Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
Science Entertainment Games Technology

League of Legends Rank Predicts IQ, Study Finds (plos.org) 85

limbicsystem writes: A new publication in the journal PLOS ONE shows that your rank in League of Legends (LoL) correlates with your intelligence quotient (IQ). Games like LoL and DOTA II apparently depend on the same cognitive resources that underlie tests of fluid intelligence. That means that proficiency in those games peaks at the same age as raw IQ -- about 25 -- while scores in more reaction-time based games like Destiny or Battlefield seem to decline from the teens onwards. The researchers suggest that the massive datasets from these online games could be used to assess population-level cognitive health in real-time across the globe. The authors have a nice FAQ (and open datasets) here.
This discussion has been archived. No new comments can be posted.

League of Legends Rank Predicts IQ, Study Finds

Comments Filter:
  • by hey! ( 33014 ) on Wednesday November 15, 2017 @08:38PM (#55558957) Homepage Journal

    Unfortunately, one of the things I've learned over the years is how little IQ correlates to anything useful; at least once you get much past 1.5 or 2 standard deviations over the mean.

    • Unfortunately, one of the things I've learned over the years is how little IQ correlates to anything useful; at least once you get much past 1.5 or 2 standard deviations over the mean.

      That's actually because the cutoff for things like IQs actually meaning anything is roughly two standard deviations from the mean--in either direction, not just over. Tests like IQ tests are functionally rulers that are just long enough to go that far on either side of the mean--once you get past that the ends, the actual number is pretty much insignificant.

      Of course, there's problems if you were to just report those scores as something other than numbers, even though the actual numbers don't mean much any

    • What a ridiculous statement. Your assumption of people's intelligence is meaningless. Unless you tested people with a valid IQ test you could not even begin to learn anything about IQ. People who do that consistently find the opposite.
      • by hey! ( 33014 )

        No, my assumption is that IQ tests are relatively meaningless. About the best you can say for them is that they're reasonably self-consistent.

        Here's I think why: intelligence isn't something directly observable, like height or length. It's something we infer. So to measure it we develop indirect tests, but how do we know that those tests measure what we think they should? We compare the results to what we expected to get, for example to they rank subjects approximately the way our intuition ranks them?

    • IQ measures your ability to give a shit about pointless and puzzling minutiae, a characteristic that leads to success in scholarly pursuits.

      I hear that League of Legends rank also correlates strongly with obesity, foul odor, number of burgers flipped, and mean proximity to grandmother's basement.
    • Unfortunately, one of the things I've learned over the years is how little IQ correlates to anything useful; at least once you get much past 1.5 or 2 standard deviations over the mean.

      But over lo these many years you inch farther and farther beyond the mean?

      If the mean was ebbing, you could be just standing still. If the mean was also increasing, you would be getting smarter faster.

      Then we come to the question of "correlate". There are a lot of useful things popping up all the time. Seems to be more stuff and more frequently on top of that, so more times more, or more squared. Does that mean if your increasing intelligence is just bumping up little by little over years and years, the cor

  • by Khopesh ( 112447 ) on Wednesday November 15, 2017 @08:52PM (#55559029) Homepage Journal

    Like any kind of basic test, IQ tests aren't terribly abstractable. Therefore, the supposed correlation between this type of games and IQ tests isn't terribly indicative of intelligence.

    Therefore: if you like IQ tests, you should really try these games.

    This also reminds me of a quote:

    I have no idea [what my IQ is]. People who boast about their IQ are losers. -- Stephen Hawking

    • by Anonymous Coward on Wednesday November 15, 2017 @09:00PM (#55559067)

      It's not true that IQ (more commonly 'g' in professional literature) means nothing, there is extensive psychometric studies about what it means and what is in it and what isn't. Some aspects of it is observable in biological measurements.

      https://source.wustl.edu/2012/07/brain-imaging-can-predict-how-intelligent-you-are-study-finds/

      It means that performance across a wide, but not comprehensive, variety of cognitive tasks is correlated in individuals, i.e. if you are good at some of them, it is highly more likely than chance you will be above average on the others. There are of course many other tasks and demands on human performance which aren't related.

      Nevertheless, it is true that people who boast about their IQ are losers.

      • Or as I recently posted in another /. discussion, the g-factor is kind of like the master clock frequency of your brain. I don't think I'm alone with such computing analogies of cognition; for another example, learning things by heart remains a useful skill in the era of Google etc., due to caching and I/O latency. Memory and I/O analogies also explain issues in large organizations (e.g. SMP vs. clustering vs. globally distributed computing).

        I also don't have much interest in measuring my own IQ. If it t

    • IQ results are proven to predict success in a lot of things. Thinking that IQ does not measure intelligence is a sign of low IQ. You can go ahead and argue that there are other types of intelligence, but that argument does little other than conflate skills with intelligence. IQ measures your ability to solve problems. IQ tests a are reproducible, consistent, and predictive. That so many people write it off as pseudoscience proves only how stupid most people are.
      • by sound+vision ( 884283 ) on Thursday November 16, 2017 @12:05AM (#55559961) Journal
        I think it's pseudoscience on the same level that the MBTI is pseudoscience - not so much in what it does, as what people *think* it does. An MBTI or IQ test very accurately measures... your ability to take an MBTI or IQ test. The basic premise behind the tests - that personality can be reduced to four scores on four linear axes, or problem-solving ability reduced to a score on one linear axis - seem to be gross oversimplifications. Incomplete pictures, vis-a-vis the Platonic ideals you think you are measuring with these tests. Which is not to say that you won't find all sorts of correlations between peoples' MBTI and/or IQ, and how their lives turn out. It *is* to say that those correlations don't mean what you think they mean. An IQ test measures very specific types of problem solving, taken in a bubble, in a manner that almost no real-world problems are presented to a person.

        FWIW, since you wanted to make this personal, I assume I did quite well on the IQ tests I was given in school. The result was always that they wanted to put me in GT or AP courses. But I am lacking in the sort of narcissism that would prompt me to go out of my way to take an IQ test at age 28, or join Mensa, or make a post like yours on Slashdot.
        • An MBTI or IQ test very accurately measures... your ability to take an MBTI or IQ test.

          ...I didn't know that an MBTI test required skill. What does being better at an MBTI test mean?

          I thought it was just a way to pigeonhole your personality into sixteen categories...

          (ISTP myself...)

          • The MBTI doesn't require skill - of the similarities I noted, that wasn't one of them. The comparison I was trying to make was that both of these tests have their limitations and paint an incomplete picture of what they are trying to measure. And that a lot of people, from employers to Slashdot posters, would do well to keep that in mind.
    • by Gussington ( 4512999 ) on Thursday November 16, 2017 @01:18AM (#55560205)

      Like any kind of basic test, IQ tests aren't terribly abstractable.

      What does that even mean? The IQ tests I've done were abstract, and by that I mean the tests where about pattern matching so required no prior knowledge, and weren't specific to any cultural or social standard. eg series of shapes guess the next one, rotating shapes to find match etc. That is about as abstract as a test can get.

      Therefore, the supposed correlation between this type of games and IQ tests isn't terribly indicative of intelligence.

      I've only played a little LoL, but it is effectively abstract too. You have a pool of characters to choose from with special powers which compliment or contradict each other (think paper/scissors/rock on steroids). The people who do well are able to process the various combinations more quickly than others, which is effectively is the same as an IQ test.

      People who boast about their IQ are losers

      Not sure how this is relevant. TFA is merely pointing out a connection being some types of games which are very similar to some IQ tests, therefore have corresponding results.

      • I've only played a little LoL, but it is effectively abstract too. You have a pool of characters to choose from with special powers which compliment or contradict each other (think paper/scissors/rock on steroids). The people who do well are able to process the various combinations more quickly than others, which is effectively is the same as an IQ test.

        Or they care more about the game and went through the pains to learn nooks and crannies.
        Or they played a LOT and managed to become proficient through repetition.
        Or they learned some cookie cutter strategies by heart, effectively knowing nothing about anything else.

        If those stats are true, then look no further for proto-Einsteins, just pick the LoL world champions and call it a day.

        • I've only played a little LoL, but it is effectively abstract too. You have a pool of characters to choose from with special powers which compliment or contradict each other (think paper/scissors/rock on steroids). The people who do well are able to process the various combinations more quickly than others, which is effectively is the same as an IQ test.

          Or, Or, Or, If...

          Of course, you could spend all day making wild guesses about it, or you could actually do a study and document the process you used and the results you found. You know, exactly like TFA has...

  • by Barny ( 103770 )

    So that an AI beat a top world ranked player in 1v1 means that AI has an incredible IQ?

    • It may be possible to write an AI that does really well on IQ tests. That would not invalidate the validity of the test on humans nor prove that the AI is more intelligent than a human. IQ tests predict success in humans, they may not in AI.
  • by Tablizer ( 95088 ) on Wednesday November 15, 2017 @09:08PM (#55559113) Journal

    I smell a marketing gimmick: play-our-game-to-feel-smart

    • by Xyrus ( 755017 )

      Which is stupid as IQ has almost nothing to do with being smart. It's at best a possible measure of potential. High IQ people aren't any more or less prone to being idiots as the rest of the population. They just don't have an excuse to fall back on.

      • Actually, since the definition of "idiot" is based on IQ, anyone with an IQ of 20 or higher has exactly 0 chance of being an idiot. They may not use their intelligence, so they may be quite dumb, but "idiot" has a definition that excludes all but the absolute least capable people in the world.

        You actually make a great example of this. I'm almost positive your IQ is 20 or higher; you're here, so it's quite likely it's over 120, even; yet you didn't know the mental health definition of "idiot". This is, as
  • ... correlates with not playing it at all.

  • Let's assume there is a real corellation between IQ and score in LoL. And let's further assume that this is good enough to use it as a predictive tool. Then you can at best identify the IQ of those 100 mio gamers using it. Unfortunately, LoL gamers are not a random selected group of any society and definitely not globally. Therefore, you cannot just scale that up, like you do with an perfectly randomized set.

    • exactly. A facebook IQ link would be more interesting. You could make a much better argument that facebook can be used to measure some aspect of populations globally and get actual interesting populations. again not complete populations is facebook had that kind of penetration they wouldn't be creating initiative to bring facebook aka "the internet" to the global south.
  • When did it jump from November to April First?

  • the massive datasets from these online games could be used to assess population-level cognitive health in real-time across the globe.

    I mean do people really think entire populations play LoL everywhere around the globe? You're talking a really really limited data set. I mean event the simple tech barrier (has computer or money to regularly visit a cyber cafe) is enough to remove a LOT of people from your supposed "population".

  • That means that proficiency in those games peaks at the same age as raw IQ -- about 25 -- while scores in more reaction-time based games like Destiny or Battlefield seem to decline from the teens onwards.

    No no no, those young punks win because they're all Low Ping Bastards! >:-(

"Marriage is low down, but you spend the rest of your life paying for it." -- Baskins

Working...