is dumb too. Skilled people know they’re skilled ; they just think others are keeping pace, when actually they’re winning the race by a mile.

15

I disagree with your interpretation. Although I will note that it’s unusual how Dunning and Kruger avoided using the word intelligence throughout their paper. They use terms like competence and logical reasoning and metacognitive ability. I hadn’t noticed this until you pointed it out. But I feel that in an online discussion like this, I’m allowed to use the term intelligence as a shorthand for metacognitive ability.

That said I don’t agree with your interpretation of what they say about domains of knowledge. When they tested people on whether they thought a bunch of jokes were funny and then asked them how well they did on the test, they weren’t seeing if these people had a well developed sense of humor. The test results told them that. What they wanted to see was if the people taking the test were able to assess themselves and determine whether or not they had a sense of humor.

They then gave similar tests in two other areas, logical reasoning and grammar. To me this shows that they were making an effort to be domain neutral. They wanted to make sure their results weren’t just relevant in one domain, like humor, but existed across domains. And they found it did. They found that regardless of the specific domain of knowledge, people who lacked competence in metacognitive ability - ie people that weren’t intelligent - displayed the same effect over significantly overestimating their performance. These people did not just have poor judgement about their humor or their grammar or their logical reasoning - they had poor judgement in general.

They also noted that the people who got the best results underestimated their performance. But they did see this as the opposite end of the same effect. (“We felt that this miscalibration had a different source then the miscalibration evidenced by bottom-quartile participants.”) They noted that the intelligent did assess their own performance as being above average - they just failed to assess their performance as well above average. And when these people were given more information, they changed their assessment and placed themselves higher. Which the unintelligent people did not; they did not significantly change their assessments even when presented with more information.

I feel this is in keeping with what I said above. The intelligent people used evidence and reason to assess their own abilities. But when they were asked were they ranked in the overall group, the only specific evidence they had was their own performance on the test. Based on the knowledge that they had done well, they reasoned that other people in the group had done well and ranked their performance closer to the average than was true. (“Top-quartile participants did not underestimate themselves because they were wrong about their own performances, but rather because they were wrong about the performances of their peers.”) But when they were given additional evidence they revised their assessment and came up with a more accurate answer.

The unintelligent people had the same initial evidence - their own performance - but produced a worse conclusion. And more pointedly, in my opinion, when they were given more evidence, they did not significantly revise their previous assessment.

("Bottom-quartile participants failed to gain insight into their own performance after seeing the more competent choices of their peers…With top-quartile participants, a completely different picture emerged…Armed with the ability to assess competence and incompetence in others, participants in the top quartile realized that the performances of the five individuals they evaluated (and thus their peers in general) were inferior to their own. As a consequence, top-quartile participants became better calibrated with respect to their percentile ranking.)

This to me shows a fundamental divide. The first group produced the initial assessment based on evidence. And when they received additional evidence, they changed their initial response and produced a better one. The second group did not see any need to change their second response based on new evidence. And my conclusion is that meant they hadn’t used evidence to form their first response.

16

I believe it’s a dynamic that’s just part and parcel of the Dunning-Kruger effect.

“The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt.” So says Bertrand Russell.

17

theres also the effect of people who are as smart as a box of rocks but are told there smart simply because everyone around them are dumber than said box … so they get told “your so smart” when they just arent outside of their own bubble …

18

I’m not getting into a rabbit hole of Dunning Kruger definitions, but I see the effect that the OP talks about a lot. It is especially prevalent with engineers and medical doctors. It is the attitude of “I know engineering and how things work, so I know how everything works.” Or doctors think that since they are doctors, they know everything. The old joke about neurosurgeons: what’s the difference between God and a neurosurgeon? God doesn’t think he is a neurosurgeon.

19

Indeed, ISTM that does look like what is closest to the phenomenon described in the OP.
Someone mentioned “impostor syndrome”, and that is what would look, to me, as more of an “inverse” of Dunning-Kruger in the sense of not believing (or not knowing) you have the knowledge and talent you do have.

20

I’m having trouble understanding your arguments here. How do you jump from this to “the ones with a sense of humor” are more intelligent? Or if that is not what you mean, what is it you mean?

The three experiments consisted of a test meant to assess ability within the domain, and then a self-assessment. They are separate experiments, but your argument seems to require assuming a correlation in skill level between the experiments, or a different x-axis completely.

Am I misinterpreting you, or DK, or are you relying on evidence outside DK to justify assuming this correlation?