If you measured someone’s IQ, what could you guess about the rest of their life?

More than you’d expect. And less than the number deserves.

Intelligence correlates with brain size, reaction time, height, income, trust, crime, religiosity, myopia, how long you live, and whether you can spot a sentence that sounds deep but means nothing. That is not a tabloid list. Every item on it has a real, replicated literature behind it.

Here is the catch. Almost all of those correlations are small. Most of them shrink the instant you look closely. And for nearly all of them, “IQ predicts X” does not mean “IQ causes X.” Usually it’s the other way around, or a third thing causes both.

So this is a tour of what actually tracks intelligence. Not the listicle version. The version where we ask, for each one, how big is it really, and why might it be there at all.

First, a ruler.

What a correlation actually buys you

IQ is a standardized score for general cognitive ability. Give people a wide enough range of mental tasks, and performance on all of them tends to move together. That shared factor is what Spearman called g, and the fact that diverse tests intercorrelate (the “positive manifold”) is one of the most replicated findings in psychology.¹²

What g is underneath, whether it’s one thing or many things that happen to travel together, is still argued about. But that diverse tests agree with each other is not in doubt.

Now the ruler. Most of the correlations in this post live around r = 0.2 to 0.4. People hear “correlation” and picture a tight line. They shouldn’t.

r = 0.2  ->  IQ explains ~4% of the variation in X
r = 0.3  ->  ~9%
r = 0.5  ->  ~25%   (rare in this whole list)

Four percent is real. It shows up across thousands of people. It is also almost invisible in any single person you meet. Keep that ruler in your hand for everything below.

The body

Start with the most literal question. Is intelligence sitting in physical tissue you could point at?

Sort of… Bigger brains do go with higher IQ. The best meta-analysis, across 88 studies and more than 8,000 people, puts it at about r = 0.24.³ That’s roughly 6% of the variance. The older, smaller studies said 0.3 to 0.4. Some of that gap is publication bias, the splashy positive results made it into print and the nulls quietly didn’t. Though not all of it: others argue the higher number mostly survives once you account for how sloppily IQ was measured in the first place.

Either way, the more careful work says size isn’t the right thing to look at. What predicts g better is where the tissue is and how well it’s wired. The fashionable account is a fronto-parietal network rather than gross volume, the integrity of the white-matter tracts connecting those regions, not how much bulk you’ve got.⁴

That wiring idea matters, because it shows up again.

Reaction time is the cleaner story. Faster, more consistent reaction times go with higher IQ. Across 172 studies and over fifty thousand people, the correlation is about -0.26, climbing as the task gets harder.⁵ The interesting bit is the variability: how much your reaction time wobbles from trial to trial predicts IQ as well as your average speed does, sometimes better.

Why would raw speed track intelligence at all? One camp (Jensen’s) says mental speed is the substrate of g. The more careful reading says speed is a readout of the wiring. White-matter tract integrity predicts intelligence, and that link is fully mediated by processing speed: intact axons, fast transmission, higher g.⁶ Speed isn’t the cause. It’s the gauge.

Hold onto that, because it’s about to tie the whole post together.

Then there’s height. Taller people are, on average, slightly smarter. The correlation is small, around 0.1 to 0.16.⁷ And this is the cleanest example of the trap this whole post is about.

Because the height-IQ link is almost entirely genetic, and it shrinks when you compare siblings instead of strangers. Common genetic variants explain about 71% of it. That genetic overlap splits roughly two ways: some genes happen to nudge both height and IQ (pleiotropy), and tall people tend to pair with smart people, so the alleles ride together down the generations (assortative mating). A real shared-gene thread does survive within families. But it’s shared genes and who-pairs-with-whom, not a taller body growing a better brain.⁷

Taller does not make you smarter. The alleles just travel in the same car.

And myopia. Smarter, more-educated people really are more likely to be nearsighted. This one has the strongest causal evidence in the body section, and it runs the surprising way. A Mendelian-randomization study using genetic variants as natural instruments found that education causes myopia, about a quarter-diopter of nearsightedness per extra year of schooling, with no detectable effect in the reverse direction.⁸ More school, worse eyes.

Why? Two theories. Near-work (reading, screens) and reduced time outdoors. The current evidence leans hard on outdoor light. Randomized trials that simply added 40 minutes of outdoor time to the school day cut three-year myopia incidence from about 40% to 30%.⁹ Bright light, not book-avoidance, looks like the lever.

(And the related folk belief, that glasses make you look smart? The research is mixed and culture-dependent. In Western samples it mostly holds; one of the studies people cite for “no effect” is a non-Western sample where glasses lowered perceived intelligence. So the stereotype is real, conditional, and a bit silly.¹⁰)

Where it comes from, and what wrecks it

Birth order is a fun one because the myth is so much bigger than the effect.

Firstborns are, on average, a sliver smarter. About 1.5 IQ points per step down the birth order, give or take.¹¹ Real, replicated across enormous datasets, including comparisons within the same family.

But notice the size. One and a half points. And the famous “firstborns are natural leaders, bossy, conscientious” story? It collapses. The same large studies find essentially no birth-order effect on personality. Just the tiny IQ sliver.¹²

Why does even that exist? The best clue is eerie. In a Norwegian conscript study, second-born men whose older sibling had died in childhood, so they grew up as the social eldest, scored like firstborns.¹¹ That points at social rank inside the family, not anything biological about being gestated first. Parents also measurably spend more one-on-one time with the firstborn, which fits the resource-dilution idea.¹³ Which exact mechanism wins is still open. But it’s environment, not birth-rank biology.

Now the things that genuinely lower intelligence, where causation is real and runs downhill.

Lead. Childhood lead exposure lowers IQ, and this is about as close to established causation as this field gets. The pooled international analysis found about 6 IQ points lost as blood lead rises from under 1 to 10 micrograms per deciliter, and the damage is worse per unit at the lowest doses. There is no safe threshold.¹⁴ The mechanism is known down to the synapse: lead blocks the NMDA receptor, which gates the long-term potentiation that learning is built on.¹⁵ This isn’t a correlation you puzzle over. It’s a poison.

Nutrition is messier, and worth being honest about. “Eat well, get smarter” is mostly not what the evidence shows. The clean causal case is narrow: iodine. Severe iodine deficiency costs a lot of IQ, and a natural experiment (US salt iodization in the 1920s) raised scores by roughly a standard deviation in the most-deficient regions.¹⁶ That’s huge, and specific.

The popular version, breastfeeding raises IQ, is the textbook example of confounding. The raw effect is about 4 points. Adjust for the mother’s IQ and roughly three-quarters of it evaporates.¹⁷ Smart mothers are more likely to breastfeed and to have smart kids. The milk gets the credit; the genes and the household did the work.

That pattern, where a wholesome-sounding cause turns out to be a stand-in for poverty or parental IQ, is everywhere in this topic. Hold the suspicion.

The mind and its temperament

Of the big five personality traits, only one tracks intelligence at all, and it’s the obvious one. Openness to experience, especially the intellectual-curiosity, ideas-loving side of it, correlates with IQ at about 0.20.¹⁸ Conscientiousness, the diligent, organized, gets-things-done trait, sits at roughly zero.¹⁹

That surprises people. Smart and hardworking feel like they should go together. They basically don’t.

There’s even a hypothesis, weakly supported, that they trade off: less able people build up conscientiousness to compensate, because they can’t coast on raw ability. To be honest, the cleaner reading is that there’s barely a correlation to explain in the first place, and the small negative ones that show up are mostly an artifact of who stays in school.¹⁹

The openness link has a nicer question buried in it. Which way does it run? Does being smart make you curious, or does curiosity build the intelligence? The “investment” idea, that a hungry mind keeps seeking out hard things and compounds into knowledge over a lifetime, has real support: intellectual curiosity predicts academic performance about as well as effort does.²⁰ Curiosity as a third engine, alongside ability and diligence. I find that one genuinely hopeful.

Then there’s the party trick everyone wants to be true: smart people can smell nonsense.

Partly. When you give people pseudo-profound sentences, the randomly-assembled “a formless void serves the mechanics of destiny” kind, people lower in analytic thinking rate them as more profound.²¹ So yes, the willingness to call out bullshit tracks cognitive ability.

But the honest version is weaker than the flattering one. The strong correlation is with receptivity (rating garbage as deep), not sensitivity (telling garbage apart from the genuinely deep). A lot of the effect is just that lower-ability people rate everything as more profound. It’s less “smart people have a finely-tuned nonsense detector”… and more “some people are generous to all sentences.”

Smart people are also more patient, in a specific sense. They discount the future less: offered money now versus more money later, higher-IQ people more often wait. The meta-analytic correlation is about -0.23, partly running through working memory.²² Small, but it’s one of the few places the stereotype of the far-sighted clever person has data under it.

And creativity? Here the folk belief is just wrong. The overall IQ-creativity correlation is weak, around 0.17. The beloved “threshold” idea, that IQ and creativity move together up to about 120 and then decouple, is measure-dependent at best and rejected by meta-analysis at worst.²³ The clean “you need IQ 120 to be creative, after which it stops mattering” line is folklore with a lab coat on.

Faith, trust, and the things we argue about

Higher IQ goes with lower religiosity. The meta-analysis lands around -0.2 for strength of belief.²⁴ Real, replicated, and small. Notice the ruler again: that’s about 4% of the variance, weighted toward American and college samples, and apparently declining over time.

The tempting explanation is that intelligence makes you analytic, and analytic thinking overrides the intuitions that religion runs on. It’s a lovely story. It’s also where I have to stop you, because it’s the best example in this whole post of a reasoning that sounds true and doesn’t survive a hard look.

The famous experiments, where nudging people to think analytically (showing them Rodin’s The Thinker, say) supposedly reduced their religious belief, failed to replicate. Twice. A large preregistered direct replication found an effect of essentially zero, and the same study was among those that flunked the big Social Sciences Replication Project.²⁵ The bare correlation between analytic style and disbelief survives, but it’s weak, around -0.18, and it partly disappears depending on which question you ask first.

So: there’s a small, fragile link. There is not good evidence that thinking harder makes you an atheist. The clean causal story is the kind of thing we should want to believe and shouldn’t.

Trust is a friendlier finding. Higher cognitive ability, verbal ability in particular, goes with more generalized trust, the willingness to say most people can be trusted. The gap across the ability range is large in the raw data.²⁶ The nicest interpretation, and the best-supported one, is that it isn’t naivety: smarter people are better at reading who’s actually trustworthy, so they end up correctly trusting more. They’re not gullible. They’re calibrated. Though, fair warning, that mechanism is inferred from the correlation surviving controls, not directly tested.

Higher verbal ability also goes with more willingness to extend free speech to groups you personally can’t stand, across the whole ideological spectrum.²⁷ The thread running through that one looks like intellectual humility and a lower sense of threat: if a hateful group’s rally doesn’t feel like an existential danger to you, you’re more relaxed about letting them have it.

And there’s a politically loaded one I’ll state carefully, because it’s easy to weaponize and easy to overread. In British cohort data, higher childhood IQ predicts more socially liberal, anti-authoritarian, lower-prejudice attitudes in adulthood, mediated by education.²⁸ Read the fine print before anyone quotes this at a dinner party. The thing being measured is an attitude bundle (anti-racist, anti-authoritarian, institution-trusting), not party affiliation and not economic policy. The same literature can link childhood IQ to economic conservatism. “Smart people agree with me politically” is not in this data. What’s in it is openness to out-groups.

Now the famous one everyone gets backwards.

The Dunning-Kruger effect is not “dumb people think they’re smarter than smart people.” It’s narrower: low-skill people overestimate their own absolute ability, because the skills you’d need to do the task are the same skills you’d need to know you did it badly.²⁹ In the original data the bottom group still rated themselves below the top performers. They were just badly miscalibrated upward.

That’s the part the meme gets wrong. Here’s the part almost nobody mentions: the famous chart is mostly a statistical artifact. When you analyze the data with the right tools instead of the seductive quartile graph, the dramatic curve largely flattens. You can even reproduce the classic “Mount Stupid” shape out of pure random numbers plotted that way.³⁰ Some small genuine effect may survive. The cartoon version doesn’t.

I love this example, because it’s the same lesson as the analytic-thinking-and-religion one. A satisfying story about who’s deluded turns out to be partly an illusion in how the graph was drawn.

What it buys in life

Does intelligence pay?

In income, somewhat. The big longitudinal meta-analysis puts IQ’s correlation with education at about 0.56, with occupational status around 0.43, and with income only about 0.2.³¹ Notice the drop. IQ predicts how far you get in school strongly, what kind of job you land moderately, and how much you earn only weakly. And it’s not a dramatically better predictor of any of these than your parents’ status was.

Then it falls off a cliff. IQ predicts income but not wealth. In one large US cohort, each IQ point was worth a few hundred dollars a year in earnings, but had no significant relationship to net worth. And financial distress, missed payments, maxed cards, bankruptcy, didn’t fall cleanly with IQ at all. The relationship there isn’t even a straight line: being smart was no guarantee against money trouble, and on some measures the high-IQ group came off slightly worse.³² The intelligence shows up in the paycheck and vanishes by the bank balance. What you do with money, save it or spend it, isn’t the same trait as earning it.

“Smart predicts earning, not having.” That one deserves to be better known.

Crime runs the other way, and I’ll be careful with it. Lower IQ, specifically lower verbal IQ, is associated with more offending. Offenders average something like 8 to 10 points below non-offenders.³³ It’s mostly a verbal-ability effect, heavily tangled with poverty and schooling, and it is not simply that dimmer criminals get caught more, the link shows up in self-reported offending too. But it’s a population-level association with huge confounds, not a verdict on any individual. Treat it as such.

Job performance? This is where a number you may have heard needs an update. The classic figure had general mental ability predicting job performance at about 0.5, the single best predictor in the toolkit. A 2022 reanalysis argued the old corrections were overcooked and put the real operational validity closer to 0.3, behind structured interviews.³⁴ Still useful. No longer king. If someone quotes you the 0.5, they’re citing the old number.

And the one that should give the whole “IQ is destiny” crowd pause: it predicts how long you live. Higher childhood IQ goes with lower mortality from basically everything, around a 20% lower risk of death per standard deviation of IQ, for heart disease, stroke, cancer, dementia.³⁵ Part of that is the better jobs and habits that IQ buys. But not all of it.

Which brings us to the thread I promised.

One wire, many gauges

Why would the same number predict your reaction time, your brain’s wiring, and your age at death?

The cleanest idea on offer is what Deary calls “system integrity.” A well-built nervous system might show up as both faster, more reliable cognition and a body that simply runs better and lasts longer. On that view IQ isn’t causing your longevity through smart choices. It’s a marker of an underlying biological soundness that drives both.³⁶

The evidence for it is genuinely clever. Reaction time, which is nearly free of culture and schooling, predicts the same causes of death as IQ does, with similar effect sizes. And when you adjust for reaction time, the IQ-mortality link shrinks.³⁶ That’s hard to explain with “education and money,” and easy to explain with “the wiring was better.”

I want to be honest that this is a hypothesis, not a fact. Deary himself calls the idea underdeveloped and undertested. But it’s the closest thing this sprawling topic has to a spine: speed, structure, and survival might all be reading off the same underlying integrity of the hardware.

The things you can actually act on

Most of this post is curiosities. A few items are levers.

Set against the correlational fog, these are the ones with real causal weight, the things where intervening actually moves cognition:

lead exposure       ->  remove it. causal, no safe dose.
high blood pressure ->  treat it. an RCT cut mild-cognitive-impairment risk.
hearing loss        ->  treat it. the largest modifiable midlife dementia factor.
iodine deficiency   ->  fix it. ~a standard deviation in deficient regions.
exercise            ->  do it. modest but real, and it grows the hippocampus.
head injury         ->  avoid it. dose-dependent, probably causal.

Blood pressure: intensive control in a large randomized trial significantly cut the risk of mild cognitive impairment (it narrowly missed significance for full dementia, mostly because the trial stopped early).³⁷ The mechanism is plumbing, chronic high pressure damages the brain’s small vessels and the white-matter wiring, which is exactly the speed-of-the-wiring story again.

Hearing loss is the headline. The Lancet Commission rates it the single largest modifiable risk factor for dementia in midlife.³⁸ And this one isn’t only a correlation. A randomized trial of hearing aids slowed three-year cognitive decline by about half in higher-risk older adults (though it was flat in the healthy overall group, so treat it as “real in the at-risk, unproven for everyone”).³⁸ Why hearing loss precedes decline is still argued, the effort of straining to hear stealing resources, social withdrawal, plain shared aging, but the fact that treating the ears changed the brain’s trajectory argues it’s at least partly causal.

Sleep apnea belongs on the watch-list too, though more weakly: it’s clearly associated with worse cognition, but CPAP’s cognitive payoff in trials is modest and patchy.³⁹ Exercise is the gentlest real one: aerobic exercise gives small but genuine cognitive gains, and in one trial a year of it measurably grew the hippocampus.⁴⁰ And head trauma is dose-dependent and probably causal, the more hits, the worse the outcome.⁴¹

If you want to do something with all this rather than just collect it, that list is the whole actionable yield. The rest is interesting. This part is useful.

Where intelligence stops

For all of that, the thing IQ conspicuously fails to deliver is the thing everyone actually wants.

It barely touches happiness. The popular claim is “no link,” and that’s not quite right either, the truth is a small, mostly indirect one: in nationally representative data, lower IQ does go with lower happiness, but adjust for income, health, and mental health and about half of it disappears.⁴² So intelligence buys happiness only by way of the things it buys, and even then weakly. It does not, on its own, make you content. That’s worth sitting with.

It also can’t be installed by the side door. The “music lessons make kids smarter” idea has one good randomized study showing a small gain, and a pile of meta-analyses showing the far-transfer to general intelligence basically vanishes once you control for study quality.⁴³ Enjoy music for music.

And the most freeing finding in the whole pile: for any specific skill you care about, practice eats raw IQ for breakfast. Deliberate practice explains a real chunk of performance, but only a chunk, around 19% on average, less in school and professional work.⁴⁴ So practice is powerful but not the whole story (the people who oversold “10,000 hours” oversold it). Still, the practical upshot stands: the person who has actually drilled the thing for years beats the person with ten more IQ points who hasn’t. You don’t need a higher IQ to get good at what you care about. You need reps and honest feedback.

How to read a correlation

So what’s the takeaway from a list this long and this real?

Almost everything correlates with intelligence. Brain size, speed, height, eyes, faith, trust, patience, income, crime, lifespan. And almost none of it is what it looks like.

The effects are small, mostly r = 0.2 to 0.4, a few percent of the variance, invisible in any one person. They shrink when you control for the obvious confound (height within families, breastfeeding for maternal IQ, happiness for income). They run the surprising direction (education causes the nearsightedness, not the reverse). And the satisfying causal stories, analytic thinking dissolving faith, the Dunning-Kruger curve, are the ones most likely to be illusions.

A handful are genuinely causal, and you can act on those: pull the lead, treat the pressure and the hearing and the iodine, move your body, protect your head.

The rest is a lesson in reading. “Predicts” is not “causes.” A real correlation can be tiny, confounded, backwards, or a shared symptom of something deeper, like a well-built nervous system reading out as both quick wits and a long life. Intelligence is one of the most powerful single variables we can measure about a person. It still explains almost nothing about the person in front of you.

The smart move was never having a high IQ. It’s knowing what a correlation is worth.

References

Footnotes

1. Charles Spearman, “‘General Intelligence,’ Objectively Determined and Measured” (1904), American Journal of Psychology. The origin of g and the positive manifold. JSTOR ↩

2. Ian J. Deary, Lars Penke, Wendy Johnson, “The neuroscience of human intelligence differences,” Nature Reviews Neuroscience (2010). Modern review of g and the positive manifold. Open PDF ↩

3. Jakob Pietschnig et al., “Meta-analysis of associations between human brain volume and intelligence differences,” Neuroscience & Biobehavioral Reviews (2015). The meta-analytic estimate is r ≈ 0.24 across 88 studies, with a note on publication bias inflating older estimates. Open PDF. Earlier, larger estimate: Michael McDaniel, “Big-brained people are smarter,” Intelligence (2005). Record. For the counter-position that the higher estimate largely survives once IQ-measurement quality is corrected: Gignac & Bates, “Brain volume and intelligence: The moderating role of intelligence measurement quality,” Intelligence (2017). Record ↩

4. Rex E. Jung, Richard J. Haier, “The Parieto-Frontal Integration Theory (P-FIT) of intelligence,” Behavioral and Brain Sciences (2007). Intelligence as a distributed fronto-parietal network rather than gross volume. PDF ↩

5. Leah Sheppard, Philip Vernon, “Intelligence and speed of information-processing: A review of 50 years of research,” Personality and Individual Differences 44(3) (2008). Reaction time and g correlate about -0.26 across 172 studies, N ≈ 53,500. Open PDF. Inspection time: Grudnik & Kranzler, “Meta-analysis of the relationship between intelligence and inspection time,” Intelligence (2001). Record ↩

6. Lars Penke et al., “Brain-wide white matter tract integrity is associated with information processing speed and general intelligence,” Molecular Psychiatry (2012). The white-matter and intelligence link is fully mediated by processing speed. Nature ↩

7. Matthew Keller et al., “The Genetic Correlation between Height and IQ: Shared Genes or Assortative Mating?” PLOS Genetics (2013). The link is almost entirely genetic, split between pleiotropy and assortative mating. Open. Riccardo Marioni et al., “Common Genetic Variants Explain the Majority of the Correlation Between Height and Intelligence,” Behavior Genetics (2014). SNPs explain about 71% of the correlation. PMC ↩ ↩²

8. Edward Mountjoy et al., “Education and myopia: assessing the direction of causality by mendelian randomisation,” BMJ (2018). About -0.27 dioptres per year of education, with the reverse direction null. PMC ↩

9. Mingguang He et al., “Effect of time spent outdoors at school on the development of myopia among children in China: a randomized clinical trial,” JAMA (2015). An extra 40 minutes a day outdoors cut three-year myopia incidence from 39.5% to 30.4%. JAMA ↩

10. Helmut Leder, Michael Forster, Gernot Gerger, “The glasses stereotype revisited,” Swiss Journal of Psychology (2011). Glasses raised perceived intelligence in a Western sample. Open summary. Contrasting non-Western result: AlRyalat et al., Cureus (2022). PMC ↩

11. Petter Kristensen, Tor Bjerkedal, “Explaining the relation between birth order and intelligence,” Science (2007). Norwegian conscripts; the effect tracks social rank, not birth rank (second-borns raised as eldest scored like firstborns). Record. Julia Rohrer, Boris Egloff, Stefan Schmukle, “Examining the effects of birth order on personality,” PNAS (2015). About 1.5 IQ points per birth position. PMC ↩ ↩²

12. Rohrer, Egloff, Schmukle, PNAS (2015). Confirms the small IQ effect but finds essentially no birth-order effect on the Big Five personality traits. PMC ↩

13. Joseph Price, “Parent-Child Quality Time: Does Birth Order Matter?” Journal of Human Resources (2008). Firstborns receive 20 to 30 more minutes of one-on-one parental time per day (American Time Use Survey). Author copy ↩

14. Bruce Lanphear et al., “Low-Level Environmental Lead Exposure and Children’s Intellectual Function: An International Pooled Analysis,” Environmental Health Perspectives (2005). About 6.2 IQ points lost from <1 to 10 µg/dL, with steeper harm at the lowest doses. PMC ↩

15. “Cognitive Impairment Induced by Lead Exposure during Lifespan: Mechanisms of Lead Neurotoxicity,” Toxics (2021). Lead acts as an NMDA-receptor antagonist that blocks long-term potentiation. Open ↩

16. James Feyrer, Dimitra Politi, David Weil, “The Cognitive Effects of Micronutrient Deficiency: Evidence from Salt Iodization in the United States,” Journal of the European Economic Association (2017). About a one-standard-deviation gain in the most-deficient areas. NBER working paper. Meta-analysis: Bougma et al., Nutrients (2013). PMC ↩

17. Geoff Der, G. David Batty, Ian Deary, “Effect of breast feeding on intelligence in children: prospective study, sibling pairs analysis, and meta-analysis,” BMJ (2006). Adjusting for maternal IQ removes about 71 to 75% of the apparent breastfeeding effect. PMC ↩

18. Jeromy Anglim et al., “Personality and Intelligence: A Meta-Analysis,” Psychological Bulletin (2022). Openness ≈ 0.20, the only Big Five trait that meaningfully correlates with intelligence. Preprint. Corroborated by Stanek & Otto, PNAS (2023). Open ↩

19. Anglim et al., Psychological Bulletin (2022). Conscientiousness sits near zero with intelligence. Preprint. On the small negative being an education/facet artifact: Rammstedt, Danner, Martin, Journal of Research in Personality (2016). Record ↩ ↩²

20. Sophie von Stumm, Benedikt Hell, Tomas Chamorro-Premuzic, “The Hungry Mind: Intellectual Curiosity Is the Third Pillar of Academic Performance,” Perspectives on Psychological Science (2011). Curiosity predicts attainment about as well as effort. Record ↩

21. Gordon Pennycook et al., “On the reception and detection of pseudo-profound bullshit,” Judgment and Decision Making (2015). Bullshit receptivity correlates negatively with analytic ability; the effect is largely receptivity, not true sensitivity. Open ↩

22. Noah Shamosh, Jeremy Gray, “Delay discounting and intelligence: A meta-analysis,” Intelligence (2008). The correlation is r ≈ -0.23 (higher IQ, less discounting). Meta-analysis. Working-memory mechanism from the companion study, Shamosh et al., Psychological Science (2008): PDF ↩

23. Emanuel Jauk et al., “The relationship between intelligence and creativity: New support for the threshold hypothesis,” Intelligence (2013). The threshold is measure-dependent and absent for creative achievement. PMC. Meta-analysis rejecting the threshold: Kim, “Can only intelligent people be creative?” (2005). Record ↩

24. Miron Zuckerman, Jordan Silberman, Judith Hall, “The Relation Between Intelligence and Religiosity: A Meta-Analysis,” Personality and Social Psychology Review (2013). About r = -0.2 for strength of belief. PDF. On the effect being weaker and declining: Webster & Duffy, Intelligence (2016). Record ↩

25. Will Gervais, Ara Norenzayan, “Analytic Thinking Promotes Religious Disbelief,” Science (2012). Science. The priming result failed to replicate: Sanchez et al., PLOS ONE (2017), N = 941, effect ≈ 0 Open; and the Social Sciences Replication Project, Camerer et al., Nature Human Behaviour (2018). Open PDF ↩

26. Noah Carl, Francesco Billari, “Generalized Trust and Intelligence in the United States,” PLOS ONE (2014). Verbal ability strongly predicts generalized trust. Open. UK cohort replication: Sturgis, Read, Allum, Intelligence (2010). Repository ↩

27. Jonas De keersmaecker et al., “Disliked but Free to Speak: Cognitive Ability Is Related to Supporting Freedom of Speech for Groups Across the Ideological Spectrum,” Social Psychological and Personality Science (2021). Mediated by intellectual humility. Open copy ↩

28. Ian Deary, G. David Batty, Catharine Gale, “Bright Children Become Enlightened Adults,” Psychological Science (2008). Higher childhood IQ predicts socially liberal, anti-authoritarian attitudes (not party or economic ideology), mediated by education. Manuscript. Replication: Schoon et al., Intelligence (2010). Record ↩

29. Justin Kruger, David Dunning, “Unskilled and Unaware of It,” Journal of Personality and Social Psychology (1999). The effect is overestimation of one’s own absolute ability, not “the unskilled think they beat the skilled.” DOI ↩

30. Gilles Gignac, Marcin Zajenkowski, “The Dunning-Kruger effect is (mostly) a statistical artefact,” Intelligence (2020). Record. The pattern reproduces from random numbers: Nuhfer et al., Numeracy (2017) Open; Magnus & Peresetsky, Frontiers in Psychology (2022) Open ↩

31. Tarmo Strenze, “Intelligence and socioeconomic success: A meta-analytic review of longitudinal research,” Intelligence (2007). IQ correlations: education ~0.56, occupation ~0.43, income ~0.2. Open PDF ↩

32. Jay Zagorsky, “Do you have to be smart to be rich? The impact of IQ on wealth, income and financial distress,” Intelligence (2007). IQ predicts income but not net worth, and financial distress is not monotonic in IQ. Open PDF ↩

33. Ttofi, Farrington, Piquero et al., “Intelligence as a protective factor against offending: A meta-analytic review of prospective longitudinal studies,” Journal of Criminal Justice (2016). Intelligence is protective, especially in high-risk youth. Record. On the link not being merely a detection artifact: Yun & Lee, Youth Violence and Juvenile Justice (2013). PDF ↩

34. Frank Schmidt, John Hunter, “The validity and utility of selection methods in personnel psychology,” Psychological Bulletin (1998). The classic ~0.51 figure. DOI. Revised downward to ~0.31 by Sackett, Zhang, Berry, Lievens, Journal of Applied Psychology (2022). Author PDF ↩

35. Catherine Calvin et al., “Childhood intelligence in relation to major causes of death in 68 year follow-up,” BMJ (2017). About 20% lower all-cause mortality per standard deviation of childhood IQ. PMC ↩

36. Ian Deary, “Looking for ‘system integrity’ in cognitive epidemiology,” Gerontology (2012). The hypothesis that IQ marks underlying biological soundness. Open. Reaction time predicting the same causes of death as IQ: Der & Deary, Intelligence (2018). PMC ↩ ↩²

37. SPRINT MIND Research Group (Williamson et al.), “Effect of Intensive vs Standard Blood Pressure Control on Probable Dementia,” JAMA (2019). Intensive control cut mild cognitive impairment (HR 0.81); the dementia result narrowly missed significance. PMC ↩

38. Gill Livingston et al., “Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission,” The Lancet (2024). Hearing loss is the largest modifiable midlife risk factor. Accepted version. The ACHIEVE randomized trial (Lin et al., The Lancet, 2023) slowed decline by about 48% in higher-risk elders. Related analysis. Cohort link: Lin et al., Archives of Neurology (2011). PMC ↩ ↩²

39. Omonigho Bubu et al., “Obstructive sleep apnea, cognition and Alzheimer’s disease: a systematic review,” Sleep Medicine Reviews (2020). OSA is associated with cognitive deficits; CPAP’s cognitive benefit in trials is modest and mixed. PMC ↩

40. Patrick Smith et al., “Aerobic Exercise and Neurocognitive Performance: a Meta-Analytic Review of Randomized Controlled Trials,” Psychosomatic Medicine (2010). Small but real gains. PMC. Hippocampal growth: Erickson et al., PNAS (2011). Open ↩

41. “The Association Between Traumatic Brain Injury and the Risk of Cognitive Decline: An Umbrella Systematic Review and Meta-Analysis,” Brain Sciences (2024). Dementia odds ratio ≈ 1.8, dose-dependent. PMC ↩

42. Afia Ali et al., “The relationship between happiness and intelligent quotient: the contribution of socio-economic and clinical factors,” Psychological Medicine (2013). Lower IQ goes with lower happiness, about half of it mediated by income, health, and mental health. Record. Small positive raw association: Nikolaev & McGee, Intelligence (2016). Working paper ↩

43. E. Glenn Schellenberg, “Music lessons enhance IQ,” Psychological Science (2004). A small RCT gain. Open PDF. Far-transfer to general cognition does not survive design-quality controls: Sala & Gobet, Memory & Cognition (2020). PMC ↩

44. Brooke Macnamara, David Hambrick, Frederick Oswald, “Deliberate Practice and Performance in Music, Games, Sports, Education, and Professions: A Meta-Analysis,” Psychological Science (2014). Practice explains about 19% of performance variance on average, far less in education and the professions. Open PDF ↩