Very High Correlation Found Between Brain Size and IQ

I posted on this study a while back, but here is the actual abstract to the study (full paper here). Researchers tried to guess IQ scores of children based upon multiple MRI images of the children’s brains. A correlation of .071 was found between estimator’s guesses and the children’s actual measured IQ’s. The guesses had an average error of ~8 IQ points. That is, the guesses tended to be within 8 IQ points of the actual IQ. This shows that there is indeed a high correlation between brain size and IQ, although it is often hard to measure given today’s technology.

Previous studies had found lower correlations of ~.41, but the recent study used many newer techniques utilizing multiple images that allow better estimation of brain size than the previous techniques.

MRI-Based Intelligence Quotient (IQ) Estimation with Sparse Learning

Abstract

In this paper, we propose a novel framework for IQ estimation using Magnetic Resonance Imaging (MRI) data. In particular, we devise a new feature selection method based on an extended dirty model for jointly considering both element-wise sparsity and group-wise sparsity.

Meanwhile, due to the absence of large dataset with consistent scanning protocols for the IQ estimation, we integrate multiple datasets scanned from different sites with different scanning parameters and protocols. In this way, there is large variability in these different datasets.

To address this issue, we design a two-step procedure for

1) first identifying the possible scanning site for each testing subject and

2) then estimating the testing subject’s IQ by using a specific estimator designed for that scanning site.

We perform two experiments to test the performance of our method by using the MRI data collected from 164 typically developing children between 6 and 15 years old.

In the first experiment, we use a multi-kernel Support Vector Regression (SVR) for estimating IQ values, and obtain an average correlation coefficient of 0.718 and also an average root mean square error of 8.695 between the true IQs and the estimated ones.

In the second experiment, we use a single-kernel SVR for IQ estimation, and achieve an average correlation coefficient of 0.684 and an average root mean square error of 9.166. All these results show the effectiveness of using imaging data for IQ prediction, which is rarely done in the field according to our knowledge.

Conclusion

In this paper, we proposed a novel framework for the estimation of a subject’s IQ score based on the neuroimaging features. Methodologically, since the number of features in neuroimaging data usually overwhelms the number of available samples, feature selection has been always an important role in the field.

To this end, considering the strong relationship between GM and WM features in MR images, we devised a feature selection method based on a dirty model [24] that efficiently considered the coupling of different feature types, but still alleviated the strong parameter overlap across features. Specifically, we penalized an objective function with a squared Frobenius norm of the element-wise sparsity matrix.

Using the MR Images acquired at different scanning sites with their own scanning parameters and protocols, we designed a two-step procedure, by which we first identified the scanning site of a test image and then estimated the test subject’s IQ by using the respective estimator. Also, we performed comparison between multi-kernel SVR and single-kernel SVR by two sets of experiments.

From a practical point of view, although the current framework is not limited to apply for the MR images obtained from only a predefined site, it would be our forthcoming research issue to develop a more generalized method for efficiently handling the inter-site variability and thus constructing a single generalized estimator model for all subjects by skipping the scanning site identification step.

Furthermore, thanks to the availability of various imaging modalities, it would be beneficiary to integrate their complementary information for more precise IQ score estimation. It should be emphasized again that our work paves a new way for a research on predicting an infant’s future IQ score by using neuroimaging data, which can be a potential indicator for parents to prepare their child’s education if needed.

35 Comments

Filed under Intelligence, Neuroscience, Psychology, Science

35 responses to “Very High Correlation Found Between Brain Size and IQ

  1. Jason Y

    But again IQ doesn’t matter in terms of coming out ahead in lfe. More important is a good work ethic, coming to work on time, being in a good vocational trade. or being a good saver in say a janitor job etc..

  2. Jason Y

    Perhaps even a smaller brain can train itself to process information better. It would be like a bodybuilder training his body. On the flip side, a large brain inside of a lazy person might become weak.

    • Jason Y

      Iv’e noted that in studying math, Iv’e become sharper and smarter than before. That’s because my brain is being put to the test more than some guy who watches TV all day.

      • Jason Y

        Perhaps lower IQ people could train the brains they do have, and in a sense become sharper despite the small brain handicap. I don’t think a lot of people, including the lower IQ range, do anything but watch TV all day, or surf the internet, or talk on cell phones. What do they do, assuming they’re not working at some job?

  3. HK woman

    Asian don’t care the brain size. We found that higher and wider forehead could reflect high intelligence quality. I always find Caucasians and East asians have these traits.

  4. Jason Y

    Someone on a forum said brain size was irrrelevant and it had more to do with the number of neural connections which would obviously be reduced as the result of poor nutrition and birth defects (from tobacco smoking, drugs etc..)

    • That’s the thing Jason, there’s a difference in those tow factors. Size, I believe, contributes with brain power/ gray matter, which relates to the ability for someone to focus.
      Neural connections, while indeed are capable of being environmentally depressed, are also selected for as brain mutations that are beneficial for more complex abilities. This has been typically selected with brainsize.

    • TJF

      Someone on a forum said brain size was irrelevant ..

      Not irrelevant but not solely relevant. You are correct that neural efficiency plays an important but it is not the sole difference.

      Encephalization quotient (EQ), or encephalization level, is a measure of relative brain size defined as the ratio between actual brain mass and predicted brain mass for an animal of a given size, which is hypothesized to be a rough estimate of the intelligence or cognition of the animal.

      Both are important and it should noted that the Encephalization quotient (EQ), or encephalization level (ratio of brain size to non obese body mass) comes into play. Blue whales have larger brains than humans but in relation to mass their brains are 1/3 the size of humans.

      A 6’6″ Norwegian lumberjack may have larger brain than a female 5’1″ Chinese physicist but it is quite possible that the ratio of the lumberjack’s brain to body mass would be smaller – especially the portions that pertain to higher functions versus involuntary life sustaining functions such as breathing, swallowing and heart rate.

      • Another William Playfair Web

        TJF- “A 6’6″ Norwegian lumberjack may have larger brain than a female 5’1″ Chinese physicist but it is quite possible that the ratio of the lumberjack’s brain to body mass would be smaller – especially the portions that pertain to higher functions versus involuntary life sustaining functions such as breathing, swallowing and heart rate.”

        That’s what I initially thought, but I was told I was wrong on another HBD blog, I’ve heard that tallness itself, for some reason is a sign of higher IQ, so the advantage stays;
        https://pumpkinperson.com/2016/03/19/estimating-rosie-odonnells-iq/#comment-23750

        Women have smaller brains, but that’s because they are more verbally oriented than men, who are spatially oriented, which takes up more space.

        • Yeah but the correlation isn’t that strong, it’s 0.2.

          Second, tall is just one measure of body volume but their also actually body mass to consider too.

        • TJF

          To Another William Playfair Web:

          I’ve heard that tallness itself, for some reason is a sign of higher IQ, so the advantage stays..

          Well that reminds me of the classic phrase for a problem in Physics – “When all other factors are the same..” when in the real world they rarely are… let’s do a mental exercise. If you were to take a random sampling of 100 Chinese guys from Shanghai whose average height is 5’7″ and then compare them to a random sampling of 100 Dinka males from South Sudan whose average height is 6’8″ (the average Dinka man is 6’8″) which group would you say would do better on an IQ test?

          Moving away from a discomforting racial comparison let’s compare 100 randomly sampled American Ashkenazim Jewish men but limit the sample to men who are no taller than 5’7″ and compare them to a random sample of 100 white West Virginian men none of whom were less than 6’2″ – which group would you think would do better on an IQ test..? My money is on the group that has guys with surnames like Goldstein and Loeb.

          Height probably has relevance within a specific group but becomes less relevant when comparing across groups – greater height within a given group would likely indicate better nutrition and health – at least during the formative years.

        • Another William Playfair Web

          Phil-
          If I recall, for cranial capacity it’s 0.4, but for cranial circumference (measured just above the eyes)- is 0.2 .
          On brainsize PP gave a conversion from cranial circumference to cranial capacity (since it’s not a perfect sphere).

          That’s how, using the fact that Oprah wears two wigs, he got the cranial circumference, and from there the “statostrohpic” cranial capacity of 2029 mL, which is like 6.5 SDs larger than av. caucasian
          6.5(0.4)= 2.6(15)+100= about 140, more accurate than 0.2 and cranial circumference.

        • I meant IQ, not measurements of the cranium. He explained it an early article though Correlations could’ve changed by that time around.

        • Another William Playfair Web

          I suppose you would have to look at the best indicators of IQ, whatever they may be, then control for them, and then go from there.

          Ashkenazis and Northeast Asian, the two smartest ethnicities in the world based on current testing, are smarter than the two groups tied for third (Europeans and SE Asian Mongoloids), who are taller and equivalent, respectively.

    • Jason Y

      Interesting comments on the forum, but I went on there and asked for sources because on blogs etc,.. a comment isn’t enough.

  5. Gay State Girl

    Lindsay

    You say there is high correlation between IQ and reaction time?

    • more likely moderate, about .4 out of 1.

      • Gay State Girl

        I have an IQ of 122 but abysmally slow reaction time and coordination. I suspect that there are Subsaharan Africans who have IQs more than 50 points below mine, who could operate a vehicle more efficiently than I could. So maybe there is something to the notion of cultural biases.

        • Another William Playfair Web

          GSG-
          Ashkenazis have very lopsided IQs, with verbal IQs like 2SDs above Whites, and Spatial IQs the same as Whites, overall about +1SD.

          Performance, which involves the speed is more accurately linked to performance IQ.

        • Gay State Girl

          I don’t know why reaction time is incorporated into one’s spatial IQ. I have above average spatial awareness probably ~ 115.

        • Gay State Girl

          Any other women in the house care to comment?

        • Another William Playfair Web

          I have terrible reaction times.
          I am honestly not sure what exactly my IQ is (regression only works in populations accurately)- but according to online charts, based on my PSAT score has an IQ of 124, and SAT 119.
          I had untreated anxiety at the time of the actual SAT, and actually did better on Math but far worse on reading.

          I did better on Math than Critical Reading, but I am skeptical that means my Spatial IQ is better. Math often does involve verbal type thinking as well, and likewise does not require spatial thinking.

        • TJF

          To Gay State Girl:

          I have an IQ of 122 but abysmally slow reaction time and coordination.

          I believe you fall somewhere on the Autism spectrum..? Results for people who do not fit within the “normal” spectrum probably diverge substantially from what is expected for a given standard IQ. I have a daughter who has been diagnosed with Asperger’s who has very high scores in certain areas but low scores in others.

        • Gay State Girl

          TJF

          I lack the hallmark characteristic of Asperger’s, I have no problem understanding body language or voice intonation. I do have developmental dyspraxia, which includes severe problems with motor control, coordination, reaction time, and organization, and also sensory sensitivity and repetitive and self stimulating behavior.

        • Gay State Girl

          TJF

          Good to know about another female who is neurodiverse. I was accustomed to being the only girl at disability rehabilitation programs as a kid.

        • Yeah, variation like that is why we use averages. A similar one is income, which is moderately correlated and as we can see many with a High IQ haven’t exactly gained the highest living.

  6. Another William Playfair Web

    Really, I never understood why, 0.4 was considered so strong, I mean, although mine crashes down because it’s so extreme- I have a head that’s like +4SDs in cranial capacity, as big as one in 10,000 caucasians, but 4(0.4)= 1.6 SDs, doesn’t even make me “genius” (+2SDs).😦

    That being said, I must be honest and say I’m skeptical it’s 0.71, I would have an IQ close to 145 if that was the case.

  7. TJF

    To Another William Playfair Web:

    That’s how, using the fact that Oprah wears two wigs, he got the cranial circumference, and from there the “statostrohpic” cranial capacity of 2029 mL, which is like 6.5 SDs larger than av. caucasian
    6.5(0.4)= 2.6(15)+100= about 140, more accurate than 0.2 and cranial circumference.

    Oprah Winfrey’s head circumference is public knowledge..?Ms. Winfrey is probably rather bright but measuring head circumference does not take into account skull thickness among other things. A much more accurate way to measure cranial capacity of a living person is through an MRI but I doubt we have many broad based studies with large population samples using this measurement as MRIs are rather expensive.

    • Another William Playfair Web

      he said 25.5 inches….which can be converted.

      I do find it interesting that, he estimates the IQ of the average self-made billionaire at 136, and through both the cranial capacity, and “idea density” analyses, he has come up close to 140.

      When I was talking about “White people being kidnapped for their presumed money” I was mainly referring to disproportionately Native American left-wing rebels in Colombia, with whom Pablo Escobar briefly allied.

      Pablo Escobar was a self-made Decabillionaire (worth about $25 billion in the 1980s), and was the 7th richest man in the world at his peak.
      He did this while headquartered out of a country with a current median income of about $12,000 annually, so a second world/second tier country in terms of income. and he died at 44

      He was also said to be a lightning calculator.
      I’d suspect, although he was a criminal (lower IQs), the other factors push his IQ back up to around 146.
      PP claims to be a leftie, but in regards to his obsession of “IQ-income” correlation he’s fairly right-wing.

      • Another William Playfair Web

        I also suspect he was a Crypto-Sephardic Jew;

        hoping to get the Blacks addicted and acting crazy. His victims in Colombia were disproportiantly White.😉

        • TJF

          To Another William Playfair Web:

          he said 25.5 inches….which can be converted.
          Ummm, who said this and how did he know Oprah Winfrey’s head circumference? Errr.. a special relationship with Oprah combined with an unusual fetish? Ok maybe that information is publicly available but I find it odd.

          https://brainsize.wordpress.com/2014/07/06/the-brain-size-of-the-worlds-most-successful-woman/

          Ahh ok I found the reference – umm I guess.. we have to trust the reporting from an old print magazine, Oprah’s word, and this guy’s memory.

          According to “the internet” – 26 for a man would place him in the 99.2 percentile. So yes 25.5 inches would be quite large for a woman.
          http://forum.bodybuilding.com/showthread.php?t=136850963

          I also suspect he was a Crypto-Sephardic Jew
          I suppose you mean that he might have been a descendant of Crypto-Jews as I don’t think there is currently any issue with practicing Judaism in Colombia. That said — nobody expects the Spanish Inquisition:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s