Average IQ by Age Across the Lifespan

One of the most common questions in intelligence research is whether IQ changes with age. The answer is genuinely nuanced: some cognitive abilities peak in adolescence, others not until middle age, and a few continue improving into the seventh decade of life. Understanding these trajectories — and the difference between population-level patterns and individual variation — is essential for interpreting test scores accurately and forming realistic expectations about cognitive aging.

How IQ Scores Are Normed by Age

IQ tests like the WAIS-V and WISC-V are normed against age-specific reference groups. A score of 100 always means average for your age group. A 70-year-old scoring 100 performs as well as the typical 70-year-old — not the typical 25-year-old. This age-norming is essential for clinical interpretation (you don’t want to label a healthy octogenarian as impaired), but it also conceals the raw cognitive changes that occur across the lifespan. To see the underlying trajectory, researchers look at raw scores — the actual number of items correct — before age-norming is applied.

When researchers examine raw scores, a clear developmental arc emerges. Cognitive abilities increase rapidly through childhood, reach near-adult levels by ages 16–18, and continue improving slightly through the early-to-mid twenties before plateauing and eventually declining at very different rates depending on the specific ability. The shape of that arc, and the way it differs across cognitive domains, is one of the most robust findings in the psychology of aging.

When Do Different Cognitive Abilities Peak?

The single most influential modern study on this question is Hartshorne and Germine’s (2015) analysis of nearly 50,000 online participants combined with archival Wechsler datasets. Their conclusion is captured by the article’s subtitle: “the asynchronous rise and fall of different cognitive abilities across the life span.” Specifically:

• Processing speed peaks earliest, around ages 18–19, then declines steadily. This is why reaction-time and digit-symbol tests show the most pronounced age effects.
• Short-term and working memory peak in the mid-twenties — around 25 in the Hartshorne & Germine data — and start a slow decline in the mid-thirties.
• Fluid reasoning — the ability to solve novel problems, as measured by Raven’s-style matrices and similar tests — peaks in the mid-twenties and begins gradually declining by the early thirties.
• Emotional perception (judging others’ emotions from facial cues) peaks in the forties and fifties, an age range when most people are deepest in family and managerial life.
• Crystallized intelligence — vocabulary, general knowledge, accumulated expertise — continues climbing through the forties, fifties, and even sixties. Vocabulary specifically peaks around ages 65–70 in the Hartshorne data and many earlier studies.

This pattern fits the long-standing Cattell–Horn distinction between fluid and crystallized intelligence (Gf/Gc): the more biologically grounded reasoning capacities decline with age, while the knowledge-based abilities benefit from continued accumulation. The practical implication is that “are you smarter at 25 or 55?” is the wrong question. You are likely faster at 25 and more knowledgeable at 55, and which matters more depends on the task.

Cross-Sectional Versus Longitudinal Data

How you measure age-related cognitive change matters enormously. Cross-sectional studies (comparing different people of different ages at one time point) tend to overstate cognitive decline because they confound true age effects with cohort effects — older generations grew up with less schooling and (because of the Flynn effect) scored lower on average even when they were young. Longitudinal studies (following the same people over time) show slower rates of decline, but they have their own biases: practice effects from repeated testing inflate scores, and selective attrition (the cognitively impaired drop out of long studies) flatters the surviving sample.

Salthouse (2010), in a widely cited review, argued that the truth lies between the two estimates, and that careful adjustment for these methodological artifacts shows real fluid-ability decline beginning in the late twenties or early thirties — earlier than the optimistic longitudinal picture suggests, but more gradual than cross-sectional comparisons imply.

The single most informative dataset is K. Warner Schaie’s Seattle Longitudinal Study, which has tracked cognitive abilities in thousands of adults since 1956 (Schaie, 2012). It finds that most adults maintain near-peak performance on most abilities into their sixties; meaningful decline on the average composite typically does not appear until the mid-seventies. Equally important, individual variation is enormous — the gap between the most- and least-preserved 75-year-olds in Schaie’s cohorts is wider than the average difference between 30- and 75-year-olds.

Stability of Rank Order Across the Lifespan

One of the most striking findings in the field comes from a Scottish natural experiment. In June 1932, virtually every 11-year-old child in Scotland (about 87,000 children) sat the same intelligence test on the same day as part of the Scottish Mental Survey. Decades later, Ian Deary and colleagues tracked down survivors and re-tested them. Deary, Whalley, Lemmon, Crawford, and Starr (2000) reported a correlation of approximately r = 0.66 between age-11 and age-77 scores — extraordinary stability across more than two-thirds of a human lifetime.

This rank-order stability has two important implications. First, it confirms that intelligence as measured by these tests is a real, durable trait — not a transient testing artifact. Second, it does not mean individuals stay at the same absolute level. Everyone slows down. But where you sit relative to your age peers tends to be remarkably similar at 11 and at 77. The follow-up Lothian Birth Cohort 1936, run from the same Scottish source, has since become one of the world’s most informative datasets on cognitive aging.

Age-Related Changes in Full Scale IQ and Index Scores

The composite Full Scale IQ (FSIQ) blends abilities that age very differently. Because FSIQ combines declining capacities (processing speed, fluid reasoning) with stable or improving ones (vocabulary, general knowledge), the net effect on raw FSIQ is relatively flat through the fifties and sixties, with gradual decline thereafter. On WAIS-IV normative data, the average raw-score decline from peak (ages 20–34) to ages 70–74 is roughly 15–20 raw-score points — a real, measurable drop that is entirely masked when results are converted to age-normed IQ.

Index scores tell a more nuanced story. The Processing Speed Index shows the steepest decline (approximately one full standard deviation from age 25 to 75), while the Verbal Comprehension Index shows the least (less than 0.3 standard deviations across the same span). For clinicians interpreting older adults’ profiles, this differential pattern is essential: an older adult with a flat profile (uniform scores across indices) often deserves more concern than one with a typical age-related dip in processing speed but preserved verbal performance.

What Affects the Rate of Cognitive Aging?

Individual variation in cognitive aging is enormous, and several factors shape the trajectory. Education is consistently associated with slower decline, likely through what researchers call cognitive reserve — the brain’s capacity to absorb age-related damage without functional symptoms, built up by decades of intellectual demand. Aerobic exercise protects against decline through cardiovascular and neurotrophic mechanisms; the effect sizes in randomized trials are modest but consistent. Social engagement and continued intellectual challenge — work, hobbies, complex relationships — also appear protective.

Conversely, cardiovascular risk factors (hypertension, diabetes, high cholesterol), chronic sleep disruption, social isolation, and untreated depression all accelerate cognitive aging. The crucial point is that all of these factors are at least partly modifiable. The trajectory is not predetermined: cognitive aging is influenced by lifestyle decisions made decades earlier and by ongoing choices in mid- and late-life.

Implications for Testing and Assessment

For clinicians, the central practical implication is that age-appropriate norms must always be used. A 75-year-old with an FSIQ of 95 is performing within the normal range for her age and is not impaired. The harder clinical question is whether she has declined from her own earlier peak. Practitioners estimate premorbid ability from education, occupation, and demographic variables, and look for divergence between current performance and that estimate. A drop relative to expected level can indicate pathological decline even when raw scores fall within “normal” range — and a stable below-average score is rarely cause for concern in a person who has always tested below average.

For individuals, the message is mostly optimistic. Cognitive decline is neither inevitable nor uniform. Many abilities are maintained well into old age, vocabulary and knowledge can keep growing for decades after physical peak, and the rate of change is meaningfully responsive to lifestyle factors. The brain remains plastic across the lifespan, and the things that build cognitive reserve — education, exercise, social connection, sustained intellectual challenge — are also things most people can choose to do more of.

Frequently Asked Questions

What is the average IQ for a 12-year-old?

By definition, the average IQ for any age group is 100, since IQ tests are normed against age-specific reference groups. A 12-year-old scoring 100 performs as well as the average 12-year-old in the normative sample.

At what age does IQ stop increasing?

Different abilities peak at different ages. Processing speed peaks around 18–19, fluid reasoning and working memory in the mid-twenties, emotional perception in the forties and fifties, and vocabulary and general knowledge as late as 65–70 (Hartshorne & Germine, 2015). There is no single age at which all cognitive abilities stop developing.

Does IQ decline after 50?

Most adults maintain their overall cognitive ability well into their sixties when measured longitudinally. Processing speed and fluid reasoning begin declining earlier, but those declines are partially offset by continued gains in crystallized intelligence. Meaningful overall decline typically does not appear until the mid-seventies in healthy individuals.

Can you raise your IQ as an adult?

You can improve specific skills with training, and educational interventions show modest broader effects. But the underlying construct (g) is highly stable in adulthood, and “IQ-boosting” claims for brain-training apps have generally failed to replicate in well-controlled trials. Maintaining cognitive function through exercise, sleep, education, and social engagement is a more realistic goal than raising baseline IQ.

Why do IQ scores stay 100 on average if abilities decline with age?

Because age-norming. IQ scores compare you to peers of the same age, so the average is always 100 by construction. The underlying raw cognitive performance does change with age — it’s just hidden by the comparison framework.

Are early-life IQ scores predictive of late-life cognition?

Yes — strikingly so. Deary and colleagues’ follow-up of the 1932 Scottish Mental Survey found a correlation of about r = 0.66 between age-11 and age-77 IQ. People generally retain their relative standing across the lifespan even as everyone’s absolute performance shifts.

The Bottom Line

“Average IQ by age” is a question with two answers depending on how you ask it. Statistically, the average is 100 at every age, because that’s how the test is built. Cognitively, abilities peak at different ages — fluid in the twenties, emotional and managerial in the forties, vocabulary in the sixties — and decline at different rates. The story is one of asynchronous rise and fall rather than a single peak followed by a single descent. Most adults preserve far more cognitive function into old age than popular stereotypes suggest, and the trajectory remains responsive to choices made throughout life.