Personality and Intelligence Are More Closely Linked Than We Thought
What two recent papers reveal - and how to read psychology research like a pro!
‘Tis the season once again! As an early Christmas present, I’m re-releasing this earlier paid post as a free post for everyone. It looks at two recent meta-analyses exploring the links between personality and intelligence. It also gives tips on how to read psychology papers, and make sense of contradictory findings. Hope you enjoy it - and Merry Christmas to all who celebrate!
Do intelligent people have different personalities than their less intelligent peers? If so, why? Does personality shape intelligence, or does intelligence shape personality? Or is it a bit of both?
These are some of the questions addressed by two groundbreaking recent papers published in two of the world’s top scientific journals. Their findings challenge the long-standing view that intelligence and personality are largely independent, and suggest instead a much deeper connection between the two domains.
The first paper, by Jeromy Anglim and colleagues, appeared in Psychological Bulletin, and was titled “Personality and Intelligence: A Meta-Analysis.” The second, by Kevin Stanek and Deniz Ones, was published in Proceedings of the National Academy of Sciences (PNAS), and titled “Meta-Analytic Relations Between Personality and Cognitive Ability.”
In this post, I’ll break down both papers, highlighting their agreements, disagreements, and broader implications. I’ll also share some tips on how to read psychology research - especially when different studies come to different conclusions. Without further ado, then, let’s find out why personality and intelligence are more intertwined than we previously thought.
The Classic View: Little Relationship Between Personality and Intelligence
Psychologists have long been interested in whether personality and intelligence are connected. The traditional answer is: not particularly. Most research on the topic has focused on the Big Five personality traits: neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness. Of these core traits, the only one reliably linked to intelligence in past research is openness to experience: Greater openness is modestly correlated with higher intelligence. Some studies also suggest a weak negative link between neuroticism and intelligence: Higher neuroticism is associated with slightly lower intelligence scores. Beyond those meagre findings, however, the consensus has been that personality is largely distinct from cognitive ability.
But is that the final word on the topic? Maybe not. Earlier research often involved small samples, which immediately casts doubt on its conclusions. Additionally, most studies examined intelligence-personality links only at the broad trait level. Traits, however, are composed of sub-traits, including aspects and facets.1 The facets of openness, for instance, include openness to ideas and openness to emotions, while the facets of neuroticism include depression and anxiety. Even if trait-level links are modest, facet-level links could still be substantial.
The Two Papers: Methods and Scope
To investigate these issues, the two meta-analyses - which I’ll refer to as the Psych Bull paper and the PNAS paper - took several steps to improve on past work.
First, they pulled together large collections of earlier studies covering large numbers of participants. The Psych Bull paper included more than 200 studies covering more than 162,000 people. By normal psychology standards, that’s an enormous sample. Amazingly, though, the PNAS paper managed to dwarf it, with 1,235 studies covering more than two million people from more than fifty nations.
Second, both studies examined personality-intelligence links not just at the broad trait level, but also at lower levels of the personality hierarchy. The Psych Bull paper did this for both the Big Five and the HEXACO (which is basically the Big Five plus an honesty-humility trait); the PNAS paper stuck to the Big Five. As well as looking at traits and sub-traits, the Psych Bull paper looked at individual questionnaire items.
Finally, both studies explored not only general intelligence, but also its two main factors: crystallized and fluid intelligence. Crystallized intelligence is acquired knowledge and skills. Fluid intelligence is the ability to process information quickly and solve novel problems, independent of acquired knowledge. (The PNAS paper referred to these as invested vs. non-invested abilities, but they’re basically the same things.) On top of that, the PNAS paper looked at more specific, lower-level cognitive abilities, such as verbal comprehension, spatial scanning, and mathematics knowledge.
What They Found: Three Main Takeaways
Despite their different samples and methods, the major findings of the two studies were essentially identical. Here are three main takeaways common to both.
1. Openness and neuroticism are the two traits most closely linked to intelligence.
Both studies found that openness to experience and neuroticism were notably more closely linked to intelligence than any other traits. Openness was in first place, and was positively correlated with intelligence. Neuroticism was in second place, and was slightly negatively correlated with intelligence. This provides powerful confirmation of earlier research on the topic.
2. Facets show much stronger links to intelligence than do broad traits.
Both studies concluded that personality-intelligence links are much stronger at the sub-trait level than the trait level. Indeed, the Psych Bull paper concluded that personality facets are twice as closely linked to intelligence as personality traits - and that individual items are even more closely linked than that.
3. Crystallized intelligence is more strongly linked to personality than is fluid intelligence.
General intelligence, crystallized intelligence, and fluid intelligence were all linked to various aspects of personality. It was crystallized intelligence, however, that showed the strongest, most pervasive associations.
Because both studies converged on these three conclusions, we can be much more certain of them than we could be otherwise. In light of psychology’s replication crisis, and psychologists’ subsequent crisis of confidence in their findings, this is welcome news. Some results in psychology are genuinely robust.
Digging Deeper: How the Big Five Link with Intelligence
To dig more deeply into the findings, let’s look at the personality-intelligence associations of each Big Five trait in turn.
Openness: The Strongest Predictor
As mentioned, both studies confirmed the earlier finding that the personality trait most closely linked to intelligence is openness to experience. The Psych Bull paper found a correlation of .17 between openness and general intelligence, whereas the PNAS paper found a correlation of .26. (Correlations span from -1 to 1, with -1 indicating a perfect negative correlation, 1 indicating a perfect positive correlation, and 0 indicating no correlation at all.)
The contrasting estimates provide a nice reminder of something you should always keep in mind when reading psychology papers: Don’t take the exact numbers too seriously. Critics of psychology sometimes point out that it’s rather hilarious that psychologists report their findings to two decimal places, as if they have anything like that degree of precision. We should always blur our eyes when reading psych papers, and convert the exact numbers into rough estimates. At the same time, though, we shouldn’t think the rough estimates are worthless. They’re not; they’re just rough. A reasonable conclusion, then, would be that openness and intelligence are correlated at about the .2 level, give or take. In everyday language, that means there’s a correlation, but it’s fairly weak.
There were two other areas of agreement between the two papers. First, openness was more closely linked to crystallized than fluid intelligence (or, if you prefer, to invested than non-invested abilities). Second, the openness facets varied a lot in how closely linked they were to intelligence. Facets related to intellectual engagement (e.g., openness to ideas and intellectual curiosity) were more closely linked than those related to experience (e.g., openness to emotions and openness to aesthetics). Lurking somewhere in the middle were facets such as creativity and unconventionality.
How might we explain the openness-intelligence links? According to the Psych Bull paper, there are two main possibilities. First, openness could accelerate the development of intelligence. People who are intellectually curious spend more time engaged in cognitively demanding tasks, thus building up their intellectual muscles and stockpiling knowledge. The latter is easier to do, which might explain why openness is more closely linked to crystallized intelligence (acquired knowledge) than to fluid intelligence (raw intellectual horsepower).
On the other hand - and perhaps more plausibly - intelligence could help cultivate openness, and especially its intellectual facets. Smarter people are better equipped to undertake cognitively demanding tasks, and obtain more benefits from doing so. As a result, they enjoy these tasks more and develop a stronger interest in them. Smarter people are also better equipped to entertain and evaluate unconventional ideas, and thus may come to enjoy those more too. Their less-smart compatriots, in contrast, may tend to prefer more conventional ways of thinking and acting.
Neuroticism: A Mild Negative Correlation
As we’ve seen, both the Psych Bull and PNAS papers concluded that neuroticism is weakly negatively correlated with intelligence. Indeed, both found a correlation of −.08. This level of convergence is rare in psychology, and is presumably just a fluke. As such, we once again shouldn’t fixate too much on the exact numbers. Our conclusion should be simply that there’s a negative correlation between intelligence and neuroticism, but that it’s weak. Plenty of smart people are neurotic; plenty of not-smart people aren’t.
Both papers also found links between intelligence and various neuroticism facets. Most notably, depression and anxiety were negatively correlated with intelligence in both. Unlike with openness, however, there was no obvious pattern in which facets correlated with intelligence and which didn’t.
As for the “why?” of the neuroticism-intelligence link, the Psych Bull paper has an interesting speculation. Intelligence, the authors note, is a useful resource for dealing with the slings and arrows of everyday life. Not only that, but it’s associated with better occupational and financial outcomes. For that reason, greater intelligence may insulate its owners from the kind of stressful emotions that constitute the trait of neuroticism.
Conscientiousness: A Mixed Bag
The relationship between conscientiousness and intelligence was less straightforward. On the one hand, both papers found little association between trait conscientiousness and intelligence. On the other, when it came to the facet level, the papers diverged more for conscientiousness than for any other trait. The Psych Bull paper found a negative correlation with a handful of facets that reflect a desire for organization, including orderliness. The PNAS paper, in contrast, found modest positive correlations with various conscientiousness facets… including orderliness.
What should we conclude in light of these discrepancies? In my view, the safest conclusion would be that, although there could be something going on at the facet level, even if there is, it’s probably not hugely important - if it were, it’d be easier to pick up a signal, and the results would be less ambiguous. Thus, in lieu of further evidence, we should conclude that conscientiousness facets probably have little meaningful connection with intelligence.
Why, though, are the links between intelligence and conscientiousness so weak? In a sense, it’s surprising, as we have good theoretical reasons to expect such links. Conscientious people are more diligent at school and in life in general. As a result, they stockpile more knowledge and hone their cognitive skills to a greater degree. Why, then, isn’t conscientiousness positively correlated with intelligence?
One possibility, discussed in depth in the Psych Bull paper, is that intelligent people don’t necessarily need to be hugely conscientious, because they can often coast through life on their smarts. Less intelligent people, in contrast, can’t afford to do as much coasting - not if they still want to be successful. As such, those less blessed in the smarts department may cultivate their conscientiousness to a greater degree. This trend may cancel out any effect of conscientiousness in boosting people’s intellectual chops.
Extraversion: Minimal Links
Both the Psych Bull paper and the PNAS paper concluded that trait extraversion is minimally related to intelligence. At the facet level, the Psych Bull paper found that sociability and warmth were negatively correlated with intelligence, whereas assertiveness was positively correlated with it. Meanwhile, the PNAS paper found that the activity facet was positively correlated with intelligence. Beyond that, though, there were few personality-intelligence links, even at the sub-trait level.
Agreeableness: Largely Independent of Intelligence
Last but not least - actually, last and least - agreeableness was largely unrelated to intelligence, at both the trait and sub-trait levels. The PNAS paper did find a weak positive association between compassion and intelligence, and a weak negative association between politeness and intelligence. But these were small islands of connection in a sea of independence. The overall trend was summed up well by the authors of the Psych Bull paper, who observed that “our results suggest that intelligence is relatively unrelated to whether someone is a kind and moral person.”
Final Thoughts: What the Two Papers Tell Us
Taken together, the Psych Bull paper and the PNAS paper radically rewrite our understanding of the links between personality and intelligence. Here are the main conclusions supported by both.
Personality is more closely linked to intelligence than previously assumed. This is largely because earlier research tended to focus on broad personality traits, when the main links between personality and intelligence are found at the facet level. It’s also worth pointing out that, although the links are often fairly modest, they may in fact be larger than they seem. As the Psych Bull paper notes,
observed correlations are likely to be substantially attenuated due to imperfections in self-rated assessments of personality and [the fact] that typical intelligence assessment falls well short of gold standard assessments such as the WAIS [the Wechsler Adult Intelligence Scale, a famous test of IQ].
Openness is the Big Five trait most closely related to intelligence: Smarter people are somewhat more likely to be open, and vice versa. The links are particularly strong for openness facets related to intellectual tasks and pursuits.
Neuroticism is in second place, with a weak negative link to intelligence: Smarter people are slightly less likely to be neurotic, and neurotic people slightly less likely to be intelligent.
Personality is more closely linked to crystallized intelligence than to general or fluid intelligence.
Beyond these specific conclusions, reading the two papers together suggests various lessons about how to make sense of research in psychology.
When two high-quality studies converge on the same conclusions, our confidence in those conclusions should increase exponentially.
When no clear picture emerges from the studies, this doesn’t mean there’s nothing there - but it does mean that, if there is, it’s probably not especially important. We can therefore safely act as if there’s nothing there.
We should always take exact numbers in psychology with a grain of salt, treating them as rough estimates rather than precise measurements. This isn’t necessarily a failure of our methods, but may instead reflect the fact that psychology is a world of fuzzy trends rather than law-like regularities.
In short, personality and intelligence are more intertwined than once believed. The key to unlocking their relationship is to look deeper than broad traits, into the rich and nuanced world of personality facets.
You can access a free copy of the Psych Bull paper here.
You can access a free copy of the PNAS paper here.
Follow me on Twitter/X for more psychology, evolution, and science.
Coming Soon to the Nature-Nurture-Nietzsche Newsletter…
Top 12 Viral Graphs Featured in The Nature-Nurture-Nietzsche Newsletter
12 Things Everyone Should Know About Animal Behavior
Is the Male-Female Distinction Valid?
How You Can Support the Newsletter
If you like what I’m doing with The Nature-Nurture-Nietzsche Newsletter, and would like to support my work, there are several ways you can do it.
Like and Restack: Click the buttons at the top or bottom of the page to boost the post’s visibility on Substack.
Share: Send the post to friends or share it on social media.
Upgrade to a Paid Subscription: A paid subscription will get you:
Full access to all new posts and the archive
Full access to excerpts from my forthcoming book A Billion Years of Sex Differences
Exclusive content such as my “12 Things Everyone Should Know” series, Linkfests, and other regular features
The ability to post comments and engage with the N3 Newsletter community
If you could do any of the above, I’d be hugely grateful. My goal is to make the newsletter a space for free thought and non-politicized psychology, and ultimately to make it a viable alternative career. The support of readers like you is what makes this possible.
Thanks!
Steve
Related Reading From the Archive
Five Traits to Rule Them All
The Big Five is the most popular personality model in psychology. Advocates claim that it captures most of the variation in people’s personalities. If that’s right, though, why are there so many stand-alone personality traits - traits such as grit, empathy, and self-esteem, which are usually considered independent of the Big Five? In a fascinating recent
Sex Differences in Personality
The graph below shows sex differences in the Big 5 personality traits across the adult lifespan. The first five panels show the differences in each of the Big 5 traits in turn; the final panel shows the trajectory of all five traits across adulthood, averaging across the sexes.
re: “our results suggest that intelligence is relatively unrelated to whether someone is a kind and moral person.” I'd be surprised if their results suggested that Agreeableness is related to whether you are a kind and moral person. It's hard to be moral if you are a spineless coward engaged in constant preference falsification. Not that cultivating a hobby of offending others is an improvement ....
Would knowing that you know nothing come under openness? It's not scientific but it seems like a common trait.
I remember reading about the categorisation of words in the original Allport research and it was something like 15,000 words he manually sorted? Long before computers and quicksort.
I think AI research has found that the LLMs see concepts regardless of their language and put them in the same vectors space.
Both the studies you've written about are really interesting, but the multinational one ties into this LLM finding slightly.