Fluid Intelligence vs. Crystallized Intelligence

Our capacity to learn the novel and recall the past is called general intelligence (Cattell, 1963). It is a construct of psychometric investigations of human intelligence and our cognitive abilities.

General intelligence encapsulates correlations among various cognitive tasks which can be categorized into two subdivisions (Cattell, 1971). These are fluid intelligence and crystallized intelligence.

The theory of fluid v. crystallized intelligence simultaneously challenges and extends what was once supposedly the single construct of general intelligence.

Cattell’s Theory of Intelligence

The theory of fluid v. crystallized intelligence was first postulated as a psychometrically based theory by psychologist Raymond B. Cattell in 1963.

Cattell argued that fluid intelligence and crystallized intelligence are two categories of general intelligence.

In his 1987 book, Intelligence: Its Structure, Growth, and Action, Cattell identified one component of general intelligence as fluid. It could apply to any new problem.

He proceeded to identify the other component as a part invested in the areas of crystalized skills. He pointed out that the latter involves knowledge acquisition and crystallized skills, which can be upset individually without impacting others.

The two concepts of fluid intelligence and crystallized intelligence were further developed by Cattell’s former student and cognitive psychologist John Leonard Horn (Horn & Cattell, 1967).

Aim: Horn and Cattell (1967) tested this directly.

Method: They combined scores from a battery of cognitive tests into estimates of fluid and crystallized ability for 297 participants. Participants were split into five age groups spanning 14 to 61 years, and scores were compared across groups.

Results: Fluid intelligence rose through adolescence and levelled off in young adulthood. It then declined in the older groups. Crystallized intelligence, by contrast, kept rising across the whole age range, with no comparable decline.

Conclusion: General intelligence is not developmentally uniform: its fluid and crystallized components follow genuinely different life-course patterns that a single g score cannot capture.

Fluid Intelligence

Fluid intelligence is the capacity to think speedily and reason flexibly to solve new problems without relying on past experience and accumulated knowledge.

Fluid intelligence allows us to perceive and draw inferences about relationships among variables and to conceptualize abstract information, which aids problem-solving. It is correlated with essential skills such as comprehension and learning.

As Raymond Cattell (1967) pointed out, it is a capacity to “perceive relationships independent of previous specific practice or instruction related to those relationships”.

Examples of fluid intelligence include solving puzzles and devising strategies for new problems. It also helps in spotting patterns in statistical data, and it underlies speculative philosophical reasoning (Unsworth, Fukuda, Awh & Vogel, 2014).

Horn (1969) called fluid intelligence formless. It relies only minimally on prior learning or acculturation, whether formal or informal.

It can flow into a wide range of cognitive activities, Horn argued. Solving abstract problems and classifying novel figures both depend on it (Horn, 1968).

Fluid intelligence peaks in the late 20s. It then gradually declines (Cacioppo & Freberg, 2012).

The decline may reflect deteriorating neurological function, reduced mental use in later life, or local atrophy in the right cerebellum (Cavanaugh & Blanchard-Fields, 2006).

Recent research challenges these assumptions: some parts of fluid intelligence may not peak until age 40.

What Is Crystallized Intelligence?

Crystallized Intelligence is the ability to use skills and knowledge built up through prior learning (Horn, 1969). It means recalling information and skills already stored.

Examples include recalling historical events and dates, remembering geographical locations, building vocabulary, and reciting poetry (Horn, 1968).

Crystallized Intelligence results from accumulated knowledge, including knowledge of how to reason, language skills and an understanding of technology. This type of intelligence is linked to eduction, experience and cultural background and is measured by tests of general information.

Riding a bike or reading a book both depend on it.

Horn (1969) explained that Crystallized Intelligence is a “precipitate out of experience” which stems from a prior application of fluid intelligence.

Language-mechanics tasks such as vocabulary building rely on Crystallized Intelligence. So does general information recall.

Crystallized intelligence rises gradually across most of the lifespan and shows no comparable decline in old age (Horn & Cattell, 1967). A large modern study confirms this. It tracked vocabulary scores into people’s seventies and eighties and found them still higher than in younger adults (Salthouse, 2019).

Despite the observance of this general trend, the age at which Crystallized Intelligence reaches its peak is yet to be ascertained (Desjardins, Warnke & Jonas, 2012).

Measurements of Crystallized Intelligence

The C-Test 

The C-Test is a text completion test initially proposed as a foreign language proficiency test that provides an integrative measure of crystallized intelligence (Baghaei & Tabatabaee-Yazdi, 2015).

The underlying construct of the C-Test corresponds to the abilities undergirding the language component of crystallized intelligence.

However, research implies that the careful selection of texts from relevant domains of knowledge can enable the C-Test to measure the factual knowledge component of crystallized intelligence as well.

The Wechsler Adult Intelligence Scale (WAIS)

The revised form of the Wechsler Adult Intelligence Scale, which has been used since 1981, comprises five performance and six verbal subtests (Kaufman & Lichtenberger 2006).

These verbal tests include comprehension, information, digit span, vocabulary, similarities, and arithmetic (Wechsler Adult Intelligence Scale-Revised). Most of these verbal tests are widely construed as capable of measuring crystallized intelligence.

How the Intelligence Types Work Together

Fluid and Crystallized Intelligence are distinct, but most real tasks call on both together.

Taking a math exam is one example. A student relies on fluid intelligence to work out a strategy for answering questions in the time allowed. They then draw on crystallized intelligence to recall the mathematical concepts needed to answer them.

An entrepreneur shows the same pattern. She might use fluid intelligence to spot a new gap in the market. Filling it takes crystallized, past-built knowledge instead.

Crystallized Intelligence is not simply fluid intelligence that has ‘set’ over time (Cherry, 2018). It is produced when fluid intelligence is invested in learning new information.

In other words, working through a problem with fluid intelligence transfers information into long-term memory. That stored information becomes part of crystallized intelligence.

Can Fluid Intelligence Be Improved?

Crystallized intelligence is known to improve over time and remain stable with age, so education and experience are generally thought to increase it (Cavanaugh & Blanchard-Fields, 2006). Fluid intelligence has been a harder story.

Until recently, it was widely held that fluid intelligence is fixed, determined largely by genetics. Some research now challenges that view.

Aim: Jaeggi, Buschkuehl, Jonides, and Perrig (2008) tested whether fluid intelligence could be improved through working-memory training.

Method: 70 healthy adults completed several sessions of an adaptive working-memory task (the dual n-back task). Their gains were compared with a control group on tests of fluid reasoning.

Results: Adults who trained showed measurable gains on fluid-reasoning tests. The more training they completed, the larger the gain.

Conclusion: Fluid intelligence, once assumed fixed, can shift with targeted practice.

A later, more rigorous synthesis tempered this optimism. Au, Sheehan, Tsai, Duncan, Buschkuehl, and Jaeggi (2015) pooled the accumulated n-back training studies in a meta-analysis.

They found only a small-to-moderate overall effect on fluid intelligence (Hedges’ g ≈ 0.24), well below Jaeggi’s original result. Working-memory training can produce a real, if modest, transfer to fluid reasoning, not the dramatic gains the first study suggested.

The key themes are challenging oneself by learning new skills, problem-solving, working memory training, and exposure to intense cognitive tasks systematically and with increasing difficulty. This seems to drive neural changes that facilitate enhanced fluid reasoning abilities.

1. Physical exercise – Aerobic exercise like running or swimming can help boost fluid intelligence by promoting brain plasticity and growth.
2. Learning new skills – Actively challenging oneself by learning new complex skills and solving mentally demanding problems can improve fluid intelligence over time. The activities should be progressively more difficult.
3. Working memory training – Targeted training on working memory tasks, like memory games or mental math sequences, can help improve problem-solving capacities related to fluid intelligence.
4. Action video game training – Training with fast-paced action video games that require quick reactions and on-the-fly decision-making provided some cognitive stimulation that improved fluid intelligence-related performance.
5. Brain stimulation – Non-invasive brain stimulation techniques like tDCS and TMS showed some preliminary evidence for temporarily and modestly improving reasoning skills tied to fluid intelligence, but more research is still needed.
6. Getting good sleep – Getting enough good quality sleep supports cognitive function and consolidation of memories important for neural changes underlying learning, which relates to fluid intelligence capacity.

Critical Evaluation

The fluid/crystallized split holds up well under decades of scrutiny, but it also has real limits worth knowing.

Strong Evidence for the Age Split

Horn and Cattell’s (1967) original finding has been replicated more than once. That is rare in this field. Schaie and Hertzog (1983) retested several age cohorts over fourteen years and found the same pattern: fluid scores fell while crystallized scores held steady.

This cohort-sequential design fixes a real weakness in the original study.

A single snapshot can confuse true ageing with the fact that older and younger cohorts had different educations and life experiences to begin with. Retesting the same cohorts over time rules that out.

Few ideas in intelligence research have this much independent support. It explains why an older adult can do worse on a new reasoning puzzle yet still out-perform a younger adult on vocabulary. A single g score cannot represent that pattern at all.

Contemporary Research

Aim: Salthouse (2019) asked whether the classic fluid/crystallized age trajectories hold up once cross-sectional, longitudinal, and hybrid research designs are directly compared.

Method: Over 5,000 adults were tested once, and almost 1,600 were retested across three occasions spanning nearly six years. Vocabulary served as the crystallized measure.

Results: Fluid measures such as processing speed and reasoning declined across every method used. This held regardless of design. Vocabulary kept rising into people’s sixties and stayed higher in older adults than in younger ones.

Conclusion: More than fifty years on, the dissociation still holds. Longitudinal studies can understate fluid decline, though, unless practice effects are controlled for (Salthouse, 2019).

A separate meta-analysis asked whether these broad scores actually help predict real outcomes.

Zaboski, Kranzler, and Gage (2018) pooled studies linking CHC broad abilities to school achievement.

The results were clear. General intelligence explained far more variance than gf or gc alone, and only crystallized ability showed a reliable link, mainly for reading rather than maths.

Limited Diagnostic Value

Fluid and crystallized scores still correlate with each other and with general intelligence. The split does not escape general intelligence entirely. It only refines it.

Reporting fluid and crystallized scores separately may add less to a diagnosis than test manuals suggest. That is because g alone predicts achievement better than either broad score on its own (Zaboski, Kranzler, & Gage, 2018).

Nor is the fluid/crystallized carve-up the only option. When several rival models were fitted to one large battery of ability tests, a structure built around test content, not the fluid/crystallized distinction, fit best (Johnson & Bouchard, 2005).

This points to a deeper reification concern.

gf and gc are convenient summaries of test correlations, not directly observed brain structures. Reasonable modelling choices differ. They can carve up the same data in more than one defensible way.

Cultural Bias in Testing

Cattell’s own ‘culture-free’ test was later renamed ‘culture-fair.’ That is itself an admission: no test built in one culture can be fully independent of it (Cattell, 1940). Testing familiarity still varies by culture.

Helms (1992) went further.

She argued that mainstream ability testing rarely checks whether a test measures the same thing across cultural and racial groups in the first place. That gap matters most for crystallized measures, which are built directly from culturally specific vocabulary and knowledge.

This is not just a technical concern. Culture-fair, fluid-reasoning tests are often used in cross-national hiring or forensic assessment. They are chosen for exactly this reason.

Comparing crystallized scores across groups is different, though. Groups with different educational access or first languages remain especially prone to unfair disadvantage rather than a genuine ability difference.

References

Au, J., Sheehan, E., Tsai, N., Duncan, G. J., Buschkuehl, M., & Jaeggi, S. M. (2015). Improving fluid intelligence with training on working memory: A meta-analysis. Psychonomic Bulletin & Review, 22 (2), 366–377.

Baghaei, Purya & Tabatabaee-Yazdi, Mona. (2015). The C-Test: An Integrative Measure of Crystallized Intelligence. Journal of Intelligence, 3 (2), 46-58.

Cacioppo, J. T., & Freberg, L. (2012). Discovering psychology: The science of mind. Cengage learning.

Cattell, R. B. (1963). Theory of fluid and crystallized intelligence: A critical experiment. Journal of Educational Psychology, 54 (1), 1–22.

Cattell, R. B. (1971). Abilities: Their structure, growth, and action. New York: Houghton Mifflin.

Cattell, Raymond B. (1987). Intelligence: Its Structure, Growth, and Action. Elsevier Science Publishers.

Cavanaugh, J. C.; Blanchard-Fields, F (2006). Adult development and aging (5th ed.). Belmont, CA: Wadsworth Publishing/Thomson Learning.

Desjardins, R., & Warnke, A.J. (2012). Ageing and Skills (PDF). OECD Education Working Papers.

Ferrer, E., O”Hare, E. D., & Bunge, S. A. (2009). Fluid reasoning and the developing brain. Frontiers in neuroscience, 3 (1), 46–51.

Flanagan, D. P., & Kaufman, A. S. (2004). Essentials of WISC-IV assessment. Hoboken, NJ: John Wiley.

Geary, D. C. (2005). The origin of mind: Evolution of brain, cognition, and general intelligence. Washington, DC: American Psychological Association

Helms, J. E. (1992). Why is there no study of cultural equivalence in standardized cognitive ability testing? American Psychologist, 47 (9), 1083–1101.

Horn, J. L. (1968). Organization of abilities and the development of intelligence. Psychological Review, 75 (3), 242-259.

Horn, J. L. (1969). Intelligence: Why it grows. Why it declines. Trans-action, 4, 23-31.

Horn, J. L., & Cattell, R. B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologica, 26, 107–129.

Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of Sciences, 105 (19), 6829-6833.

Johnson, W., & Bouchard, T. J., Jr. (2005). The structure of human intelligence: It is verbal, perceptual, and image rotation (VPR), not fluid and crystallized. Intelligence, 33 (4), 393–416.

Kaplan, J. T., Gimbel, S. I., & Harris, S. (2016). Neural correlates of maintaining one’s political beliefs in the face of counterevidence. Scientific reports, 6, 39589.

Kaufman, Alan S.; Lichtenberger, Elizabeth (2006). Assessing Adolescent and Adult Intelligence (3rd ed.). Hoboken (NJ): Wiley.

Martin, JH (2003). Lymbic system and cerebral circuits for emotions, learning, and memory. Neuroanatomy: text and atlas (third ed.) . McGraw-Hill Companies.

Pardo, J. V., Pardo, P. J., Janer, K. W., & Raichle, M. E. (1990). The anterior cingulate cortex mediates processing selection in the Stroop attentional conflict paradigm. Proceedings of the National Academy of Sciences, 87 (1), 256-259.

Raven, J. C. (1983). Manual for Raven’s progressive matrices and vocabulary scales. Standard Progressive Matrices.

Salthouse, T. A. (2019). Trajectories of normal cognitive aging. Psychology and Aging, 34 (1), 17–24.

Schaie, K. W., & Hertzog, C. (1983). Fourteen-year cohort-sequential analyses of adult intellectual development. Developmental Psychology, 19 (4), 531–543.

Schrank, F. A.; Flanagan, D. P. (2003). WJ III Clinical use and interpretation. Scientist-practitioner perspectives. San Diego, CA: Academic Press.

Unsworth, Nash; Fukuda, Keisuke; Awh, Edward; Vogel, Edward K. (2014). Working memory and fluid intelligence: Capacity, attention control, and secondary memory retrieval. Cognitive Psychology, 71, 1–26.

Wechsler Adult Intelligence Scale–Revised. LIST OF TESTS Available from the CPS Testing Library. Center for Psychological Studies at Nova Southeastern University.

Wechsler, D. (2003). WISC-IV technical and interpretive manual. San Antonio, TX: Psychological Corporation.

Woodcock, R. W.; McGrew, K. S.; Mather, N (2001). Woodcock Johnson III. Itasca, IL: Riverside.

Zaboski, B. A., Kranzler, J. H., & Gage, N. A. (2018). Meta-analysis of the relationship between academic achievement and broad abilities of the Cattell-Horn-Carroll theory. Journal of School Psychology, 71, 42–56.

Further Reading

• Kievit, R. A., Davis, S. W., Griffiths, J., Correia, M. M., & Henson, R. N. (2016). A watershed model of individual differences in fluid intelligence. Neuropsychologia, 91, 186-198.
• General Intelligence
• Multiple Intelligence