In the early 1900s, the French government asked a psychologist named Alfred Binet for a practical tool: a way to identify which schoolchildren needed extra help before they fell behind.
The test Binet built, according to a standard account of the test’s history in OpenStax’s psychology textbook, “included many verbally based tasks.” A Stanford professor named Louis Terman adapted it in 1916 into what became the Stanford-Binet scale, and by 1939 David Wechsler had built his own version, later known as the WAIS, drawing on decades of earlier verbal and nonverbal subtests. More than a hundred years later, a version of that same idea, asking someone to define words they know, is still sitting inside modern intelligence tests, largely unchanged in purpose if not in form. Test designers have swapped out plenty of other tasks over that century, dropping ones that turned out to measure something narrower than intended and adding new ones as theories of intelligence evolved.
The vocabulary question kept its seat through all of it.
Why word knowledge became the anchor
Vocabulary earned that staying power for a fairly unglamorous reason: it works. Among the various subtests that have come and gone across a century of test design, tasks that ask someone to define or recognize words tend to correlate unusually well with performance across the whole test, and with other measures of reasoning and problem-solving besides. The value has little to do with the words themselves being impressive. Building that vocabulary requires the same underlying machinery an intelligence test is trying to measure in the first place: noticing patterns, making inferences from context, and holding onto new information once it’s been encountered.
Nobody sits down and memorizes a dictionary. Most words get absorbed sideways, guessed at from context in a book or conversation, tested against later encounters, and either kept or quietly discarded. Someone hears “attrition” used about a company losing staff, files away a rough sense of what it means, and refines that guess a little more the next three or four times the word turns up somewhere else. A vocabulary that keeps growing is really a record of how many times that quiet inference-and-retention process has succeeded, which is why it ends up correlating with broader thinking skill instead of just reflecting how many books someone happened to read.
That distinction matters, because it’s easy to mistake vocabulary size for something closer to trivia, a stockpile of impressive words with no real bearing on how well someone reasons. The research points the other way. A person who can define an unusual word usually acquired it the same way they solve an unfamiliar problem: by testing a hypothesis against evidence, revising it, and holding onto the version that keeps working. The word is just the visible residue of a much less visible skill.
The part that surprises people: it doesn’t stop in your twenties
Most cognitive skills follow a familiar arc: rise through childhood, peak in early adulthood, decline gradually after that. Vocabulary breaks the pattern. A large study of nearly 50,000 people, run by researchers at MIT and Massachusetts General Hospital and reported in an Association for Psychological Science release, found that vocabulary knowledge kept climbing well past the point where raw processing speed and working memory had already started to fade, not leveling off until people reached their late 60s or early 70s. As co-author Joshua Hartshorne put it, “at any given age, you’re getting better at some things, you’re getting worse at some other things, and you’re at a plateau at some other things.”
The timeline has also been drifting later across generations. According to reporting on the same research in the Harvard Gazette, adults tested between 1974 and 1987 tended to peak on vocabulary measures in their early 40s, adults tested between 1988 and 1997 peaked closer to 50, and adults tested between 1998 and 2012 didn’t peak until around 65. More education, more reading-heavy jobs, and more everyday exposure to unfamiliar language all plausibly push that ceiling higher with each generation, though the exact mix of causes is still being worked out. Either way, someone whose vocabulary is still expanding at 55 isn’t an outlier fighting the normal aging curve. On the current data, they’re closer to the norm than someone whose word knowledge stalled out at 25. That’s a genuinely unusual shape for a cognitive skill to have, and it’s part of why researchers keep coming back to vocabulary as a separate case worth tracking on its own rather than folding it into the general decline story that fits most other mental abilities.
What that says about how a person thinks
None of this means a large vocabulary makes someone smarter in some general, superior sense, and it certainly doesn’t mean using complicated words in conversation signals anything at all. The correlation runs through the underlying process, not the display of it. Someone whose vocabulary keeps expanding into middle age is someone whose brain is still routinely doing the small, unglamorous work that built the vocabulary in the first place: catching an unfamiliar term in context, forming a rough guess at its meaning, and testing that guess against the next few times the word shows up.
That’s a strange kind of intelligence measure, precisely because it isn’t measuring a peak. Most of the traits people associate with sharp thinking, quick reactions, fast recall, rapid-fire problem solving, are the ones that fade earliest. Vocabulary tracks something slower and more cumulative: the habit of staying curious enough about language to keep letting new words in, decade after decade, long after the twenties have ended and the rest of the cognitive profile has already started trading speed for something else. A century of test design keeps landing back on the same quiet measure for a reason.
