Announcement

Collapse
No announcement yet.

Dunning-Kruger effect

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Dunning-Kruger effect

    http://en.wikipedia.org/wiki/Dunning-Kruger_effect

    I was reading a Slashdot story about a tiff among the Linux kernel developers and this article was linked in the discussion. Kind of an interesting characterization of the attitudes of the "competent" and the "incompetent".
    Dude, seriously, WHAT handkerchief?

    snooggums' density principal: "The more dense a population, the more dense a population."

    Iliana: "You're a great friend but if we're ever chased by zombies I'm tripping you."

  • #2
    Re: Dunning-Kruger effect

    Sometimes this is a good thing. How many times have you heard the following.

    Q: "How did you do that? The experts thought it was impossible."
    A: "We where to ignorant to know that it couldn't be done."
    Last edited by El_Gringo_Grande; 07-30-2009, 10:52 AM.
    Iím not racists, I have republican friends. Radio show host.
    - "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
    - "A foolish consistency is the hobgoblin of little minds" - Emerson
    - "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
    - If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette
    - "Let me now state what seems to me the decisive objection to any conservatism which deserves to be called such. It is that by its very nature it cannot offer an alternative to the direction in which we are moving." -Friedrich Hayek
    - "Don't waist your time on me your already the voice inside my head." Blink 182 to my wife

    Comment


    • #3
      Re: Dunning-Kruger effect

      I'm not sure I've ever heard that before. Where have you heard it so many times? Primary sources please - none of this I heard it from guy who heard it.

      Comment


      • #4
        Re: Dunning-Kruger effect

        I think I saw it on Star Trek once. :p
        Dude, seriously, WHAT handkerchief?

        snooggums' density principal: "The more dense a population, the more dense a population."

        Iliana: "You're a great friend but if we're ever chased by zombies I'm tripping you."

        Comment


        • #5
          Re: Dunning-Kruger effect

          Originally posted by sordavie View Post
          I'm not sure I've ever heard that before. Where have you heard it so many times? Primary sources please - none of this I heard it from guy who heard it.
          Really? You have never heard or experienced that before? You can't be serious.

          First what I posted was obviously anecdotal. How can I have a "primary source" for anecdotal evidence?

          It happens to me often. (How is that for a primary source.) Ten years developing software makes me an "expert" by most standards.

          In my early days I would do something and then later find out most other experts would think what I did was unique or would never had considered even attempting it. For example way back in the day I was using the xmlhttprequest to dynamically update not only web pages but also applications. This was not generally thought of as possible until google came out with Google maps. After that AJAX became all the rage and was considered as brand new way of doing something by veteran, expert web developers.

          Today I will be working with younger programmers and they will do something that I find novel. Or they will accomplish something that I thought would require to much work or would not be able to perform well enough.

          One of the problems that comes with expertise is that it can blind you or prevent you from attempting something. Even the linked to article suggests this. Those with more skill have lower confidence in their ability. Usually it is justified. But at times it is not and could prevent the more skillful to even attempt something. While the less skilled may be overconfident and that leads them to try and do what an expert would reject out of hand.

          You have certainly read about this type of thing before? Take plate tectonics. Most of the experts of the time thought the idea was crazy. They rejected the idea out of hand even though a child could see that the continents fit together like a puzzle. Those experts couldn't even see the obvious.

          There was a good NOVA special about it. And you see the same thing in all kinds of fields and professions.

          I am not saying that all experts will not be able to overcome their expertise but many, many will not.
          Last edited by El_Gringo_Grande; 07-30-2009, 10:54 AM.
          Iím not racists, I have republican friends. Radio show host.
          - "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
          - "A foolish consistency is the hobgoblin of little minds" - Emerson
          - "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
          - If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette
          - "Let me now state what seems to me the decisive objection to any conservatism which deserves to be called such. It is that by its very nature it cannot offer an alternative to the direction in which we are moving." -Friedrich Hayek
          - "Don't waist your time on me your already the voice inside my head." Blink 182 to my wife

          Comment


          • #6
            Re: Dunning-Kruger effect

            I'm not sure I understand your situation. Are you telling me the younger programmers who do something you find novel are incompetent? Are you saying that you were incompetent in your early days? Should the term 'experts' in your post be surrounded by scare quotes, because they refer to people who the incompetent think are experts but aren't really experts?

            I don't think expert geologists or whatever back in the day rejected plate tectonics out of hand. Based on the evidence they had, they thought it wasn't true. And it's not like kids then knew that plate tectonics was correct.

            The problem with the statement is that the answerer in your dialogue implicates a contradiction.

            Q: "How did you do that? All the experts thought it was impossible."
            A: "We where to ignorant to know that it couldn't be done."
            Assume that A did X, which Q and the experts thought was impossible. A goes on to state that he was too ignorant to know that X was impossible. But 'know' is a success term like 'found'. If you know that P, then it follows that P is true. If P turns out to be false, then you didn't know P after all. To say that one is too ignorant to know that S is impossible is to imply that one didn't know that X is impossible even though it's true that X is impossible. Of course, since A did X, X couldn't have been impossible after all. So it couldn't be possible to know that X was impossible. So A couldn't be too ignorant to know that X was impossible.

            Still, I suppose I understand what A was trying to say. But being ignorant of where the current field stands doesn't imply that one is incompetent in the field. There are many people who are ignorant of the current status in the field, who go on the discover many things that established experts in the field marvel at. The famous Indian mathematician Srinivasa Ramanujan is a good example. I've never heard of such people being incompetent. The Dunning-Kruger effect is about incompetence not ignorance. One can be ignorant but not incompetent.

            Comment


            • #7
              Re: Dunning-Kruger effect

              http://www.apa.org/journals/features/psp7761121.pdf
              http://www.psy.fsu.edu/~ehrlinger/Se...et_al_2008.pdf

              I've also encountered this in the workplace. It seems to be somewhat common in software development ;)



              TacticalGamer TX LAN/BBQ Veteran

              Comment


              • #8
                Re: Dunning-Kruger effect

                Good meta-cognition has it's limits and presents problems unique to it. It can be limiting. Over confidence, even if misplaced can be a good thing.

                That's all I'm sayin'.

                *Edit Just to be clear, it is normally a bad thing. But sometimes it is a good thing.
                Last edited by El_Gringo_Grande; 07-30-2009, 02:14 PM.
                Iím not racists, I have republican friends. Radio show host.
                - "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
                - "A foolish consistency is the hobgoblin of little minds" - Emerson
                - "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
                - If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette
                - "Let me now state what seems to me the decisive objection to any conservatism which deserves to be called such. It is that by its very nature it cannot offer an alternative to the direction in which we are moving." -Friedrich Hayek
                - "Don't waist your time on me your already the voice inside my head." Blink 182 to my wife

                Comment


                • #9
                  Re: Dunning-Kruger effect

                  Originally posted by El_Gringo_Grande View Post
                  Over confidence, even if misplaced can be a good thing.
                  Do you work at the White-House by any chance?
                  sigpic




                  Comment


                  • #10
                    Re: Dunning-Kruger effect

                    Originally posted by INSUNABULA View Post
                    Do you work at the White-House by any chance?
                    He's probably the Imperial Beer Taster. :p
                    Dude, seriously, WHAT handkerchief?

                    snooggums' density principal: "The more dense a population, the more dense a population."

                    Iliana: "You're a great friend but if we're ever chased by zombies I'm tripping you."

                    Comment


                    • #11
                      Re: Dunning-Kruger effect

                      Originally posted by El_Gringo_Grande View Post
                      Good meta-cognition has it's limits and presents problems unique to it. It can be limiting. Over confidence, even if misplaced can be a good thing.

                      That's all I'm sayin'.

                      *Edit Just to be clear, it is normally a bad thing. But sometimes it is a good thing.
                      I don't think it's ever inherently a good thing. Having an attribute A and accidentally doing something good doesn't entail that having A is a good thing. I take it that's what happens when incompetent people sometimes accidentally stumble on the correct solutions or answers. The article isn't talking about mere overconfidence. It's a certain kind of overconfidence that's given rise to by profound incompetence.

                      Good meta-cognition shouldn't be limiting by definition. If your meta-cognition doesn't allow you to see that it's somehow limiting your cognition, then it's not very good. Perhaps your concept of meta-cognition is inaccurate. A good meta-cognition applies to itself. That must be part of the concept of meta-cognition or else you result in an infinite regress or no ability to distinguish people who can gauge their own ability accurately from those who can't (except at a certain meta-level, but that doesn't help much if there are an infinity of meta-levels). Meta-cognition presumably is the ability to monitor accurately one's cognitive abilities. But since meta-cognition is a cognitive ability itself, it must be included as part of what's monitored.

                      Comment


                      • #12
                        Re: Dunning-Kruger effect

                        I am talking about real human beings that are experts, not mythical intellectual perfection.

                        I am reading competence to be much like expertise*. It isnít necessarily what you are capable of. It is what you can accomplish at that particular point in time. Just because you fail a test today doesnít mean you will fail the same test tomorrow. The fact that those that fail a test are over confident of their ability to pass the test does not detract from what they are actually capable of.

                        Maybe you think that is what meta-cognition is? Understanding what you can accomplish not only now, but also in the future? If it is I believe you are wrong. Meta-cognition is just knowledge about your knowledge up till that point in time. It isnít about what your knowledge or ability will be in the future.

                        And meta cognition is a two edged sword. As you said if it canít be applied to itself then it isnít very good. I have never read anywhere that it does apply to itself. I would guess that any individuals could have, at most, 5-7 levels of it because that seems to be the limit of short term memory. It is, I think, a short term memory phenomenon. You might be able to recall that in the past your knowledge level was x. And really I think it is a summation of long term memory instances of meta cognition awareness.

                        So in the present your reaction to your meta cognition will depend on two things. Whether you are an optimist or a pessimist and how often your meta cognition has been correct. (Donít forget that the subjects of the study where college students. Most would be optimists to begin with and that is a confounding factor. Add to that the fact most were probably young. What kind of meta cognition do most young people have?)

                        If there is no meta cognition due to age or other factors I would say the optimist would try things and the pessimist would not. The pessimist that had high potential would fail to achieve but the pessimist that had low potential would not exert undo effort. The optimist with high potential would achieve but the optimist with low potential would exert undo effort for nothing.

                        The tricky part is knowing who has high potential. How do you know what the actual potential is? Many studies have shown that one key factor to success is the ability to delay gratification. This is not an always innate trait and can be a skill that is taught. Delaying gratification is relevant in this conversation because those that believe they can succeed at something will work longer at the task with need of immediate success.

                        So sordavie, it is a complex issue. I could go on and on about this. There are other factors like external support and encouragement. The individualís ability to deal with, and even excel in the face of, failure. The contentment one gets from success can ruin future possible accomplishments. It isnít just about skill and ability at a particular time. It plays a part but is only one part of the puzzle.

                        ==================
                        *As I was taught an expert is generally considered somebody that has had about 4-5 years of experience within a field and is considered competent by his/her peers.

                        Also I am referring to experts as a group. I know there are exceptional individuals within any group and they are a special case and may not act as I describe below.

                        Experts rely heavily on heuristics. They do not think through the problems presented. They often use these "broken leg cues" to come to a conclusion about a situation. There is no deep logic used. They don't analyze the problem at hand in a step by step way. For experts often analytical skill has become secondary and is not often used. Not only has their skill become rusty, their expertise can actually prevent them from using any problem solving ability they have.

                        It is why they are often wrong in such profound ways and can screw up really basic things. When they do actually try to understand the problem they can be constricted by what they have memorized. In fact it has been found that experts often donít perform any better than non-experts in certain domains.

                        But it is also why they are usually right and can come to the correct solution so fast. In life novelty is actually rare. New problems donít often crop up. In almost all cases the problem is understood well enough and a solution has already been created.

                        Pre-experts, like medical interns, are almost always much better at actually analyzing the problem at hand using their skill/intelligence and detailed knowledge. That is one reason non-experts are slower at solving problems; they must logically work through the problem. It is also why these pre-experts are much better teachers because they still have to explain to themselves why they are reaching a certain conclusion. Experts don't always know why they came to the conclusion they did.

                        It must be noted that humans are not very good at problem solving using logic etc. So it is understandable why non-experts are wrong more often than the experts on common problems. Not because they are less skilled but because they canít use the long term memory as efficiently (because of heuristic use).

                        Another example is chess grandmasters. When they analyze a board they don't really use short term memory to try and figure out their next move. They usually have a massive set of board arrangements stored in long term memory. They look at the board as is and work through what is possible from the current configuration from the stored arrangements. They ďrememberĒ that from the current configuration there is a set of moves that will lead to victory. They do not move one piece then figure out what pieces the opponent can or will move then figure out the next move etc.

                        So being a non-expert does not mean you canít eventually do whatever an expert can. It usually just means it will take longer and you will be wrong more often if the problem presented is known within the domain. Especially with novel problems they can do as well as the experts if they have just a bit of intelligence and perseverance. I think they actually can have an advantage over the expert because they must logically work through the problem. They have to break it down and analyze it for understanding. During this process they may discover something or find a better solution. They may even take on a problem that an expert would reject out of hand as unimportant or consider ďsolvedĒ.

                        Consider also that doing something correctly the second time means you learn nothing. Failure and the first success is the only time learning really takes place. Everything else is just re-enforcement. (Also there is pattern recognition that can really help, and screw over, the expert.)

                        So when you said ďI don't think expert geologists or whatever back in the day rejected plate tectonics out of hand. Based on the evidence they had, they thought it wasn't true. And it's not like kids then knew that plate tectonics was correct.Ē I donít think you really understood what was going on.

                        The fact is there was evidence that the earth moved around in an organized way. There was obvious evidence that what they thought was wrong. The map was evidence. The fact that all the continents fit together like a big puzzle was, and is, overwhelming evidence. But it was ignored. Why? Because, in part, their expertise they had blinded them to the obvious. They had already learned why the earth looked the way it did. Some simply where not capable of learning anything else because they overvalued the knowledge of what they knew. They had a set of long term memory structures and heuristics to manage them in place that answered certain questions with ease. There were other problems that they knew they didnít know and focused on them.

                        I am not saying experts are irrelevant. They are very important and contribute much. I think we all should strive to become one in our respective field. But they are not everything we need and the overconfident novice should be respected.

                        If you want to actually understand the problem better, look up Ericsson and expertise. He was one of the main authors I remember when I studied this 10 or so years ago and that will get you started.
                        Iím not racists, I have republican friends. Radio show host.
                        - "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
                        - "A foolish consistency is the hobgoblin of little minds" - Emerson
                        - "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
                        - If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette
                        - "Let me now state what seems to me the decisive objection to any conservatism which deserves to be called such. It is that by its very nature it cannot offer an alternative to the direction in which we are moving." -Friedrich Hayek
                        - "Don't waist your time on me your already the voice inside my head." Blink 182 to my wife

                        Comment


                        • #13
                          Re: Dunning-Kruger effect

                          Did you even read the article?

                          You appear to be talking about competent novices who may be ignorant of the latest developments in their field or who haven't earned the reputation of expert yet. That's not what Dunning and Kruger were studying at all. Their conclusions are about seriously incompetent people.

                          The chess example is crazy. There's no GM who's ever going to lose to a seriously incompetent chess player. You appear to be thinking of some novice chess player who doesn't know his opening theory, doesn't know end game patterns, but who can calculate much better than the strongest chess computers. (By the way, our strongest chess computers, while they can destroy all but the strongest human GMs, they'd get whipped without their opening books or endgame tables by even IMs.) In any case, such a novice would certainly not be considered seriously incompetent at chess. Someone who's seriously incompetent at chess would not be able to anticipate their opponents moves, would not be able to recognize impending forks, doesn't know what a skewer or pin is, has no clue of opening theory, pawn structure, good/bad bishops, strategy, tactics, mid or endgame. I don't see how such a player would possibly be able to calculate a win from even a moderately competent novice, much less an expert in chess. This is the kind of person that Dunning and Kruger are talking about - the seriously incompetent. Their findings are that seriously incompetent people overestimate their abilities far more relative to competent people. You're whole discussion on long term working memory and short term memory aren't on topic.

                          I gave you an argument that there are no meta meta levels of metacognition, that it must collapse to just the second level. Indeed, that's how Dunning and Kruger view it. Actually, come close to arguing that it all collapses to the first level in the last paragraph of page 1121.

                          J. H. Flavell's - the first to use the word - definition of 'metacognition' implies that there is only one meta-level, by the argument I gave.

                          " Metacognition refers to one’s knowledge concerning one’s own cognitive processes or anything related to them..."
                          —J. H. Flavell (p 232 in Metacognitive aspects of problem solving. In L. B. Resnick (Ed.), The nature of intelligence (pp.231-236). Hillsdale, NJ: Erlbaum).

                          Since metacognition itself is a cognitive process, metacognition includes one's knowledge concerning one's own metacognitive processes.

                          I'm sure you can understand how this would work. You could easily write a computer program that records its own processes. You can even make it record the process that does the recording. You needn't some second program to do that, and some third program to record the second's recording process, and a third, and so on.
                          Last edited by sordavie; 08-01-2009, 12:53 AM.

                          Comment


                          • #14
                            Re: Dunning-Kruger effect

                            Originally posted by sordavie View Post
                            Did you read the article?
                            You mean the wiki article? Yes.

                            But I am not responding to the article. I don't really have a problem with the article. And I don't have access to the original paper so I am not questioning it either.

                            I am responding to your seeming assertion that over confidence (at any point in time) is always bad and high levels of meta cognition (at any point in time) is always good. I am saying that overconfidence can be beneficial and knowing your lack knowledge can be bad given the right circumstances.

                            Did you read and understand my response? I am saying you are reading way to much into it. It is one set of studies and you must understand all the other studies that relate , and are even tangential, to it.

                            I am saying you are making conclusions based upon your biases.
                            Last edited by El_Gringo_Grande; 08-01-2009, 12:53 AM.
                            Iím not racists, I have republican friends. Radio show host.
                            - "The essence of tyranny is the denial of complexity". -Jacob Burkhardt
                            - "A foolish consistency is the hobgoblin of little minds" - Emerson
                            - "People should not be afraid of it's government, government should be afraid of it's People." - Line from V for Vendetta
                            - If software were as unreliable as economic theory, there wouldn't be a plane made of anything other than paper that could get off the ground. Jim Fawcette
                            - "Let me now state what seems to me the decisive objection to any conservatism which deserves to be called such. It is that by its very nature it cannot offer an alternative to the direction in which we are moving." -Friedrich Hayek
                            - "Don't waist your time on me your already the voice inside my head." Blink 182 to my wife

                            Comment


                            • #15
                              Re: Dunning-Kruger effect

                              No, the Dunning-Kruger article. That's what I'm talking about. That's the title of this thread, after all.

                              You started the whole thing by responding to Scratch's posting of the wiki article to the Dunning-Kruger effect by saying "sometimes this is a good thing." By the word 'this', I took it you refer to the results of their study about how the seriously incompetent overestimate their abilities by much more relative to their competent counterparts. Did you mean to not be anaphorically referencing the subject of Scratch's link and be talking about something else? You were the second to post in this thread. What were you responding to if not the article? It sure looks like you were. If you weren't, how are your readers to know what 'this' refers to, if not the subject of the link in the first post?

                              I only assumed you were talking about the Dunning-Kruger effect, given your second post; and I have been responding under the assumption you've been defending what you said in that second post as it relates to the Dunning-Kruger effect. I'm not sure it's fair to say that I'm reading too much in to your response given what you actually said in this thread.

                              Tau was kind enough to post a link to the original paper as well as a larger follow up study.

                              Comment

                              Connect

                              Collapse

                              TeamSpeak 3 Server

                              Collapse

                              Advertisement

                              Collapse

                              Twitter Feed

                              Collapse

                              Working...
                              X