Thoughts on the Dunning-Kruger effect

[jimbob]

Neanderthals and Cro Magnon? man coexisted for thousands of years without either gaining the advantage.

Man has the most efficient cooling system of any land animal. Many large prey can be fatally overheated by pursuit.

The Neanderthals were stronger, but physically incapable of long pursuit. There bodies required more calories and were adapted to cold weather.

Mentally the Neanderthals may have been comparable, based on brain mass. There brains may have allowed for more precise musculature control or better visual processing. The rear of their brain developed while humans frontal lobes were being developed.

Environmental change appears to have defeated the Neanderthals. Ambush hunters needed forests, while humans could function on the open plains. Europe experienced a long drying out that shrunk the forests. Humans got lucky.

There is uncertain evidence that modern humans brains have shrunk from pre-civilization levels. Wolves have twice the brain size of dogs.

Wolves are a threat.

In Dilbert, brains are a threat to the hair pointed boss. Brains were a threat to kings and tyrants (most of human history was under these conditions). Intelligence was bred out. Idiocracy

Although the whole story is more complicated. Asians rate high on brain size AND they have been under strict civilizations for a longer period of time than many other groups.

Yes total, crushing, and irrefutable defeat can help with humility, if you survive.

And remember that modern humans nearly became extinct about 70k years ago.

An aside. Europeans seem to have some Neanderthal genetic material, and Asians some genetic material from their local equivalent*, should these actually have been considered as separate species?



*cant google at the moment...

 

[JSFolk]

Earlier in the thread someone mentioned that the DK effect should be seen less when people have direct evidence of their actual performance. That reminded me of a passage in "The Invisible Gorilla", which I was able to find mentioned in short in this NY Post article:

http://www.nypost.com/p/news/opinio..._delusions_of_grandeur_kmSEG1YrE1Uhfh1fL4tdWP

"It turns out that the illusion of confidence can survive even the measurement of skill.

Chess, for instance, has a mathematical rating system that provides up-to-date, accurate and precise numerical information about a player’s “strength” (chess jargon for ability) relative to other players. Ratings are public knowledge and are printed next to each player’s name on tournament scoreboards. Ratings are valued so highly that chess players often remember their opponents better by their ratings than by their names or faces. “I beat a 1600” or “I lost to a 2100” are not uncommon things to hear in the hallway outside the playing room.

Armed with knowledge of their own ratings, players ought to be exquisitely aware of how competent they are. But what do they actually think about their own abilities? Some years ago, in a study we conducted with our colleague Daniel Benjamin, we asked a group of chess players at major tournaments two simple questions: “What is your most recent official chess rating?” and “What do you think your rating should be to reflect your true current strength?”

As expected, all of the players knew their actual ratings. Yet 75% of them thought that their rating underestimated their true playing ability. The magnitude of their overconfidence was stunning: On average, these competitive chess players estimated that they would win a match against another player with the exact same rating as their own by a two-to-one margin — a crushing victory. Of course, the most likely outcome of such a match would be a tie."

 

[MRC_Hans]

Your confirmation bias added "letting just any fool in" which was not in what I said.

No, that was what I said. You said "Don't close the door", I said "I don't but I won't let just any fool in"

ETA: I'm not talking about you, or me for that matter. I'm talking about those fools who complain that we close the door to the patent office because we turn their silly ideas down.

Hans

 

[MRC_Hans]

Earlier in the thread someone mentioned that the DK effect should be seen less when people have direct evidence of their actual performance. That reminded me of a passage in "The Invisible Gorilla", which I was able to find mentioned in short in this NY Post article:

http://www.nypost.com/p/news/opinio..._delusions_of_grandeur_kmSEG1YrE1Uhfh1fL4tdWP

"It turns out that the illusion of confidence can survive even the measurement of skill.

Chess, for instance, has a mathematical rating system that provides up-to-date, accurate and precise numerical information about a player’s “strength” (chess jargon for ability) relative to other players. Ratings are public knowledge and are printed next to each player’s name on tournament scoreboards. Ratings are valued so highly that chess players often remember their opponents better by their ratings than by their names or faces. “I beat a 1600” or “I lost to a 2100” are not uncommon things to hear in the hallway outside the playing room.

Armed with knowledge of their own ratings, players ought to be exquisitely aware of how competent they are. But what do they actually think about their own abilities? Some years ago, in a study we conducted with our colleague Daniel Benjamin, we asked a group of chess players at major tournaments two simple questions: “What is your most recent official chess rating?” and “What do you think your rating should be to reflect your true current strength?”

As expected, all of the players knew their actual ratings. Yet 75% of them thought that their rating underestimated their true playing ability. The magnitude of their overconfidence was stunning: On average, these competitive chess players estimated that they would win a match against another player with the exact same rating as their own by a two-to-one margin — a crushing victory. Of course, the most likely outcome of such a match would be a tie."

I think it is a human trait to overestimate our own abilites. Probably a beneficial trait, because even if it makes us embark on adventures we are not really qualified for, we also have the ability to learn as we go. So, on the whole, overestimating ourselves gets us further.

You can see it every week-end at the DIY shop. Just look for those with the largest heap of boards and stuff on their cart. Those are the "It cant be too difficult to build a carport/put up a division wall/lay that floor" people. Many of them really can't, but, after they're done, most of them can. The Dunning Krugers among them are those who cannot see that their carport will fall flat in high wind or that their new floor is warped.

Hans

 

[Skeptic Ginger]

No, that was what I said. You said "Don't close the door", I said "I don't but I won't let just any fool in"

ETA: I'm not talking about you, or me for that matter. I'm talking about those fools who complain that we close the door to the patent office because we turn their silly ideas down.

Hans

But to assume the game is unwinnable is like saying we've invented all there ever is to invent, might as well close the patent office.

Gaah! That was the favorite argument of another incorrigible DK sufferer I spent too much time with! Essentially: If you won't accept my unfounded claims you are closing the door to the patent office. Yahh, come back wnen you can demonstrate your perpetuum mobile to me.

Hans

It does sound like you are saying my argument, "might as well close the door", is the same as the "favorite argument of another incorrigible DK sufferer."

I shall correct my interpretation according to this new information.

 

[Skeptic Ginger]

I wonder if the D-K effect applies to my continual false belief that I have enough time to get to an appointment on time or to get one more thing done in the time I have?

 

[MRC_Hans]

It does sound like you are saying my argument, "might as well close the door", is the same as the "favorite argument of another incorrigible DK sufferer."

I shall correct my interpretation according to this new information.

It gave ME that, purely subjective, association. But I think I just noted that I wasn't talking about you. You happened to press a button you could not know was there. OK?

Hans

 

[MRC_Hans]

I wonder if the D-K effect applies to my continual false belief that I have enough time to get to an appointment on time or to get one more thing done in the time I have?

Since you acknowledge that it is false, I don't think so. Just normal human tendency to overestimate your ability. As I said, I consider that a trait that probably got us out of the jungle, for better and worse.

Hans

 

[pgwenthold]

As expected, all of the players knew their actual ratings. Yet 75% of them thought that their rating underestimated their true playing ability. The magnitude of their overconfidence was stunning: On average, these competitive chess players estimated that they would win a match against another player with the exact same rating as their own by a two-to-one margin — a crushing victory. Of course, the most likely outcome of such a match would be a tie.[/I]"

To be picky, this doesn't follow at all. Just because two players are perfectly equal in ability, does not mean the most likely outcome is a draw. It means that the outcomes of matches that are won will be equally distributed to the two players. It does not tell us what the distribution of draws vs a winner will be.

Now, it is true that two equal competitors are more likely to reach a stalemate, and it could absolutely be the case that as you reach higher levels, the probability of reaching a draw increases, such that the combination between the two does indeed lead to the majority of matches being a draw, but it is not inherent to the situation. For example, my wife and I are probably pretty close in ability in chess, and we both stink. So if we play each other, we don't end up in draws, someone ends up winning (because the other person messed up pretty bad, probably). Who wins is basically a 50/50 matter, so that means we are still equal in ability (according to an ELO/Chess interpretation)

Heads and tails of a coin are each just as likely to result on a coin flip; that doesn't mean the coin will land on the edge

 

[Perpetual Student]

I wonder to what degree the D-K effect is a function of age. I have noticed that as I have gotten older, I have become quite aware of superior performance, knowledge and abilities in others -- where I might have been more reluctant to do so at a younger age. At 72, I am also aware of the decline in my abilities. Perhaps facing that reality has enabled be to be more realistic about where I stand now and where I actually stood when I was younger. Is it only me, or is there some universality to this change of perspective as we age?

 

[MRC_Hans]

I wonder to what degree the D-K effect is a function of age. I have noticed that as I have gotten older, I have become quite aware of superior performance, knowledge and abilities in others -- where I might have been more reluctant to do so at a younger age. At 72, I am also aware of the decline in my abilities. Perhaps facing that reality has enabled be to be more realistic about where I stand now and where I actually stood when I was younger. Is it only me, or is there some universality to this change of perspective as we age?

I can certainly recognize it. It makes a lot of sense: Young people barge ahead, confident that they either know how to do things or can find out as they go. Old people are cautious, move slower, aware of their limitations.

Hans

 

[zooterkin]

I've just remembered what could be another example of the Dunning Kruger effect. You often see it in reality TV shows such as Strictly Come Dancing/Dancing with the Stars. There are generally three types of people competing; actors, sportsmen and miscellaneous celebrities. The celebrities tend to be the least able to take criticism, which is usually constructive (though not always tactfully framed), being unable to separate the amount of effort they've put in from the results they produce, and often disagreeing with the judges as to how well they've performed. The actors and sportsmen seem to be more used to taking outside opinions and pointers for improvements and more prepared to accept someone else's view of how they did.

 

[MRC_Hans]

You can also see something like it in competent people, althoug I suspect it is a different mechanism. Those are the clever, sometimes near geniouses, who end up believing that they are infallible, sometimes disappearing off into woo.

Hans

 

[Alex Cured]

Recently, someone linked to info on the URL-REDACTEDDunning–Kruger effect

I hadn't seen a formal description before, but I think we see this a lot, and it gave me a new understanding of many posters, here and elsewhere.

In my view, a synopsis of the D-K effect is:

A person has an unrealistic estimation of own competency on some subject, which is caused by ignorance(or lack of ability) of that subject.

Hans

Interesting. This seems closely related to "Illusory superiority", the Wikipedia article on the subject mentions :

In a survey of faculty at the University of Nebraska, 68% rated themselves in the top 25% for teaching ability.

Illusory superiority: The Dunning–Kruger Effect for people who really ought to know better?

 

[Alex Cured]

oops .. the wikipedia page on Dunning-Kruger links to this very page

 

[Dr. Imago]

Skeptic Ginger:

Check me if I'm wrong...

I hear in your counterargument that people, while they may not possess similar intelligence quotients (i.e., measure of rapidity of ability to obtain, retain, assimilate, and apply new knowledge), they do have an inherently equal aptitude. It just takes time to develop it.

If that's the case, I have to disagree with you. I am convinced, having spent 43 years on this planet, that most people differ vastly in their aptitude, with differing abilities in differing areas being more developed or less developed in a given individual.

For example, I have a terrible time with naming and recalling the names of things. By the very nature that I understand this deficiency, I don't suffer from the D-K effect in that area. If I believed, OTOH, that I was great at recalling and recollecting names of things, people, places (etc.), the D-K case (in that instance) would be proved in me. However, I'm otherwise a highly competent individual in other areas of my intellectual performance.

I don't believe any amount of time you could spend with me trying to overcome this would improve this area of my mental performance. The difference between me and a D-K sufferer? I recognize it.

~Dr. Imago

 

[Dr. Imago]

Let me employ another example of what I'm talking about...

At my sister's wedding, I work a very dark navy suit. My aunt was there and, when she saw me at the ceremony, she came up and said, "That's a really sharp suit. I love a black suit on a man."

I told her that it was navy, but thank you. She insisted, no, that I was wrong. It was black. My mother, standing next to us, agreed it was navy. So, did my father. She continued to persist in her stance that it was black. We asked a few other attendees. Navy. A couple more. Navy.

It actually became a running joke even at the reception. She insisted that we were all wrong and needed to collectively get our eyes checked.

Now, if you had quizzed her - without any prompting - to answer whether or not I had a black or navy suit on, she would've have said "black". If you had asked her how certain she was, she would have said 100%.

The D-K effect is a problem of perception in the individual. Sometimes you just can't reason with misconception, whatever its etiology.

~Dr. Imago

 

[dlorde]

... For example, I have a terrible time with naming and recalling the names of things. By the very nature that I understand this deficiency, I don't suffer from the D-K effect in that area. If I believed, OTOH, that I was great at recalling and recollecting names of things, people, places (etc.), the D-K case (in that instance) would be proved in me. However, I'm otherwise a highly competent individual in other areas of my intellectual performance.

I don't believe any amount of time you could spend with me trying to overcome this would improve this area of my mental performance. The difference between me and a D-K sufferer? I recognize it.

On the other hand... I also have a terrible time with naming and recalling the names of things, and almost everyone else I know says the same thing - they all think they have it worse than most other people... [see where I'm going?]

Perhaps our ignorance of the general lack of faith people have in their naming and recall abilities causes us to feel that our own performance is particularly poor

 

[Skeptic Ginger]

Skeptic Ginger:

Check me if I'm wrong...

I hear in your counterargument that people, while they may not possess similar intelligence quotients (i.e., measure of rapidity of ability to obtain, retain, assimilate, and apply new knowledge), they do have an inherently equal aptitude. It just takes time to develop it.

If that's the case, I have to disagree with you. I am convinced, having spent 43 years on this planet, that most people differ vastly in their aptitude, with differing abilities in differing areas being more developed or less developed in a given individual.

For example, I have a terrible time with naming and recalling the names of things. By the very nature that I understand this deficiency, I don't suffer from the D-K effect in that area. If I believed, OTOH, that I was great at recalling and recollecting names of things, people, places (etc.), the D-K case (in that instance) would be proved in me. However, I'm otherwise a highly competent individual in other areas of my intellectual performance.

I don't believe any amount of time you could spend with me trying to overcome this would improve this area of my mental performance. The difference between me and a D-K sufferer? I recognize it.

~Dr. Imago

How would you get the impression I thought everyone equal in mental aptitude?

My premise is we need to discover new approaches to reaching people and assuming that because we did our best to impart knowledge and were unsuccessful means the task is impossible is wrong. It's new approaches we need, not longer sessions of the same approach.

And I'm an optimist.


As for the name thing, I can relate to that. I can see people's faces in my memory very clearly and can't for the life of me recall their names. New approach*: in the future we might have face recognition software in our computer equipped glasses that will call up the person's name and display it on our retina so we no longer need to remember names.

*Michio Kaku, "Physics of the Future"

 

[JSFolk]

It would be MUCH easier for your Augmented Reality glasses to broadcast a unique ID via Bluetooth, which other glasses then would use to identify you.

But what do I know, I'm just an AutoCAD monkey.