What is crystallized intelligence?

Short answer: Crystallized intelligence (Gc) is the ability to use knowledge and skills you have learned through education, experience and your culture, such as vocabulary and general facts. Raymond Cattell first described it in 1943. On average it keeps growing until about the 60s, a pattern Tucker-Drob and colleagues (2022) describe as well known; estimates of its peak run from the late 40s to the early 70s, depending on the study and the measure.

Crystallized intelligence at a glance

• What it is: what you know, and how well you can use it. The CHC model also calls it comprehension-knowledge.
• Typical examples: vocabulary, analogies and general knowledge, as Bajpai and colleagues list them (2022), plus the mechanics of language.
• Where it comes from: schooling, reading, work and the culture you grow up in.
• When it peaks: later than fluid reasoning, somewhere from about 60 into the early 70s in most of the studies below (see the age section).
• How tests measure it: with vocabulary and general knowledge questions, such as Vocabulary and Information on the Wechsler scales.
• Its partner: fluid intelligence (Gf), the ability to reason through new problems. Matrix puzzles measure mostly fluid reasoning.

What is crystallized intelligence?

A 2024 paper in the Journal of Intelligence by Scherrer, Breit and Preckel, citing Schneider and McGrew (2018), puts it this way: "Crystallized intelligence (Gc) refers to the ability to understand and utilize culturally valued knowledge, encompassing information and skills acquired through experience, education, and acculturation".

Two parts matter: the knowledge itself (words, facts, learned procedures) and the ability to use it. Gc covers knowing what a word means and being able to apply it or reason with it.

Tucker-Drob and colleagues draw the line between the two main kinds of ability in Science Advances (2022). Fluid abilities "require effortful processing at the time of assessment", while crystallized abilities "require the retrieval and application of knowledge." A fluid task asks you to work something out on the spot. A crystallized task asks what you already know.

Where Gc sits in the CHC model

CHC theory (short for Cattell-Horn-Carroll) joined the Cattell-Horn Gf-Gc theory with Carroll's three-stratum model, as a 2023 retrospective by McGrew and colleagues explains. General intelligence (g) sits at the top. Scherrer, Breit and Preckel (2024) describe Gf and Gc as "broad cognitive abilities positioned at Stratum 2, hierarchically below a g-factor of general intelligence at Stratum 1."

Is it the same as knowledge or memory?

Close to knowledge, but not the same as memory.

CHC separates general knowledge (Gc) from domain-specific knowledge (Gkn), which Rusche and Ziegler (2023) describe as knowledge "not all members of a society are expected to have". A tax lawyer's grasp of tax law is Gkn. Their vocabulary is Gc.

Memory abilities are separate CHC factors, such as working memory and retrieval fluency (Biesok and colleagues, 2024). Gc is what you know. The memory factors describe how well you store and retrieve it. The tasks overlap in practice: Pearson's WAIS-IV mapping counts Vocabulary and Information toward both crystallized knowledge and long-term storage and retrieval.

Who came up with it: 1943 or 1963?

You will see both dates, and both are partly right.

• 1943, the first statement. Raymond Cattell introduced fluid and crystallized ability in "The measurement of adult intelligence" in Psychological Bulletin. A 2016 history by Brown in Frontiers in Human Neuroscience quotes his wording: "Adult mental capacity is of two kinds, the chief characteristics of which may be best connoted by the use of the terms “fluid” and “crystallized.”"
• 1963, the first test. Cattell's paper "Theory of fluid and crystallized intelligence: A critical experiment" in the Journal of Educational Psychology tested it on 277 children in the 7th and 8th grades, as Brown (2016) reports.
• 1966 and 1967, the full theory. Cattell's student John Horn developed the theory with him in two papers, one in the Journal of Educational Psychology (1966) and one in Acta Psychologica (1967).

Brown (2016) adds that Cattell himself dated the idea to 1940, 1941 or 1942 in different papers, and argues from archival letters that Cattell adopted psychologist Donald Hebb's ideas of "intelligence A and B" and renamed them. Cattell's 1943 description, as Brown quotes it, still works: crystallized ability "consists of discriminatory habits long established in a particular field, originally through the operation of fluid ability". Something first worked out with effort becomes something you simply know.

Examples of crystallized intelligence

Bajpai and colleagues (2022) write that crystallized ability "encompasses vocabulary, analogies, and general knowledge which are learned through experience and education". Tucker-Drob and colleagues (2022) add "domain-specific skills."

In everyday life:

• Vocabulary. Knowing what "frugal" means and using it correctly.
• General information. Facts most adults in your culture are expected to know, such as why the seasons change.
• Word analogies. Seeing that "author is to book as composer is to symphony".
• The mechanics of language. Wikipedia's summary lists "the mechanics of language" among crystallized tasks; grammar and spelling are everyday examples. The NIH Toolbox counts oral reading recognition toward crystallized intelligence (Doucet and colleagues, 2022).
• Learned procedures. Using the algebra you learned in school to solve a word problem.
• Skills from training or a job. A nurse reading a chart. CHC treats deep specialist knowledge as a separate ability (see above).

Horn's classic example, as Wikipedia summarizes it, is a word puzzle about shoes worn by 100 hospital patients. Someone who knows high school algebra can set up an equation, which is the crystallized route. A child without algebra can still reason it through with fluid ability.

Where crystallized intelligence comes from

Gc is learned. Wilson and colleagues write in the Journal of Intelligence (2023) that "The language-based, culturally shared knowledge that is tested in Gc is learned through education and one's environment".

Cattell's investment theory

Cattell proposed that people build Gc by investing their fluid ability in learning. Scherrer, Breit and Preckel (2024) summarize Cattell (1987): "an individual's current level of Gc is the cumulative result of several years of Gf investment." The same paper suggests curious, learning-seeking traits may explain who takes up chances to learn, "such as reading challenging literature".

This is a theory, and it has a challenger. Kievit and colleagues tested vocabulary and matrix reasoning in 785 adolescents and young adults and reported in Psychological Science (2017) that a mutualism model fit best: abilities such as vocabulary and reasoning help each other grow.

Schooling

Education raises test scores. Ritchie and Tucker-Drob's meta-analysis in Psychological Science (2018) pooled 142 effect sizes from 42 data sets with over 600,000 participants. They found "beneficial effects of education on cognitive abilities of approximately 1 to 5 IQ points for an additional year of education." A 2020 summary by Stankov and Lee puts the average at 3.4 IQ points per extra year.

A natural experiment in Norway tested the same question. Writing in PNAS in 2012, Brinch and Galloway used a 1960s reform that raised compulsory schooling from 7 to 9 years to estimate the effect of education on IQ.

One caution. These studies measured overall IQ and broad abilities rather than Gc alone. Ritchie and Tucker-Drob found effects "on all broad categories of cognitive ability studied", so it would be wrong to say schooling raises only crystallized intelligence.

When does crystallized intelligence peak?

Later than fluid reasoning. The exact age depends on the study, because studies measure different things in different ways.

• Late 60s or early 70s, for vocabulary. Hartshorne and Germine (2015, in Psychological Science) analyzed data from 48,537 online participants plus normative IQ and memory test data. They found some abilities "do not peak until subjects reach their 40s or later." MIT News reported on March 6, 2015, that in their online data, vocabulary peaked in the late 60s or early 70s.
• Late 40s, in the Wechsler norms. The same MIT News report says data from the Wechsler tests "suggested that vocabulary peaks in the late 40s".
• Gains until about the 60s, the established pattern. Tucker-Drob and colleagues (Science Advances, 2022) open their discussion with what they call well known: on population averages, "crystallized abilities exhibit age-related gains until approximately the seventh decade of life", which is the 60s. They present this as background, not as their own result.
• About 60, with education controlled. Kaufman, describing the early Cattell-Horn distinction in cross-sectional data, wrote in the Journal of Intelligence (2021) that learned facts and skills "increase to about age 60 (as long as education is controlled) before beginning its inevitable decline."

A fair summary: on average, Gc keeps growing through middle age, and most of these estimates put its peak somewhere from about 60 into the early 70s. The studies differ in what they measure (vocabulary alone or a broader Gc score), in their data (online test-takers or test norms) and in whether they adjust for education.

Does crystallized intelligence decline?

Eventually, on average, but much later than fluid reasoning. Kaufman (2021) writes that "Even the maintained ability of crystallized knowledge has been shown to decline steadily after about age 70 or 75" in cross-sectional studies controlled for education.

Researchers have argued about this for decades. The University of Alberta Dictionary of Cognitive Science records an older debate: Horn (1970) held that Gc stays relatively stable and may even increase, while Belsky (1990) argued that it declines with age.

Averages also hide wide differences. Tucker-Drob and colleagues (2022), using two large longitudinal studies, found that people whose fluid abilities fell slowly showed some of the largest gains in crystallized ability. People whose fluid abilities fell steeply showed "very little gain, or even decline" in crystallized ability over the same years.

If you are worried about memory or thinking changes in yourself or a relative, a doctor or psychologist is the right person to ask. A test score or an article cannot tell you what is happening.

How IQ tests measure crystallized intelligence

Gc is usually measured with knowledge questions. Scherrer, Breit and Preckel (2024) note that "Gc is typically measured through factual knowledge questions." In clinical IQ tests, that means verbal subtests.

• WAIS-IV (adults). The Verbal Comprehension Index is built from Similarities, Vocabulary and Information, as Chen and colleagues describe it in Scientific Reports (2019). They treated the index as the measure of Gc. Pearson's own mapping of the WAIS-IV to CHC terms lists Vocabulary, Information and Comprehension as "crystallized knowledge" and puts Similarities under fluid reasoning. Pearson also says the WAIS-IV full-scale score "de-emphasizes crystallized knowledge" compared with the WAIS-III, with Comprehension now supplemental. Our guide to the WAIS IQ test explains its index scores.
• WAIS-5 (adults). Pearson lists "New crystallized and fluid expanded index scores", including a "verbal (expanded crystallized)" index.
• WISC-V (children). Pearson's WISC-V FAQ gives Gc as the CHC name for the Verbal Comprehension index, which Watkins and colleagues (2023) describe as built from Similarities and Vocabulary.
• Woodcock-Johnson IV. Bulut and colleagues (2021) measured Gc with the Oral Vocabulary and General Information subtests.
• Stanford-Binet 5. It has a Knowledge factor among its five, and testing begins with two routing tests, in Nonverbal Fluid Reasoning and in Verbal Knowledge, according to Texas Statewide Leadership for Autism Training. That page does not label the Knowledge factor as Gc. Read more in our guide to the Stanford-Binet test.

By Pearson's mapping, Vocabulary, Information and Comprehension are the WAIS-IV's crystallized-knowledge subtests. Similarities is borderline: it sits in the Wechsler verbal index, but Pearson's mapping to CHC terms counts it as fluid reasoning.

For a wider view of what a full test covers, see what an IQ test measures.

Crystallized vs fluid intelligence, in brief

Fluid intelligence is reasoning on problems you have not met before. Crystallized intelligence is using what you have already learned.

The two are linked. In adulthood, Tucker-Drob and colleagues (2022) found that "rates of change are strongly correlated across fluid and crystallized abilities."

They part ways with age. The same team writes that "On average, fluid abilities decline throughout adulthood, whereas crystallized abilities show gains into old age." So an older adult may find new puzzles harder than they once did while still adding words and facts.

Our full guide to fluid intelligence covers how it is measured and how it changes with age.

Can you increase crystallized intelligence?

Yes. By definition, Gc is learned, so learning adds to it.

The strongest evidence is for education. Ritchie and Tucker-Drob (2018) conclude: "Education appears to be the most consistent, robust, and durable method yet to be identified for raising intelligence." Their meta-analysis found the effects persisted across the life span.

Outside formal schooling, the evidence is thinner. Investment theory suggests habits like reading challenging books feed Gc over the years, but the studies we read do not measure how much they add.

Brain-training apps are a weaker bet. Simons and colleagues reviewed the evidence in Psychological Science in the Public Interest (2016) and found that training improves the trained tasks, with "little evidence that training enhances performance on distantly related tasks or that training improves everyday cognitive performance."

Our guide on whether you can increase your IQ covers the wider evidence.

Why culture and language matter

Gc is culture-bound by design. The CHC definitions describe it as "culturally valued knowledge": the words and facts a particular culture, in a particular language, expects adults to know. That has practical effects:

• A Gc score depends on the test matching your language and schooling. Someone tested in a second language can score lower on vocabulary questions for reasons unrelated to their reasoning. That is our inference from the definition.
• Specialist knowledge tests cannot simply be translated. Rusche and Ziegler (2023) write that specialist knowledge tests "from different cultural circles cannot simply be translated as they need to be culture specific."
• Test makers offer reduced-language routes. Holden and Tanenbaum (2023) write that common cognitive assessments "have attempted to remediate fairness and equity problems by reducing verbal load, reducing unnecessary focus on specific forms of crystallized knowledge". Half of the Stanford-Binet 5 has nonverbal content, according to Texas Statewide Leadership for Autism Training, and the WAIS-5 Nonverbal Index leaves out Verbal Comprehension.

What an online matrix test can and cannot tell you about it

Matrix puzzles are a standard measure of fluid reasoning. Scherrer, Breit and Preckel (2024) write that "Gf is frequently assessed through tests such as Ravens matrices and the culture fair intelligence test", while Gc is measured with factual knowledge questions. The APA task force report "Intelligence: Knowns and Unknowns" describes Raven's as "a nonverbal, untimed test that requires inductive reasoning about perceptual patterns". Our guide to Raven's Progressive Matrices explains the format, and abstract reasoning covers the skill behind it.

So a matrix test can estimate your fluid reasoning on the standard IQ scale. It says little about your vocabulary, general knowledge or career expertise. For both, you need a full clinical test such as the WAIS, and even then the result is a range rather than a single point, and it can differ between tests and sittings. Our guide on whether IQ tests are accurate explains the error that comes with any score.

The WhatIQ test is a matrix reasoning test of 24 picture puzzles in about 15 minutes, built to test "reasoning without words or arithmetic" (whatiq.com, read October 27, 2026), so by our reading of the research above it measures mostly fluid reasoning, not crystallized knowledge. It gives a score on the standard IQ scale with the range it most likely sits in, as an estimate rather than a clinical assessment.

Frequently asked questions

At what age does crystallized intelligence peak?

The studies disagree, so expect a range. Hartshorne and Germine's online data (2015) put the vocabulary peak in the late 60s or early 70s, while the Wechsler norms they also analyzed suggested the late 40s. Tucker-Drob and colleagues (2022) describe average gains until about the 60s as well known, and Kaufman (2021) puts the peak at about 60 when education is controlled.

What is an example of crystallized intelligence?

Knowing what a word means and using it correctly is the classic example. Bajpai and colleagues (2022) list vocabulary, analogies and general knowledge, all learned through experience and education. Other examples are the mechanics of language, such as grammar and spelling, and oral reading recognition, which the NIH Toolbox counts toward crystallized intelligence (Doucet and colleagues, 2022). Using the algebra you learned in school to solve a word problem draws on it too.

Does crystallized intelligence decline with age?

Later than fluid reasoning, on average. Kaufman (2021) reports that it declines steadily after about age 70 or 75 in studies controlled for education. Tucker-Drob and colleagues (2022) found wide differences between people, with some still gaining and others declining over the same years.

How is crystallized intelligence tested?

With knowledge questions, such as vocabulary and general information. On the WAIS-IV, the Vocabulary, Information and Comprehension subtests measure it, according to Pearson's mapping to CHC terms. The WISC-V, Woodcock-Johnson IV and Stanford-Binet 5 include similar vocabulary and knowledge tasks. Matrix puzzles do not measure it directly.

Is crystallized intelligence the same as knowledge or memory?

It is broad, culturally shared knowledge and the ability to use it. CHC treats specialist expertise as a separate ability, domain-specific knowledge, and treats memory as its own set of factors, such as learning efficiency and retrieval fluency. Pearson's WAIS-IV mapping counts its Vocabulary and Information subtests toward both crystallized knowledge and long-term storage and retrieval.

Can you improve crystallized intelligence?

Yes. Because it is learned, it grows with what you learn. For intelligence scores overall, Ritchie and Tucker-Drob's 2018 meta-analysis found about 1 to 5 IQ points per extra year of education, across all the broad abilities studied. Brain-training games show little transfer to everyday thinking, according to Simons and colleagues (2016).