5 ms·

I’ve wondered recently if speed and intelligence have anything to do with each other at all. It does in IQ tests but Darwin by his account would have been avera

I’ve wondered recently if speed and intelligence have anything to do with each other at all. It does in IQ tests but Darwin by his account would have been average at best. Further, the argument goes that there exists a g factor, and performance in a test that is sufficiently complex will be predictive of performance at anything else sufficiently complex ... Yet if we allow time to vary this seems to be so obviously false...

some_account 8y ago
I think most people with high iq are awful to work with since they obviously lose interest trying to explain things to less intelligent people. This makes people feel bad around them, and they avoid asking them things. I have seen nice people with high iq but I think it's more common that they are not great around people.
- lifeisstillgood 8y ago
  I am not sure about that - those people I know who are waaay cleverer than me have bad days of course, but mostly if you ask them to explain the subject they love, they will talk (down) for days without getting bored. Perhaps it's true for us all, we love to talk about what we do.
- castlecrasher2 8y ago
  How high an IQ are you talking here? I have a moderately high IQ (~130) and being able to explain complex topics to family and friends in a simple enough way that they understand it is one of the things I enjoy most, probably because having a basic summary then the meaty material is how I best learn complex topics. I don't think humility, empathy, and high IQ are mutually exclusive, though given some of my experiences with smart people I can see why you might think that way.
- jacobush 8y ago
  I also think some people with high iq learn to fake a lower iq and thus, go below the radar.
  - zentiggr 8y ago
    There are some times I wish I had gone with this option... would have made a lot of life so much easier than the "but you're so smart, how come you're having trouble with ______?" Wow that sounds whiny but true.
- throwawaymath 8y ago
  How could you tell the IQ of people you were working with? I'm struggling to see how you could arrive at this conclusion unless you've 1) worked with an inordinate number of both high IQ individuals, who 2) explicitly mentioned their IQ scores to you. Personal hypotheses such as this one spread like wildfire online because they're easy to state, play to a narrative and are extremely difficult to disprove. They're stated confidently while skipping over so many little gaps - for example, I don't see anything obvious about how high IQ individuals would stop trying to explain things to low IQ individuals. It makes sense as a heuristic for explaining the behavior of one particular individual who is perhaps a misanthrope or has little patience. But it's very poorly generalizable.
kqr 8y ago
When I had my IQ test results it seemed like standard practise where I was to give two numbers: one total, and one when not counting the tests relying on speed. Not too surprisingly to anyone who knows me, the difference between the two numbers was for me unusually high. On the other hand, contrary to the Darwin quote, I scored very high on abstraction capabilities -- although before getting my results, I had assumed I was doing very bad at that. Conclusion: you probably want to measure all subtests individually to make sense of your strengths and weaknesses.
- fjsolwmv 8y ago
  This is indeed known in the psychological practice in general (it's called "processing speed"), but most lay people aren't aware of it. Ask your friends if they ever got a professional IQ score and if they got the the different scores you describe. You tend to see it with people whose parents notice a high intelligence with learning difficulty and fight for thorough diagnostic testing and educational services. Common tests don't explicitly measure it, but the test design either implicitly demands processing speed (timed tests, or rapidly spoken prompts/questions) or avoid it.
growlist 8y ago
I've often thought about this - I suffered at school because the subject matter quickly seemed to arrive at one or more foundational concepts to not be explored further, whereas my brain would constantly ask, 'but why?'. The relentless pace didn't suit me at all. Nevertheless I learnt to override this in time and now have a BSc and MSc.
- andai 8y ago
  Did you begin a parallel learning project for actually understanding the material?
  - growlist 8y ago
    When I was at school you mean?
- billysielu 8y ago
  I had that problem at school too. It felt like a denial of service attack.
- techstrategist 8y ago
  That’s interesting, I shared the “but why” until mid way through a BSc in Physics. The foundational concepts of “this is how {thermo|mechanics|quantum} works” was exciting, but the remainder of each semester was “now let’s work through all the edge cases” and I had difficulty keeping up.
- handsomechad 8y ago
  Same here! Glad other people felt this way.
usgroup 8y ago
I kind of think that an intelligence tests should moreso try to evaluate how “Turing complete” you are , rather than how quick the processor is at a narrow range ... Kind of reminds me of the programming language shootout dilemma . X is better than Y because it does Fibonacci sequences quicker ... mostly has no real world utility
- throwawaymath 8y ago
  Properly administered IQ tests with real psycometric rigor evaluate both, insofar as either can be empirically measured.
  - usgroup 8y ago
    Except it doesn’t ... Compare Rationality Quotient to Intelligence Quotient. You’d think they relate but they don’t ... https://www.nytimes.com/2016/09/18/opinion/sunday/the-difference-between-rationality-and-intelligence.html https://www.nytimes.com/2016/09/18/opinion/sunday/the-differ... This is without considering give-a-f??k quotient, the most important of all, and that by which every test is confounded.
    - throwawaymath 8y ago
      Yes, they actually do. IQ tests evaluate not just mental processing speed but also a number of orthogonal cognitive faculties. You cited an opinion article from the New York Times regarding the divergence between rationality and intelligence. I'm not talking about rationality, and in fact I would not immediately believe rationality and intelligence can be adequately tested using the same examination. One relates to decision making processes and the other relates to learning and aptitude processes. There is some overlap, but they're categorically different things. I could direct you to a survey of studies that would more accurately describe the breadth of IQ assessments [1], but I think using my own IQ examination as an example might be more instructive. I've taken a formally administered IQ assessment under a qualified clinician. This assessment included the usual tests, such as the Wechsler Adult Intelligence Scale and Wechsler Memory Scale (among others). The Wechsler Adult Intelligence Scale includes composite domains for "Full Scale IQ", "Verbal Comprehension", "Perceptual Reasoning", "Working Memory" and "Processing Speed." Note that mental processing speed is merely one of the measured dimensions of intelligence here. Each of these is given its own percentile score. Within the Verbal Comprehension Scale domain there are subtests for "Similarities", "Vocabulary" and "Information." Under Percentual Reasoning there are subtests for "Block Design", "Matrix Reasoning" and "Visual Puzzles." Working Memory includes tests for "Digit Span" and "Arithmetic", and finally Processing Speed includes "Symbol Search" and "Coding." During the verbal comprehension battery, you'd be asked questions such as, "How are an apartment and a house alike", or "What is the definition of this word?" You'll also be assessed on your knowledge of various culturally relevant facts. Perceptual reasoning will test spatial orientation, integration performance and fine motor skills. This might include arranging blocks of various colors and shapes into different patterns and being asked to find the missing portion in a complex visual pattern. The working memory examination will test for how much you can focus on simultaneously, and explicitly does not test for your processing speed. You'll be asked about sequences of numbers and tested for both long and short term recollection, as well as how accurately you can mentally perform operations on sequences while keeping them in your mind. That's the digit span test; the arithmetic tests are self-explanatory. The only portion of the examination which explicitly tests for processing speed are the ones focusing on symbol search and coding. Therein you're asked to quickly scan a group of symbols and find a match of one or two targets, or to copy numbers lexicographically coded to symbols as quickly as possible. I am interested in what measure of analogous "Turing Completeness" in human intelligence should thus be measured that is not already reflected in the examination I've just outlined. If you'd like to define that as rationality, then that's fine. But as I said before, I don't believe rationality should be coupled with intelligence. And moreover, I don't think "Turing Completeness" is a good analogy for rationality, since computers are (almost definitionally) neither rational nor irrational, regardless of their computational capabilities. ______________________________ 1. http://www.gwern.net/Iodine http://www.gwern.net/Iodine
trentmb 8y ago
My undergrad was in mathematics. I routinely failed midterms, but would ace the final exams. I generally felt that, by the end of the term, I understood the material and simply had more time during the final exam than the midterms.

some_account 8y ago

I think most people with high iq are awful to work with since they obviously lose interest trying to explain things to less intelligent people. This makes people feel bad around them, and they avoid asking them things. I have seen nice people with high iq but I think it's more common that they are not great around people.

lifeisstillgood 8y ago
I am not sure about that - those people I know who are waaay cleverer than me have bad days of course, but mostly if you ask them to explain the subject they love, they will talk (down) for days without getting bored. Perhaps it's true for us all, we love to talk about what we do.
castlecrasher2 8y ago
How high an IQ are you talking here? I have a moderately high IQ (~130) and being able to explain complex topics to family and friends in a simple enough way that they understand it is one of the things I enjoy most, probably because having a basic summary then the meaty material is how I best learn complex topics. I don't think humility, empathy, and high IQ are mutually exclusive, though given some of my experiences with smart people I can see why you might think that way.
jacobush 8y ago
I also think some people with high iq learn to fake a lower iq and thus, go below the radar.
- zentiggr 8y ago
  There are some times I wish I had gone with this option... would have made a lot of life so much easier than the "but you're so smart, how come you're having trouble with ______?" Wow that sounds whiny but true.
throwawaymath 8y ago
How could you tell the IQ of people you were working with? I'm struggling to see how you could arrive at this conclusion unless you've 1) worked with an inordinate number of both high IQ individuals, who 2) explicitly mentioned their IQ scores to you. Personal hypotheses such as this one spread like wildfire online because they're easy to state, play to a narrative and are extremely difficult to disprove. They're stated confidently while skipping over so many little gaps - for example, I don't see anything obvious about how high IQ individuals would stop trying to explain things to low IQ individuals. It makes sense as a heuristic for explaining the behavior of one particular individual who is perhaps a misanthrope or has little patience. But it's very poorly generalizable.

kqr 8y ago

When I had my IQ test results it seemed like standard practise where I was to give two numbers: one total, and one when not counting the tests relying on speed. Not too surprisingly to anyone who knows me, the difference between the two numbers was for me unusually high. On the other hand, contrary to the Darwin quote, I scored very high on abstraction capabilities -- although before getting my results, I had assumed I was doing very bad at that. Conclusion: you probably want to measure all subtests individually to make sense of your strengths and weaknesses.

fjsolwmv 8y ago
This is indeed known in the psychological practice in general (it's called "processing speed"), but most lay people aren't aware of it. Ask your friends if they ever got a professional IQ score and if they got the the different scores you describe. You tend to see it with people whose parents notice a high intelligence with learning difficulty and fight for thorough diagnostic testing and educational services. Common tests don't explicitly measure it, but the test design either implicitly demands processing speed (timed tests, or rapidly spoken prompts/questions) or avoid it.

growlist 8y ago

I've often thought about this - I suffered at school because the subject matter quickly seemed to arrive at one or more foundational concepts to not be explored further, whereas my brain would constantly ask, 'but why?'. The relentless pace didn't suit me at all. Nevertheless I learnt to override this in time and now have a BSc and MSc.

andai 8y ago
Did you begin a parallel learning project for actually understanding the material?
- growlist 8y ago
  When I was at school you mean?
billysielu 8y ago
I had that problem at school too. It felt like a denial of service attack.
techstrategist 8y ago
That’s interesting, I shared the “but why” until mid way through a BSc in Physics. The foundational concepts of “this is how {thermo|mechanics|quantum} works” was exciting, but the remainder of each semester was “now let’s work through all the edge cases” and I had difficulty keeping up.
handsomechad 8y ago
Same here! Glad other people felt this way.

usgroup 8y ago

I kind of think that an intelligence tests should moreso try to evaluate how “Turing complete” you are , rather than how quick the processor is at a narrow range ... Kind of reminds me of the programming language shootout dilemma . X is better than Y because it does Fibonacci sequences quicker ... mostly has no real world utility

throwawaymath 8y ago
Properly administered IQ tests with real psycometric rigor evaluate both, insofar as either can be empirically measured.
- usgroup 8y ago
  Except it doesn’t ... Compare Rationality Quotient to Intelligence Quotient. You’d think they relate but they don’t ... https://www.nytimes.com/2016/09/18/opinion/sunday/the-difference-between-rationality-and-intelligence.html https://www.nytimes.com/2016/09/18/opinion/sunday/the-differ... This is without considering give-a-f??k quotient, the most important of all, and that by which every test is confounded.
  - throwawaymath 8y ago
    Yes, they actually do. IQ tests evaluate not just mental processing speed but also a number of orthogonal cognitive faculties. You cited an opinion article from the New York Times regarding the divergence between rationality and intelligence. I'm not talking about rationality, and in fact I would not immediately believe rationality and intelligence can be adequately tested using the same examination. One relates to decision making processes and the other relates to learning and aptitude processes. There is some overlap, but they're categorically different things. I could direct you to a survey of studies that would more accurately describe the breadth of IQ assessments [1], but I think using my own IQ examination as an example might be more instructive. I've taken a formally administered IQ assessment under a qualified clinician. This assessment included the usual tests, such as the Wechsler Adult Intelligence Scale and Wechsler Memory Scale (among others). The Wechsler Adult Intelligence Scale includes composite domains for "Full Scale IQ", "Verbal Comprehension", "Perceptual Reasoning", "Working Memory" and "Processing Speed." Note that mental processing speed is merely one of the measured dimensions of intelligence here. Each of these is given its own percentile score. Within the Verbal Comprehension Scale domain there are subtests for "Similarities", "Vocabulary" and "Information." Under Percentual Reasoning there are subtests for "Block Design", "Matrix Reasoning" and "Visual Puzzles." Working Memory includes tests for "Digit Span" and "Arithmetic", and finally Processing Speed includes "Symbol Search" and "Coding." During the verbal comprehension battery, you'd be asked questions such as, "How are an apartment and a house alike", or "What is the definition of this word?" You'll also be assessed on your knowledge of various culturally relevant facts. Perceptual reasoning will test spatial orientation, integration performance and fine motor skills. This might include arranging blocks of various colors and shapes into different patterns and being asked to find the missing portion in a complex visual pattern. The working memory examination will test for how much you can focus on simultaneously, and explicitly does not test for your processing speed. You'll be asked about sequences of numbers and tested for both long and short term recollection, as well as how accurately you can mentally perform operations on sequences while keeping them in your mind. That's the digit span test; the arithmetic tests are self-explanatory. The only portion of the examination which explicitly tests for processing speed are the ones focusing on symbol search and coding. Therein you're asked to quickly scan a group of symbols and find a match of one or two targets, or to copy numbers lexicographically coded to symbols as quickly as possible. I am interested in what measure of analogous "Turing Completeness" in human intelligence should thus be measured that is not already reflected in the examination I've just outlined. If you'd like to define that as rationality, then that's fine. But as I said before, I don't believe rationality should be coupled with intelligence. And moreover, I don't think "Turing Completeness" is a good analogy for rationality, since computers are (almost definitionally) neither rational nor irrational, regardless of their computational capabilities. ______________________________ 1. http://www.gwern.net/Iodine http://www.gwern.net/Iodine

trentmb 8y ago

My undergrad was in mathematics. I routinely failed midterms, but would ace the final exams. I generally felt that, by the end of the term, I understood the material and simply had more time during the final exam than the midterms.