The premise of the “what highly intelligent people notice about others” genre is that raw brainpower comes with a kind of social X-ray vision. Smart people, the story goes, read the room, clock the microexpression, and see through you in a way the rest of us cannot. It is a flattering idea, and it has the shape of common sense. The research on whether intelligence actually tracks the ability to read other people is less flattering and more interesting: the link is small, and on some measures it is not there at all.
This is a Brain and Behavior piece about person perception. The short version is that being smart and being socially perceptive are closer to separate abilities than the genre assumes.
What the meta-analysis actually found
The most direct evidence comes from a 2011 meta-analysis by Nora Murphy and Judith Hall, published in the journal Intelligence and titled Intelligence and interpersonal sensitivity. They pooled 38 independent samples, covering nearly 3,000 people, that had measured both general intelligence and how accurately participants could decode other people’s emotions and intentions from cues like faces, voices, and body language.
The two were correlated, but weakly. The overall relationship came out at about r equals 0.19. That is a real, statistically reliable effect, and it is also small. A correlation of that size means intelligence accounts for only a few percent of the differences between people in how well they read others. Knowing someone scores high on an IQ test tells you very little about whether they will accurately catch what a colleague is feeling.
So there is a grain of truth in the premise. Smarter people are, on average, very slightly better at reading others. The genre takes that faint signal and inflates it into a defining trait.
A classic test found no link at all
The picture gets sharper with a specific, well-known instrument. In 2001, Simon Baron-Cohen and colleagues at Cambridge published a revised version of the Reading the Mind in the Eyes Test in the Journal of Child Psychology and Psychiatry. Participants look at 36 photographs cropped to just the eye region of a face and choose, from four options, the word that best describes what the person is thinking or feeling.
It is a narrow task, capturing only the first step of reading a mind, attaching a label like “cautious” or “regretful” to an expression. But it is a genuine measure of a social skill, and it reliably separates groups known to differ in social cognition. In that study, adults with autism or Asperger syndrome scored lower than other adults, which is part of what the test was designed to detect.
Here is the relevant number. When the researchers checked whether scores on the Eyes Test tracked IQ, they found no significant correlation, reporting a value of about 0.09. The authors concluded plainly that performance was independent of general, nonsocial intelligence. On this particular measure of reading others, being smart bought nothing.
Why the two keep getting confused
General intelligence and the ability to read people feel like they should be the same thing, which is probably why the genre sells. Both seem to be about being perceptive. But psychologists treat them as distinct, and the data support the separation.
Part of the confusion comes from a halo effect. Once we decide someone is intelligent, we tend to credit them with other good qualities, including insight into people, whether or not it is there. A person who is articulate and quick can seem to understand others better simply because they describe their impressions more fluently, even if those impressions are no more accurate.
It is worth noting how much of the genre leans on a related and equally shaky idea, that perceptive people spot fleeting microexpressions and thereby see the truth others hide. The ability of people, including trained ones, to detect lies from such cues is generally poor in controlled studies, hovering near chance. If even deliberate training does not reliably confer this power, there is little reason to think a high IQ hands it over for free. The mythology of the person who reads everyone tends to outrun what any measured skill can do.
There is also a large, separate research literature on emotional intelligence, the proposed ability to perceive and manage emotions. It overlaps only partly with cognitive ability, and its status is contested, with ongoing debate about how well it can be measured and how much it adds beyond ordinary personality traits. The honest summary is that the skill of reading people is real, but it does not ride along with IQ the way the headlines imply.
What these tests can and cannot show
None of this proves that intelligence and social perception are entirely unrelated. The meta-analysis found a small positive link, not a zero, and pooling many studies is a better guide than any single result. The Eyes Test finding of no correlation came from particular samples, including a group of Cambridge students whose narrow, high IQ range makes any correlation with IQ harder to detect.
The measures have limits of their own. The Eyes Test uses static photographs of actors, not moving people in real conversations, and it has drawn methodological criticism over the years. Decoding accuracy tasks capture a slice of social skill, not the whole of it. What none of this supports is the opposite overcorrection, the notion that intelligent people are somehow worse at reading others. The evidence points to a weak positive link, not a negative one.
What it does undercut is the specific promise of the genre. There is no good evidence for a package of special perceptions that comes bundled with a high IQ. The person who reads a room well may be intelligent, but the reading is a distinct skill, unevenly distributed among clever and ordinary people alike, and better predicted by things like attention to others and practice than by anything a reasoning test measures.
The most that can be said is modest and worth saying anyway. Intelligence helps a little with reading people, and only a little, and the rest of that ability has to be accounted for some other way.