Angelman Syndrome and IQ

Posted on October 27, 2026 by Mahesh

Understanding IQ

Angelman Syndrome and IQ: Developmental Scores, Subtypes and Strengths

Angelman syndrome and IQ: severe intellectual disability that standard IQ tests rarely measure, so researchers use developmental scales. Bayley-III scores average a developmental quotient near 40. See how genetic subtype, seizures and the Prader-Willi link shape development.

By Alden Cross, Intelligence Research WriterPublished October 27, 2026 · 8 min readReviewed by Aarav Vardhan, Editor-in-Chief

Diagram of chromosome 15 region q11-q13 in three rows: typical, Angelman syndrome with the maternal copy missing and Prader-Willi syndrome with the paternal copy missing, beside a dot chart showing 106 of 119 children with Angelman syndrome had a mental age under 24 months.

Angelman syndrome and IQ are hard to put together, because the syndrome causes severe intellectual disability and standard IQ tests usually cannot be used, so researchers describe cognition with developmental scales instead. On the Bayley-III, children aged 5 months to 5 years averaged a cognitive developmental quotient of about 40, with a range of roughly 25 to 56 for the middle two-thirds. In a Chinese cohort of 119 children, 89% had mental ages below 24 months. Skills keep growing, slowly, into the teenage years. This guide explains what those numbers mean, how Angelman syndrome compares with its sister disorder Prader-Willi syndrome and IQ, and why genetic subtype and seizures matter more than any single score.

What is Angelman syndrome?

Angelman syndrome is a rare neurogenetic disorder caused by the loss of function of the UBE3A gene on the long arm of chromosome 15 (15q11-q13). In neurons only the copy inherited from the mother is active; the father’s copy is silenced by genomic imprinting. So anything that removes or disables the maternal copy produces the syndrome: a deletion of the region on the maternal chromosome (the most common cause), two copies of chromosome 15 from the father (uniparental disomy), a fault in the imprinting control centre, or a variant within UBE3A itself. A recent consensus statement puts the frequency at about 1 in 12,000 and describes the core picture as developmental delay, cognitive impairment, movement problems, seizures and an abnormal EEG background. A British paediatrician, Harry Angelman, first described children with this pattern in 1965.

Diagnosis is genetic and clinical. A consensus update of the diagnostic criteria lists absent or minimal speech, a movement or balance disorder, a happy and easily excited demeanour and developmental delay as consistent features; an IQ score is not part of it. Anyone with concerns about a child should speak to a paediatrician or clinical geneticist; this page is general information, not medical advice.

How are Angelman syndrome and Prader-Willi syndrome related?

They are the textbook pair of imprinting disorders. In 1989 Knoll and colleagues showed that people with Angelman syndrome and people with Prader-Willi syndrome often carry the same deletion of 15q11-q13, yet look completely different, and that the difference lies in which parent the deleted chromosome came from: maternal in Angelman syndrome, paternal in Prader-Willi. It was an early demonstration that a gene’s effect can depend on its parent of origin.

Angelman syndrome and Prader-Willi syndrome side by side
Feature	Angelman syndrome	Prader-Willi syndrome
Chromosome region	15q11-q13	15q11-q13
Which parent’s copy is lost or silent	Maternal copy	Paternal copy
Main gene involved	UBE3A (active only from the maternal copy in neurons)	Several paternally expressed genes in the region
Typical cognitive level	Severe; developmental scales, not IQ tests	Mild to moderate; IQ near 60 on average
Speech	Minimal or absent	Delayed, but usually present
Hallmark features	Seizures, ataxia, a happy demeanour, poor sleep	Hunger that cannot be satisfied, low muscle tone, short stature
Estimated frequency	About 1 in 12,000	About 1 in 15,000 to 1 in 27,000
Our IQ guide	This page	Prader-Willi syndrome and IQ

The IQ contrast is stark. Prader-Willi syndrome is usually in the mild to moderate range of intellectual disability, with average scores near 60 on Wechsler-type tests, as set out in our Prader-Willi syndrome and IQ guide. Angelman syndrome sits well below, to the point where a Wechsler IQ is usually not obtainable.

Why are IQ tests rarely used for Angelman syndrome?

An IQ test needs a child to follow spoken instructions, answer or point on request, sit through a long session and attempt tasks built for typical development. The Wechsler scales print full-scale IQs from 40 upward, while the ICD-10 places severe intellectual disability at 20 to 34 and profound below 20 (our IQ of 65 guide sets out the bands). Children with Angelman syndrome typically have minimal or no speech, limited attention for formal tasks, and motor difficulties that make pointing and manipulating blocks hard, so standard IQ results are either impossible or floor-bound.

Researchers therefore use developmental tests designed for young children, mainly the Bayley Scales of Infant and Toddler Development and the Griffiths Mental Development Scales, together with caregiver-rated adaptive scales such as the Vineland. They report age equivalents (the age at which a typical child passes the same items) and a developmental quotient, roughly a child’s developmental age as a share of real age. These tools have limits of their own. In an international study of 142 children, expressive language scores on the Bayley-III were invalidated by the speech difficulties of Angelman syndrome, and scores were unstable between visits: more than half of the 52 children tested more than once got items wrong that they had passed before, which the authors link to motivation, attention and anxiety as well as true loss of skills.

What do the Bayley and Griffiths scores show?

The best-known profile is the 2010 natural history study by Gentile and colleagues, which assessed 92 children aged 5 months to 5 years. The mean Bayley-III cognitive developmental quotient was 40.5 with a standard deviation of 15.5. Cognitive scores were higher than motor and language scores, and receptive language (understanding) was higher than expressive language (speech). A 2022 study of 119 Chinese children (mean age 42 months) used the Griffiths scales and found a median general quotient of 29.6, with 106 of 119 (89%) functioning below the 24-month level in all five areas. The children had relatively good locomotor and personal-social skills but poor language.

To make a quotient concrete: a developmental quotient of 40 means a child is functioning at about 40% of their real age. The table below applies the 40.5 average, and the range for one standard deviation either side (25 to 56). It is illustrative arithmetic, not a claim about any individual child.

What a developmental quotient of about 40 implies in months
Real age	Developmental age at DQ 40.5	Range at one SD (DQ 25 to 56)
2 years	10 months	6 to 13 months
3 years	15 months	9 to 20 months
4 years	19 months	12 to 27 months
6 years	29 months	18 to 40 months
8 years	39 months	24 to 54 months

A quotient is not an IQ, and the two should not be mixed: a DQ is built from age equivalents and can fall as a child ages even when skills are still being learned, because the real-age denominator can grow faster than the developmental age. A longitudinal analysis of 236 children found that people with Angelman syndrome keep making slow gains at about 1 to 2 months of age equivalent per year through at least age 12. In other words, development does not stop; it is slow.

Does the genetic subtype change the picture?

Yes, more than any other single factor. Children with a deletion are more delayed than those with the other mechanisms in every Bayley domain except expressive language. A 2021 analysis of 250 individuals and 848 assessments ranked the groups: those with a deletion are most affected, then paternal uniparental disomy, then imprinting defects and UBE3A variants, and among the variants, truncating changes are more severe than missense ones. Interestingly, the researchers found little evidence that the size of the deletion (class I or class II) changes severity much, which points to genes other than UBE3A in the region contributing.

Genetic subtypes of Angelman syndrome and severity
Cause	Share of children	Severity relative to the rest
Deletion of 15q11-q13 (UBE3A and neighbouring genes)	74% of children in the Gentile cohort	Most affected
Paternal uniparental disomy (UPD)	Part of the other 26%	Less affected than deletion, more than the next two
Imprinting defect	Part of the other 26%	Less affected than UPD
UBE3A variant, truncating	Part of the other 26%	Less affected than UPD; more than missense
UBE3A variant, missense	Part of the other 26%	Milder than truncating variants

Your own number

Where would your own score land?

Take the IIF-certified assessment and get your score with the scale it was measured on, the percentile it corresponds to and the confidence range around it — the three figures most online tests leave out.

Find your IQ score now! →

Secure & encryptedInstant results10–20 minutes

What about seizures?

Epilepsy is a core feature: up to 90% of people with Angelman syndrome have seizures, usually in the early years. In the Chinese Griffiths study the group with epilepsy had lower quotients than the group without, on the locomotor, personal-social, speech, language and eye-hand scales. Whether seizures lower development, or whether the severe neurological changes cause both, cannot be decided from these studies; the practical point is that seizure control is a central part of care, and that scores differ by seizure history as well as by genetics.

What are the strengths?

The Gentile study found cognition to be the strongest developmental domain and understanding stronger than speech, which fits the clinical picture of children who take in far more than they can say. Communication through signs, pictures and devices therefore matters, and the developmental scales may understate what a child understands because they rely on a child showing it. The happy, sociable demeanour is a real feature of the syndrome, but an older nickname built on it is now widely regarded as outdated and is not used here.

Why does this matter for treatment trials?

Clinical trials in Angelman syndrome need ways to measure change. A 2025 study of clinicians who give them describes the Bayley-4 and the Vineland-3 as outcome measures often considered as primary endpoints, with barriers such as a child’s lack of interest in the tasks; a 2026 panel of 19 caregivers set out what size of change on those scales is meaningful in daily life. None of this involves an IQ.

How does this compare with other genetic conditions?

Angelman syndrome is at the severe end of the genetic syndromes we cover. For comparison, our guides to Down syndrome, Williams syndrome, fragile X syndrome, Turner syndrome and Klinefelter syndrome show mild to moderate effects on IQ, which can be measured with Wechsler tests, and at the opposite end of the scale see what an IQ of 155 means. If you want to see how an ordinary IQ result is built, take the IQ Metrics IQ test or browse the IQ tests hub, and read the IQ classification bands.

Sources

Gentile, J.K. et al. (2010): A neurodevelopmental survey of Angelman syndrome with genotype-phenotype correlations, Journal of Developmental & Behavioral Pediatrics.
Li, S. et al. (2022): Epilepsy and molecular phenotype affect the neurodevelopment of pediatric Angelman syndrome patients in China, Frontiers in Psychiatry.
Sadhwani, A. et al.: Developmental skills of individuals with Angelman syndrome assessed using the Bayley-III, Journal of Autism and Developmental Disorders.
Keute, M. et al. (2021): Angelman syndrome genotypes manifest varying degrees of clinical severity and developmental impairment, Molecular Psychiatry.
Ten Hooven-Radstaake, M. et al. (2025): Criterion validity, scalability and stability of scoring on the Bayley-III in children with Angelman syndrome, Journal of Intellectual Disability Research.
Duis, J. et al. (2022): A multidisciplinary approach and consensus statement to establish standards of care for Angelman syndrome, Molecular Genetics & Genomic Medicine; Williams, C.A. et al. (2006): Angelman syndrome 2005: updated consensus for diagnostic criteria, American Journal of Medical Genetics; Buiting, K., Williams, C. and Horsthemke, B. (2016): Angelman syndrome: insights into a rare neurogenetic disorder, Nature Reviews Neurology.
Knoll, J.H. et al. (1989): Angelman and Prader-Willi syndromes share a common chromosome 15 deletion but differ in parental origin of the deletion, American Journal of Medical Genetics.
Halpin, S.N. et al. (2025): Using the Bayley-4 and Vineland-3 in Angelman syndrome, Orphanet Journal of Rare Diseases; Powers, S. et al. (2026): Developing meaningful score differences for the Bayley-4 and Vineland-3 in Angelman syndrome using a Delphi panel, Value in Health; Ebrat-Mancilla, E. et al. (2024): Epilepsy in Angelman syndrome and the most common electroencephalographic findings, Revista de Neurologia.

Share this article

Know someone who keeps seeing these numbers quoted without the scale they were measured on? Send it to them.

Keep reading

All articles →

Number line of IQ scores from 35 to 100 with the moderate (35 to 49) and mild (50 to 69) intellectual disability bands shaded and four Prader-Willi syndrome studies marked: a mean of 62.3 and range 39 to 96 in 26 youths, about 60 in a UK population sample, a median of 52 in 85 adults, and verbal IQ of 60.8 with a deletion against 69.9 with maternal UPD.

Understanding IQ

Prader-Willi Syndrome and IQ

Prader-Willi syndrome and IQ: average scores sit near 60, about 40 points below the general population, with a range from the 40s to the 90s. See how deletion and maternal UPD differ, the jigsaw-puzzle strength what a growth hormone trial found and how IQ fits into diagnosis.

October 26, 202610 min read

Understanding IQ

Down Syndrome and IQ

Most people with Down syndrome score in the mild to moderate range of intellectual disability, with an average IQ of around 50 and a typical range of about 30 to 70. The number hides a wide spread, a consistent profile of strengths, and test limits that matter more here than almost anywhere else.

September 16, 202610 min read

Understanding IQ

Williams Syndrome and IQ

Williams syndrome and IQ: mean full-scale IQ was 61 in 62 adults, and 292 children averaged in the borderline range on the KBIT-2, stable from age 4 to 17. Language is a relative strength and visuospatial construction an extreme weakness. Here is what the studies show.

October 10, 20269 min read

What Does an IQ of 155 Mean?

Posted on October 27, 2026 by Mahesh

Scores & Scales

What Does an IQ of 155 Mean? Rarity, Percentile and Reliability

An IQ of 155 means a score 3.7 standard deviations above average: the 99.988th percentile, about 1 in 8,140 people or a million worldwide. See how it differs by scale, the odds a retest repeats it, what it means for a child as a mental age, and which historical figures Cox rated at 155.

By Leon Falkner, Data & Research AnalystPublished October 27, 2026 · 8 min readReviewed by Aarav Vardhan, Editor-in-Chief

Large 155 beside a vertical IQ ladder from 131 to 171 on the standard deviation 15 scale: Mensa line 131, Triple Nine line 146, 150, 155 highlighted at 1 in 8,140, Prometheus line 160 and Mega Society line 171.

An IQ of 155 is about 3.7 standard deviations above the mean, the 99.988th percentile on the standard scale, roughly 1 in 8,140 people. That puts it between the Triple Nine Society line (146) and the Prometheus Society line (160), and means about a million people alive today would score this high on a bell curve. This guide gives the exact rarity on each scale, how many people that is in your country, why a reported 155 usually comes down on a retest, what it means for a child, and who history’s ratings put near it.

How rare is an IQ of 155?

On the scale used by the Wechsler tests and the current Stanford-Binet, which fix the mean at 100 and the standard deviation at 15, a score of 155 is 55 points above average, or 3.67 standard deviations. The bell curve leaves about 1 in 8,140 of the population above that point. The old Stanford-Binet Form L-M and Cattell’s scales used a wider spread, so the same number meant something very different on them.

An IQ of 155 on three scales
Scale	Standard deviation	Standard deviations above mean	Percentile	Rarity
Wechsler tests, Stanford-Binet 5	15	3.67	99.988%	1 in 8,140
Stanford-Binet Form L-M (older edition)	16	3.44	99.971%	1 in 3,410
Cattell scales	24	2.29	98.9%	1 in 91

Always ask which scale a quoted 155 belongs to. A 155 on a standard-deviation-24 scale is only about the top 1%; on the standard scale it is 1 in 8,140. If you want to convert between them, the IQ score converter does it, and the IQ percentile calculator gives the percentile for any score.

How many people would score 155 or higher (normal-curve arithmetic)
Population	People at 155 or above
The world (8.2 billion)	1,000,000
United States (336.8 million)	41,000
United Kingdom (68.5 million)	8,400
Canada (40.0 million)	4,900
Australia (26.7 million)	3,300
New Zealand (5.2 million)	640

Set beside its neighbours, 155 is a long way from either. The last column shows what a single sitting on a very reliable test (reliability .98) would be expected to give on a second try, which previews a point we return to below.

Neighbouring scores compared
IQ	Rarity	People in the world	Expected retest, reliability .98
150	1 in 2,330	3,500,000	149
155	1 in 8,140	1,000,000	154
160	1 in 31,600	260,000	159

These are the numbers a perfect bell curve predicts, not a count of people who have taken a test. Real score distributions have thicker tails at the very top and bottom than the textbook curve, and few people ever sit a test that can reliably report a score this high, so the real number of documented 155s is a small fraction of these counts. Our guide to how rare a high IQ score is compares more scores.

Where does 155 sit between Mensa and the Mega Society?

Each high-IQ society draws a line at a percentile, and each step up is roughly a tenfold jump in rarity. A 155 clears the first three lines easily and stops short of the fourth. The scores below are for the standard-deviation-15 scale.

The high-IQ ladder
IQ (SD 15)	Percentile	Rarity	Reference point
131	98th	1 in 50	Mensa line (top 2%, usually quoted as 130)
135	99th	1 in 100	One-in-a-hundred societies such as Intertel
146	99.9th	1 in 1,000	Triple Nine Society line
150	99.96th	1 in 2,330	Our IQ of 150 guide
155	99.988th	1 in 8,140	This page
160	99.997th	1 in 30,000	Prometheus Society line; our IQ of 160 guide
171	99.9999th	1 in 1,000,000	Mega Society line; see IQ of 170

So 155 is roughly 3.5 times rarer than 150 and 3.9 times more common than 160. For the societies themselves, and which tests they accept, see high-IQ societies and Mensa score requirements.

Can a test even report an IQ of 155?

Fewer than you might think. The Wechsler Adult Intelligence Scale, fourth edition, reports a full-scale IQ from 40 to 160, so 155 sits only five points from the printed maximum and needs near-top scaled scores on almost every subtest. The WISC-V has extended norms for children that continue past 160, but ordinary scoring stops at the same ceiling. Short online quizzes and puzzle tests often cannot separate a 140 from a 155 at all, because they contain too few hard items; our guide to the IQ ceiling effect explains why, and the highest IQ score a test can report lists the cut-offs by test.

Older ratio-IQ tests had a ceiling of a different kind. On the Stanford-Binet Form L-M the highest mental age was about 22 years 10 months, so a ratio IQ of 155 was out of reach for anyone older than 14 years 9 months (our arithmetic: 274 months divided by 1.55). For the full story of ratio versus deviation IQ, see deviation IQ vs ratio IQ.

If you score 155, how likely is it to be real?

A single high score is less certain than it looks. Every test has measurement error, and an extreme score tends to be partly luck: on a retest the expected result moves back toward 100. The standard correction (Kelley’s formula) gives the expected true score as 100 plus the test’s reliability times 55. The table applies it, and also shows the chance that a second sitting lands at 155 or higher.

What an observed 155 implies, by test reliability
Test reliability	Expected retest score	95% range for the true score	Chance a retest is 155+
0.80 (A short unsupervised online test)	144	132 to 156	12%
0.90 (A good group-administered test)	150	141 to 158	22%
0.95 (A strong individual test)	152	146 to 159	32%
0.98 (A full individually administered battery)	154	150 to 158	42%

Even on the most reliable battery, only about four in ten people who score 155 will score 155 or higher again, and on a short online test it is about one in eight. That is not a flaw to fix; it is what a score near the extreme of any measured trait does. It is the reason serious assessments report a confidence band rather than a single number, and the reason a result like this should be checked on a second, independent test. To see how your own result is built, take the IQ Metrics IQ test.

Your own number

Where would your own score land?

Find your IQ score now! →

Secure & encryptedInstant results10–20 minutes

What does an IQ of 155 mean for a child?

For a child the number was once defined as mental age divided by real age, times 100. That means a ratio IQ of 155 is a child functioning like someone 55% older. Modern tests no longer divide this way, but the arithmetic still shows how far ahead a 155 implies:

A ratio IQ of 155 as a mental age
Real age	Mental age	Reachable on Stanford-Binet L-M?
5 years	7 years 9 months	Yes
6 years	9 years 4 months	Yes
8 years	12 years 5 months	Yes
10 years	15 years 6 months	Yes
12 years	18 years 7 months	Yes
14 years	21 years 8 months	Yes

A child with this profile is usually described as highly gifted in school terms, a common label for 145 to 159, with 130 to 144 “moderately gifted” and 160 or more “exceptionally gifted”. These labels are conventions, not diagnoses, and a child’s profile is usually uneven: a high overall score can hide large gaps between verbal, visual and working-memory skills. Mental-age conversions are explained further in IQ and mental age.

Is an IQ of 155 genius level?

There is no official genius threshold, and 140 is the most common informal line. A 155 is well past it by any usage, but “genius” in the sense of an original contribution to a field is not something a test score shows; see what genius IQ level means and whether 130 is genius. The people who produce the work history remembers have generally had high scores, but the score is a floor for opportunity and effort to build on, not a prediction of eminence.

Who has been estimated at an IQ of 155?

No historical figure sat an IQ test, so every figure below is an estimate. In 1926 Catharine Cox rated 301 eminent people from biographies, and she concluded that the group’s true average was not lower than 155 to 165. The table lists some of those whose corrected rating was exactly 155. Her raters gave lower figures; the correction pushed each one up.

People whose corrected rating in Cox (1926) was 155
Person	Field	Raters’ young-adult estimate	Corrected
John Adams (1735-1826)	Statesman	145	155
Alexander Hamilton (1757-1804)	Statesman	140	155
Ralph Waldo Emerson (1803-1882)	Essayist	145	155
Thomas More (1478-1535)	Statesman, writer	135	155
Rembrandt van Rijn (1606-1669)	Artist	135	155
Albrecht Dürer (1471-1528)	Artist	135	155

The key point is how common 155 is on Cox’s own scale: 253 of the 301 people carry a corrected rating of 155 or more, 36 of them exactly 155. Her scale was stretched and cannot be compared with a modern test, which is why quoting a historical genius as “IQ 155” tells you little. The comparison with a real test sample is Terman’s: the 1,070 children in his study averaged 151.0 on the Stanford-Binet of the time, with scores from 135 to 200, so a 155 would be a typical member of that group, on a ratio scale whose top end is stretched. Our pages on the IQs of 301 historical geniuses and the Terman study of the gifted give the full tables.

What an IQ of 155 does not tell you

Your future. The correlation of IQ with later outcomes is moderate at best, and at the extreme top other traits matter as much.
Your strengths. A composite hides the shape of a profile; two people with 155 can be very different.
That the score is exact. Add or subtract several points for measurement error, as in the table above.
That it will stay put. Norms age; see why IQ norms expire.

For neighbouring scores see an IQ of 145, 150, 160 and 170. To see how rare your country’s highest scorers are, see average IQ in Canada and average IQ in Australia.

Sources

Cox, C.M. (1926): The Early Mental Traits of Three Hundred Geniuses (Genetic Studies of Genius, Vol. II), Stanford University Press, Tables 12A and 12B, as set out in our study of 301 historical geniuses.
Terman, L.M. et al. (1925): Mental and Physical Traits of a Thousand Gifted Children, as summarised in our Terman study page.
Kelley, T.L. (1923): Statistical Method, Macmillan (the regression-to-the-true-score formula). Normal-curve arithmetic is ours; the high-IQ society percentiles are those published by the societies, summarised in our guide.
Pearson: WAIS-IV score ranges (full-scale IQ 40 to 160) and WISC-V extended norms technical report.

Share this article

Know someone who keeps seeing these numbers quoted without the scale they were measured on? Send it to them.

Keep reading

All articles →

A bell curve for IQ scores from 55 to 175 with the top sliver above 150 shaded gold, showing 150 at the 99.96th percentile, about one person in 2,331.

Scores & Scales

What an IQ of 150 Means

An IQ of 150 is about 1 person in 2,331 on a standard deviation 15 scale, the 99.96th percentile. See how rare it really is, what "genius" cutoffs actually mean, and why scores this high need a real test to trust.

September 23, 20266 min read

Scores & Scales

What an IQ of 160 Means

An IQ of 160 is about 1 person in 31,600 on a standard deviation 15 scale, but only 1 in 161 on the Cattell scale. What a 160 means on each scale, which tests can print one, how accurate it is, what it predicts and where the claims come from.

September 21, 20268 min read

Scores & Scales

Average IQ in Canada (vs the US)

Average IQ in Canada vs the US: both are 100 on their own tests, so compare skills instead. Canada is ahead on PISA 2025 maths by 22 points and on adult skills by 16. See the provinces, why Canadian and US WAIS-IV norms give different scores, and how far Canada has fallen since 2012.

October 27, 20268 min read

Average IQ in Australia

Posted on October 27, 2026 by Mahesh

Scores & Scales

Average IQ in Australia: Scores, Rankings and Skills Data

Average IQ in Australia is 100 on an Australian-normed test, so compare skills instead. Australia ranks 5th of 137 online but 10th on PISA 2025 school tests, trails Canada and the UK in maths, leads on research output and shows an 80-point gap by family background. See how Australian IQ norms are built too.

By Leon Falkner, Data & Research AnalystPublished October 27, 2026 · 7 min readReviewed by Aarav Vardhan, Editor-in-Chief

Bubble chart of the 20 highest-scoring countries in an online IQ test table, average score against number of test-takers on a log scale: Australia scores 104.45 from only 4,245 takers, with a side panel showing countries with under 1,000 takers average 93.6 and those with 10,000 or more average 100.1.

The average IQ in Australia is 100 on any IQ test normed on Australians, so the real question is how Australia compares with other countries on measures that are comparable. On those, Australia sits in the upper tier: about 99.2 in the studies behind our average IQ by country table, 104.5 among online test-takers (fifth of 137, from a small crowd), and 493 on the OECD’s PISA 2025 school tests, 10th of 86 systems on our index and well above the OECD average. This guide sets out each number, the gaps inside Australia, the fall in scores since 2012, and what Australian IQ norms actually are.

What is the average IQ in Australia?

By definition, 100. A test such as the Wechsler Adult Intelligence Scale, fourth edition, Australian and New Zealand adaptation (WAIS-IV A&NZ, published in 2008 for ages 16 to 90) is standardised on a local sample and rescaled so that the typical person scores 100 with a spread of 15. That is also why Canada and the UK and the US all average 100 on their own tests. A national “average IQ” only becomes a comparison when the same instrument is used in several countries, and the table shows where Australia lands on each such measure.

Australia on five measures
Measure	Australia	Note
IQ tests, each normed on its own country	100	None by construction
Pooled IQ and school-test studies (our country table)	99.2	5,962 people in 6 studies
Online International IQ Test, 2026 table	104.5	5th of 137, but only 4,245 takers
PISA 2025, average of maths, reading, science	493	10th of 86 on our combined index
Adult skills 2023	Not in the table	Australia is not among the 31 entries we use

Why is Australia 5th online but 10th on school tests?

The 104.45 comes from the online International IQ Test table that our report on the 2026 country ranking analysed, where Australia ranks fifth of 137 countries behind South Korea, China, Japan and Iran, ahead of Russia, Singapore, Mongolia and New Zealand. It rests on 4,245 test-takers, a modest crowd. Across the table, countries with fewer than 1,000 takers average 93.6, those with 1,000 to 9,999 average 98.3, and those with 10,000 or more average 100.1, and four of the top ten (Australia, Singapore, Mongolia and New Zealand) have fewer than 10,000 takers. The pattern fits a selection effect: who chooses to take an online test, and how they found it, varies by country, so a small crowd can drift well above or below the national picture.

Sampling error is not the issue here. With 4,245 takers and a spread near 15, the mean is pinned to about ±0.2 of a point by chance alone; the uncertainty is about who took the test. The school-test comparison, by contrast, samples more than 14,600 sampled 15-year-olds from 766 schools, which is why it is the sturdier of the two comparisons when they disagree.

Australia vs Canada, the UK, New Zealand and the US on PISA 2025

PISA 2025 is the most recent like-for-like comparison of English-speaking school systems. Australia averages 493 across maths, reading and science, a little behind Canada (495), the UK (498) and New Zealand (495) and ahead of the US (485). The differences between Australia and its closest peers are small, a few points, and several sit inside the margin of error. The one clear weakness is maths, where Australia (478) trails Canada (485) and the UK (488).

PISA 2025, age 15, English-speaking countries
Country	Maths	Reading	Science	Average	Low in all three (%)
Canada	485	490	510	495	12.0
United Kingdom	488	494	511	498	12.0
New Zealand	480	497	509	495	13.8
Australia	478	491	509	493	14.0
United States	463	490	502	485	17.5

Against the OECD as a whole, Australia is comfortably ahead in all three subjects. Its share of students scoring low in all three, 14.0%, is higher than Canada’s or the UK’s (about 12%) and well below the US figure of 17.5%. The Australian Council for Educational Research (ACER) ranked Australia 10th in science and in reading and 16th in maths among the countries that took part in both 2022 and 2025; our index covers all 86 systems, including some that were not in the 2022 round, so its places differ.

Australia against the OECD average, PISA 2025
Subject	Australia	OECD average	Difference
Maths	478	463	+15
Reading	491	461	+30
Science	509	482	+27
Learning in the Digital World (new in 2025)	532	500	+32

Does Australia punch above its weight in research?

One part of our smartest-countries index has nothing to do with school tests: it counts articles in the core scientific journals from 2021 to 2025 for every million residents (OpenAlex data). On that measure Australia, with a population under 27 million, ranks 9th of 86 and is well ahead of the other English-speaking countries here, which lifts its 12th place on school skills to 10th overall. Research output is not intelligence either, but it is a harder-to-game national indicator than an online quiz average.

Journal articles per million people per year, 2021-2025
Country	Articles per million per year	Rank among 86
Australia	3,708	9th
United Kingdom	2,823	16th
New Zealand	2,703	19th
Canada	2,662	20th
United States	1,825	29th

Gaps inside Australia

The national average hides large differences. ACER’s report on the 2025 results measures four of them, in PISA points, where a gap of about 20 points is a rough rule of thumb for a year of schooling:

Score gaps inside Australia, PISA 2025 (points)
Comparison	Science	Reading	Maths
Highest vs lowest socio-economic quarter	80	74	83
First Nations students vs other students (lower by)	72	76	74
Major cities vs regional schools (higher by)	24	27	30
Independent vs government schools (higher by, before adjusting for background)	about 50	about 50	about 50

The socio-economic gap is the biggest and most consistent: students from the top quarter of family background outscore those from the bottom quarter by roughly four school years in maths on that rule of thumb. The gap between independent and government schools shrinks substantially once students’ backgrounds are taken into account, according to ACER, so it mostly reflects who attends rather than what each sector adds. First Nations students’ science scores rose 16 points between 2022 and 2025, the one bright spot in that comparison. None of these are differences in ability; they are differences in opportunity and measured outcomes.

Your own number

Where would your own score land?

Find your IQ score now! →

Secure & encryptedInstant results10–20 minutes

Australian scores have been falling since 2012

Like most countries, Australia has slipped. The OECD’s own comparison with 2012 shows a clear fall in maths of about 27 points and a clear fall in reading of about 21, with science down 12 but within the margin of error. Our combined school score is also down about 5 points since 2022. Falling test scores do not mean Australians have become less intelligent; they reflect changes in curriculum, attendance, reading habits and screen use that tests of school learning pick up and IQ tests largely do not. For why norms drift, see why IQ norms expire.

Change in Australia’s PISA scores, 2012 to 2025
Subject	Change (points)	Standard error	Verdict
Maths	-27	4.5	Clear fall (z = -5.9)
Reading	-21	6.6	Clear fall (z = -3.2)
Science	-12	6.6	Within error (z = -1.9)

How are IQ tests normed in Australia?

Australia shares its test norms with New Zealand. Pearson’s documentation lists a WAIS-IV A&NZ adaptation from 2008, and for children the WISC-V A&NZ was standardised on 528 children aged 6 to 16, stratified to match the 2011 Australian and 2013 New Zealand censuses, with about 5% of the Australian sample identifying as Aboriginal or Torres Strait Islander and 23% of the New Zealand sample as Māori (figures from Pearson’s own training slides, which we have not seen in the technical manual). Because the reference sample mixes both countries, a typical Australian and a typical New Zealander are both defined as 100 against the same tables.

Norms also age. A table built from a 2011 census sample reflects 2011 children, so test publishers renorm every decade or so, and a WAIS-5 A&NZ is now on sale. Our guide to the WAIS-IV and WAIS-5 index scores explains what changed.

How many Australians have a very high IQ?

On a bell curve with an average of 100 and a spread of 15, applied to Australia’s population of about 26.7 million (World Bank, 2023), the upper tail looks like this. These are theoretical counts, not a census of test scores.

High scores among about 26.7 million Australians (normal-curve arithmetic)
IQ score	Rarity	People in Australia
130 (Mensa line)	1 in 44	610,000
145	1 in 741	36,000
155	1 in 8,140	3,300
160	1 in 31,600	840

The 130 row is the usual line for Mensa; the rarer rows are explained in our guides to an IQ of 155 and an IQ of 160.

Is Australia smarter than the UK or New Zealand?

The honest answer is that the data cannot say. On PISA 2025, Australia is a few points behind the UK and Canada and level with New Zealand; on the online table, Australia is ahead of all three; on pooled studies Australia, Canada, the UK and New Zealand are within one point of each other. Gaps this small, from instruments this different, are noise. Our report on what country IQ rankings cannot support explains why a league table of national IQ is not something the evidence backs, and the smartest countries index shows the school and research data for 86 systems.

To see where you stand yourself, take the IQ Metrics IQ test and read the result against the IQ classification bands.

Sources

Australian Council for Educational Research (8 September 2026): Australian students continue to shine in the world’s largest learning assessment (means, rankings, gaps).
OECD (2026): PISA 2025 results, Volume I, Table I.B1.2a, as processed in our Smartest Countries Index (changes since 2012, low and top performers, rank among 86 systems).
Pearson Clinical Australia: WAIS-IV A&NZ product information; WISC-V A&NZ standardisation (webinar slides).
IQ Metrics: Average IQ by country (201 entries) and the 2026 online ranking of 137 countries. Population: World Bank (2023).

Share this article

Know someone who keeps seeing these numbers quoted without the scale they were measured on? Send it to them.

Keep reading

All articles →

Map of North America with Canada in blue and the United States in orange, above three paired scores: pooled IQ estimate 99.5 for Canada against 97.4 for the US, online IQ test average 102.1 against 101.0, and PISA 2025 maths 485 against 463.

Scores & Scales

Average IQ in Canada (vs the US)

October 27, 20268 min read

Scores & Scales

Average IQ in the UK vs the US

The average IQ in the UK vs the US is 100 in both, because each country’s tests are normed on its own people. See what can be compared: PISA 2025 maths (UK 488, US 463), adult skills (266 vs 251), online samples, the Flynn effect in each country and where the real gaps sit.

October 26, 202610 min read

Understanding IQ

PISA Test vs IQ Test

Is PISA an IQ test? No. PISA surveys applied school literacy across whole education systems and returns no individual score, while an IQ test diagnoses one person. See the new PISA 2025 results, published September 2026, and the real debate over using PISA as a national IQ proxy.

September 28, 20267 min read

Average IQ in Canada (vs the US)

Posted on October 27, 2026 by Mahesh

Scores & Scales

Average IQ in Canada vs the US: What the Data Shows

By Leon Falkner, Data & Research AnalystPublished October 27, 2026 · 8 min readReviewed by Aarav Vardhan, Editor-in-Chief

The average IQ in Canada is 100 on any IQ test normed on Canadians, and the average IQ in the US is 100 on a test normed on Americans, so an IQ test cannot say which country scores higher. What can be compared are pooled estimates and skills tests, and they put Canada slightly ahead: about 99.5 against 97.4 in the studies behind our average IQ by country table, 102.1 against 101.0 among online test-takers, and 495 against 485 on the OECD’s PISA 2025 school tests. This guide lays out each source, how big the Canada–US gap really is, what the provinces show, and why the same Canadian can score differently depending on which norm table is used.

What is the average IQ in Canada?

By construction, 100. Every major IQ test is standardised on a representative sample of one country’s people and then rescaled so that the middle of that sample scores 100 with a spread (standard deviation) of 15. A Canadian-normed test therefore returns 100 for the typical Canadian, and a US-normed test returns 100 for the typical American. The same rule is why the UK and the US both average 100 and why the figure is not a fact about a country’s brains. For a score of 100 in plain terms, see what an IQ of 100 means.

To say anything about Canada relative to another country you need something measured the same way in both. There are three kinds of evidence, and the table puts Canada and the US side by side on each.

Canada and the US on five measures
Measure	Canada	United States	What it tells you
IQ tests, each normed on its own country	100	100	None by construction
Pooled IQ and school-test studies (our country table)	99.5	97.4	2 points; sizes 4,229 vs 62,649 people
Online International IQ Test, 2026 table	102.1	101.0	1 point; 7,685 vs 22,605 takers
PISA 2025, average of maths, reading, science	495	485	10 points; driven by maths
Adult skills 2023, average of three	267	251	16 points; 9th vs 21st of 31

Read the table from top to bottom and the pattern is steady but modest: Canada is ahead on every comparable measure, by a point or two on the IQ-style estimates and by more on the skills tests. None of the gaps tells you about the genes or the potential of either population; they describe schooling, adult learning and who happened to take a test.

Why do estimates of Canada’s IQ differ?

Because they measure different groups with different instruments. The 99.5 comes from a pool of 4,229 people in ten studies, a modest base; the US figure (97.4) pools 62,649 people in 58 studies, the largest sample in that dataset. The 102.1 comes from 7,685 Canadians who chose to take an online test in 2026, in a table where Canada ranks 13th of 137 countries and the US 18th at 101.0 from 22,605 takers (our report on that table).

The sampling error in the online figure is tiny: if the scores spread about 15 points, 7,685 takers pin the mean down to roughly ±0.2 of a point. The problem is not chance but selection. People who search for an IQ test are not a random slice of the country, and the share who do so differs from place to place. In the same table, countries with fewer than 1,000 takers average 93.6 and countries with 10,000 or more average 100.1, which is a sign that crowd size and who shows up matter as much as national ability. A one- or two-point gap between Canada and the US, in any of these sources, is inside that noise.

Canada vs the US on PISA 2025

PISA tests 15-year-olds in maths, reading and science every three years, and the 2025 results are the freshest comparison of the two school systems. Canada averages 495 across the three subjects and the US 485, but the gap is not spread evenly. It is large in maths, where the difference of 22 points is about 3.9 standard errors, and it disappears in reading, where the two countries are separated by a third of a point. Science favours Canada by 8 points, which is within the margin of error.

PISA 2025, age 15: Canada against the US
Measure	Canada	US	Gap	Reading of the gap
Maths	485	463	22	Clear (z = 3.9)
Reading	490	490	0	No gap (z = 0.1)
Science	510	502	8	Within error (z = 1.4)
Top performers in at least one subject (%)	17.4	17.8	-0.4	Level
Low performers in all three subjects (%)	12.0	17.5	5.5	Canada lower
Rank among 86 school systems (school-skills score)	10th	15th	5 places

A caution that applies to both countries: the OECD flags Canada and the United States (along with Albania, the Netherlands, New Zealand and Norway) for sampling caution in 2025, a flag that the sample did not fully meet the OECD’s technical standards, so the averages carry extra uncertainty. The US is also the less precise of the two, with standard errors three times Canada’s (5.4 against 1.9 in maths), which is part of why a 10-point gap can be clear in one subject and not in another. For what these tests do and do not share with IQ, see PISA vs an IQ test and our PISA 2025 results report.

Adult skills: Canada 9th, the US 21st of 31

The OECD’s Survey of Adult Skills (PIAAC 2023) tests adults aged 16 to 65 in literacy, numeracy and adaptive problem solving. Canada averages 267 across the three, ninth among the 31 countries and regions in our table, against 251 for the US in 21st. The largest gap is in numeracy, 22 points. Adults are closer to a general-population picture than a school sample, and this is the comparison where Canada’s lead is clearest.

Survey of Adult Skills 2023: Canada against the US
Skill	Canada	US	Gap
Literacy	271	258	13
Numeracy	271	249	22
Adaptive problem solving	259	247	12
Average of the three	267	251	15

Canada’s scores have fallen faster than America’s since 2012

Both countries have slipped in maths since PISA 2012, and the OECD’s own comparison shows Canada dropping further. The picture differs by subject: Canada fell clearly in all three, while the US change is within error in reading and science. In reading the extra Canadian fall (about 25 points) is itself large enough to be unlikely to be chance; in maths the two falls are not statistically different from each other.

Change in PISA scores, 2012 to 2025 (points)
Subject	Canada	Canada, tested against its error	US	US, tested against its error
Maths	-33	Clear fall	-19	Clear fall
Reading	-33	Clear fall	-8	Within error
Science	-15	Clear fall	+4	Within error

Falling scores and a high rank can coexist: Canada is still 10th of 86 school systems on the combined score. For the wider decline and why it does not mean people have become less intelligent, see why IQ norms expire.

Your own number

Where would your own score land?

Find your IQ score now! →

Secure & encryptedInstant results10–20 minutes

Which Canadian provinces score highest?

Canada’s provinces run their own school systems, so PISA reports them separately. The Council of Ministers of Education, Canada, said Alberta, Ontario, British Columbia and Quebec performed at the top of the provinces in science. Alberta’s government reported scores of 520 in science, 499 in reading and 489 in maths for the province, first among provinces in science and reading and second in maths behind Quebec. Commentary in Policy Options notes the gap between the top province in science and the lowest, New Brunswick, is more than two grade levels, while a second piece there warns that many ranking differences between neighbouring provinces fall within measurement error. We have not seen the full provincial table, so treat single rankings as indicative.

There is no credible province-by-province “IQ”: no IQ test has been normed to compare the provinces, and the school results above reflect funding, language of instruction, immigration patterns and curriculum as well as anything about students.

Do Canadian and US IQ norms give different scores?

Yes, and this is the most Canada-specific fact in the IQ literature. Psychologists in Canada can score the WAIS-IV with either the US norm tables or a separate Canadian set. In a sample of 432 Canadian college and university students who had been referred for assessment, Harrison and colleagues found that the Canadian norms produced systematically lower IQ, index and subtest scores than the American norms, with agreement between the two systems (scores within five points, or in the same classification range) of only 49% to 76%, depending on the index. Because the sample was a clinical referral group it cannot say what the general population would get, but it shows the mechanism: an “average” score is relative to whichever reference group sits behind the table.

The practical lesson is the same one the Flynn effect teaches: before comparing two scores, check which norm table and which year they were converted with. A result of 100 means “typical of the norm sample”, not “typical of the country”.

How many Canadians have a very high IQ?

Assuming a bell curve with an average of 100 and a spread of 15, and a population of about 40 million (World Bank, 2023), the upper tail works out as follows. The numbers are theoretical counts for a normal curve, not a census of test scores.

High scores among about 40 million Canadians (normal-curve arithmetic)
IQ score	Rarity	People in Canada
130 (Mensa line)	1 in 44	910,000
145	1 in 741	54,000
155	1 in 8,140	4,900
160	1 in 31,600	1,300

The 130 line is the usual entry point for Mensa, and the rarer rows are covered in our guides to an IQ of 155 and an IQ of 160.

So is Canada smarter than the US?

Not in any way an IQ test can establish. On the evidence above, Canadian 15-year-olds do better in maths and Canadian adults do better in numeracy and literacy, by margins that are real but modest, while reading and science for teenagers are about level. The online and pooled IQ estimates lean the same direction by one or two points, which is within the noise of how they were collected. Anyone claiming a national IQ difference from numbers like these is stretching them; our report on what country IQ rankings cannot support sets out why, and the smartest countries index shows the school and research data for all 86 systems.

If you are curious where you stand rather than your country, take the IQ Metrics IQ test and read the result against the IQ classification bands. For another English-speaking comparison, see the average IQ in Australia.

Sources

OECD (2026): PISA 2025 results, Volume I, Table I.B1.2a (maths, reading and science means; changes since 2012), as processed in our Smartest Countries Index.
OECD (2024): Survey of Adult Skills 2023 (PIAAC), Annex A tables.
Council of Ministers of Education, Canada (September 2026): OECD report ranks Canadian students among top performers globally in science; provincial scores for Alberta as reported by the Government of Alberta; Policy Options (September 2026) on provincial gaps.
Harrison, A.G., Armstrong, I.T., Harrison, L.E., Lange, R.T. and Iverson, G.L. (2014): Comparing Canadian and American normative scores on the Wechsler Adult Intelligence Scale-Fourth Edition, Archives of Clinical Neuropsychology.
IQ Metrics: Average IQ by country (201 entries) and the 2026 online ranking of 137 countries. Population: World Bank (2023).

Share this article

Know someone who keeps seeing these numbers quoted without the scale they were measured on? Send it to them.

Keep reading

All articles →

Paired dot chart of UK and US PISA 2025 scores: maths 488 against 463 (a clear gap), reading 494 against 490 and science 511 against 502 (within error), with bars showing 12.0 percent of UK and 17.5 percent of US students low in all three subjects, beside the line 100 equals 100 for average IQ.

Scores & Scales

Prader-Willi Syndrome and IQ

Posted on October 26, 2026October 27, 2026 by Mahesh

Understanding IQ

Prader-Willi Syndrome and IQ: Average Scores, Subtypes and Strengths

By Alden Cross, Intelligence Research WriterPublished October 26, 2026 · 10 min readReviewed by Aarav Vardhan, Editor-in-Chief

Prader-Willi syndrome and IQ: people with the syndrome typically have a mild intellectual disability, with average scores around 60, roughly 40 points below the general population and individual results ranging from the 40s to the average range. A UK population-based study found an approximately normal distribution shifted about 40 points down, a study of 26 young people found a mean of 62.3 (range 39 to 96), and adults in a French cohort had a median Full Scale IQ of 52 on the WAIS-III. The genetic subtype changes the profile more than the total: people with maternal uniparental disomy (UPD) tend to have higher verbal IQ, and those with a deletion higher performance IQ. This guide covers the genetics, the published IQ figures, the deletion-versus-UPD differences, the visual-spatial strengths (including jigsaw puzzles), what a growth hormone trial found and how IQ fits into diagnosis. It is general information, not medical advice: questions about a particular child or adult belong with their specialist team.

Average: about 60 in two samples (62.3 in 26 youths; roughly 40 points under the population mean in a UK population sample) and a median of 52 in French adult inpatients.
Genetics: loss of paternally expressed genes on chromosome 15q11.2-q13: a deletion in 65 to 75% of cases, maternal UPD in 20 to 30% and an imprinting defect in 1 to 3%.
Subtypes: a meta-analysis of 744 cases found full-scale IQ 2.7 points lower with a deletion, verbal IQ 7.5 lower and performance IQ 4.0 higher than with UPD.
Strength: visual-spatial tasks. Children with the syndrome placed more than twice as many jigsaw pieces as typically developing children in one study.

What is Prader-Willi syndrome?

Prader-Willi syndrome (PWS) is a rare genetic disorder caused by the lack of expression of genes on the paternally inherited copy of chromosome 15q11.2-q13. There are three molecular routes to it: a deletion of that region on the father’s chromosome (65 to 75% of cases), two copies of the mother’s chromosome 15 with none from the father, called maternal uniparental disomy (20 to 30%), and an imprinting defect (1 to 3%). Newborns are floppy and feed poorly. Later come short stature, an insatiable appetite (hyperphagia), delayed development, cognitive disability and behavioural difficulties, probably because of hypothalamic dysfunction. Population studies put the birth incidence between about 1 in 15,830 (Victoria, Australia, 2003 to 2012) and 1 in 26,676 (Flanders, 1993 to 2001), with 1 in 22,000 in a UK population-based study. The same region on the mother’s chromosome is the one that goes missing in Angelman syndrome, its sister disorder, which is usually too severe for a standard IQ test.

Cognitive disability is one feature among many, alongside hunger, weight and health. That context matters for how IQ is read: the score is one input into a picture that also includes adaptive skills, health and behaviour. Compare the patterns in Down syndrome and IQ, Williams syndrome and IQ, fragile X and IQ, Klinefelter syndrome and IQ and Turner syndrome and IQ.

What is the average IQ in Prader-Willi syndrome?

Four studies give the clearest picture. They used different samples and tests, so the figures are not interchangeable, but they agree on the centre of the distribution.

Published IQ figures in Prader-Willi syndrome
Study	Sample	Finding
Curfs et al. (1991)	26 youngsters	Mean IQ 62.3, range 39 to 96
Whittington et al. (2004)	People over age 3 found through a UK health region, plus others	IQ roughly normally distributed, mean about 40 points below the general population
Copet et al. (2010)	99 adult inpatients, 85 able to complete the WAIS-III	Median Full Scale IQ 52 (middle half 46 to 60), Verbal 53, Performance 52.5
Roof et al. (2000)	38 people (24 with a deletion, 14 with UPD)	Verbal IQ 60.8 with a deletion, 69.9 with UPD

Against the ICD-10 bands, a score near 60 sits in the mild range and a score near 52 sits just inside it, at the border of moderate. A small share of people with PWS score in the borderline or average range: the top of the youth sample was 96. The UK study found the whole distribution shifted, not a separate low group, which is why individuals vary so much around the average.

ICD-10 bands for intellectual disability (WHO)
Band	IQ range	Adult mental age
ICD-10 F70 mild	50 to 69	Adult mental age 9 to under 12 years
ICD-10 F71 moderate	35 to 49	Adult mental age 6 to under 9 years

Two cautions on the figures. The adult median of 52 is lower than the younger samples, and its authors noted that their scores were slightly lower than most other studies, in a cohort of inpatients on a specialist unit. And the published numbers describe people who could take the test: in the French adult cohort 14 of 99 could not cope with the test situation, so they are not in the 52. See what an IQ of 65 means and IQ and mental age for what scores in these bands imply day to day, and low IQ vs learning disability for how the terms differ. Scores at either extreme of the curve are also less precise than scores near 100, because a test has few items at the ends of its range; the ceiling problem at the top is set out in the guide to an IQ of 200.

Does the genetic subtype change IQ?

The two main subtypes, paternal deletion and maternal UPD, are the best-studied split. The totals are close; the profiles differ. A meta-analysis pooling 13 studies (423 people with a deletion and 318 with UPD) found that the deletion group had a full-scale IQ 2.69 points lower (95% interval 0.52 to 4.86), a verbal IQ 7.5 points lower, and a performance IQ 4.02 points higher. Individual studies agree on the direction.

Deletion versus maternal UPD: what four sources found
Source	Sample	Verbal IQ	Performance IQ	Full-scale IQ
Roof et al. (2000)	24 deletion, 14 UPD	UPD higher by 9.1 points (69.9 vs 60.8); 50% of UPD and 17% of deletion cases at 70 or above	No difference	–
Copet et al. (2010)	57 deletion, 27 non-deletion	No significant difference	Deletion higher	No significant difference
Yang et al. (2013), meta-analysis	13 studies, 744 cases	Deletion lower by 7.5	Deletion higher by 4.0	Deletion lower by 2.7
Whittington et al. (2004)	Population-based sample	UPD better verbal ability	Some deletion cases strong visuospatial; UPD weaker coding	–

A gap of about three points in full-scale IQ is smaller than the margin of error of a single score (see the margin-of-error explainer), so the subtype tells you little about the total and more about the shape. A related split is the size of the deletion. In 96 people (49 with UPD and 47 with a deletion), Milner and colleagues found few cognitive differences between the larger type I and smaller type II deletions, though ability tended to be lower with type I. UPD cases showed more autistic-like social difficulty. For how an uneven profile shows up on a test, see verbal vs non-verbal IQ scores.

Strengths and weaknesses: the cognitive profile

The profile is uneven, and the strongest finding is visual-spatial. In the 26-youth study, the highest scores came on performance tasks, with Block Design the best subtest in nine children and Coding or Mazes in five, and the group did better on visual-motor discrimination than on auditory skills. Dykens (2002) tested the old clinical impression that people with PWS are unusually good at jigsaws. Children with PWS scored higher than age- and IQ-matched peers with mixed intellectual disability on standard visual-spatial tasks (Object Assembly, Triangles and a visual-motor integration test), though below typically developing children of average IQ. On jigsaw puzzles they placed more than twice as many pieces as the typically developing group, and scored on par with them on word searches. Puzzle skill was not predicted by IQ, age, gender, obesity or obsessive-compulsive symptoms, but by genetic subtype.

A later study of 26 people with PWS and 26 typically developing children matched for mental age found that the PWS group solved puzzles by piece shape while the comparison children used the picture. The pattern is the reverse of Williams syndrome, where jigsaws are a weakness. The practical point for IQ is that a single Full Scale score hides a spiky profile: a performance strength can sit beside a verbal weakness, and which pattern appears depends on the subtype.

Does growth hormone change IQ in Prader-Willi syndrome?

One randomised trial looked directly. Siemensma and colleagues followed 50 prepubertal children aged 3.5 to 14, first in a two-year randomised trial and then for four years of treatment, testing cognition every two years with short forms of the WPPSI-R or WISC-R. Over the two-year trial, the treated children’s estimated total IQ stayed level while that of untreated controls fell, significantly so on the Similarities and Vocabulary subtests. After four years of treatment, Similarities and Block Design scores were above baseline, total IQ was unchanged from baseline, and children with the larger baseline deficits gained most. The authors concluded that treatment prevented deterioration of some skills and improved abstract reasoning and visuospatial skills. That is a finding about change over time in a trial, not advice: treatment decisions belong to the child’s endocrinologist and family.

Your own number

Where would your own score land?

Find your IQ score now! →

Secure & encryptedInstant results10–20 minutes

Does IQ predict behaviour or health in Prader-Willi syndrome?

Less than people expect. In 25 children and 61 adults with PWS, Dykens and Cassidy (1995) found no difference in maladaptive behaviour between higher and lower IQ groups, so the score did not sort people by how they behaved. A diagnosis of intellectual disability needs weaknesses in everyday adaptive skills, not a score alone. That is why two people with the same score can need very different support, and why how to read an IQ test report recommends reading the subscores and not the headline.

How is IQ tested in Prader-Willi syndrome?

Test choice. Research uses the Wechsler family: the WAIS-III in the French adult cohort and short forms of the WPPSI-R and WISC-R in the growth hormone trial. Clinical teams choose by age and ability (see the WPPSI-IV and the KBIT-2).
Who can sit it. Some adults cannot complete a full Wechsler test, so published scores are weighted toward those who can.
Reading the result. Look at the verbal and performance split and the subtests, and at the confidence interval, not only the Full Scale number.
Re-testing. Scores shift with age, treatment and test version, so compare like with like.

Quick answers

What is the average IQ of someone with Prader-Willi syndrome? About 60 in two studies (a mean of 62.3 in 26 youths and roughly 40 points under the general mean in a UK sample); a median of 52 in French adult inpatients.
Can a person with Prader-Willi syndrome have a normal IQ? A few score in the borderline to average range (the top of one sample was 96), but the whole distribution is shifted down, so it is uncommon.
Is Prader-Willi syndrome an intellectual disability? Most people with it have a mild intellectual disability. Diagnosis also requires deficits in adaptive functioning.
Which subtype has the higher IQ? Maternal UPD is about three points higher on full-scale IQ and seven higher on verbal IQ in a meta-analysis; a deletion is about four points higher on performance IQ.
Does growth hormone raise IQ? In a randomised trial it prevented a decline in untreated controls and improved some reasoning and visuospatial scores, with total IQ unchanged after four years.

The bottom line

Prader-Willi syndrome and IQ come down to a distribution shifted about 40 points below average, centred in the mild intellectual disability range, with enough spread that a few people score near average and some score in the moderate range. The subtype shapes the profile (verbal strengths with UPD, performance and visual-spatial strengths with a deletion) more than the total, and a growth hormone trial suggests scores can be protected. Read any single number alongside adaptive skills and the person’s own strengths, and to see how a score of your own is built, start with the IQ Metrics IQ test and the IQ tests hub.

Sources

Curfs, L.M.G. et al. (1991): Strengths and weaknesses in the cognitive profile of youngsters with Prader-Willi syndrome, Clinical Genetics.
Whittington, J. et al. (2004): Cognitive abilities and genotype in a population-based sample of people with Prader-Willi syndrome, Journal of Intellectual Disability Research.
Copet, P. et al. (2010): Cognitive profile in a large French cohort of adults with Prader-Willi syndrome: differences between genotypes, Journal of Intellectual Disability Research.
Roof, E. et al. (2000): Intellectual characteristics of Prader-Willi syndrome: comparison of genetic subtypes, Journal of Intellectual Disability Research.
Yang, L. et al. (2013): Psychiatric illness and intellectual disability in the Prader-Willi syndrome with different molecular defects: a meta analysis, PLoS ONE.
Milner, K.M. et al. (2005): Prader-Willi syndrome: intellectual abilities and behavioural features by genetic subtype, Journal of Child Psychology and Psychiatry.
Dykens, E.M. (2002): Are jigsaw puzzle skills “spared” in persons with Prader-Willi syndrome? and Verdine, B.N. et al. (2008): Strategies and correlates of jigsaw puzzle and visuospatial performance by persons with Prader-Willi syndrome; Ly, T.M. and Hodapp, R.M. (2005): Children with Prader-Willi syndrome vs. Williams syndrome on a jigsaw puzzle task.
Siemensma, E.P.C. et al. (2012): Beneficial effects of growth hormone treatment on cognition in children with Prader-Willi syndrome, Journal of Clinical Endocrinology & Metabolism.
Dykens, E.M. and Cassidy, S.B. (1995): Correlates of maladaptive behavior in children and adults with Prader-Willi syndrome, American Journal of Medical Genetics; Angulo, M.A., Butler, M.G. and Cataletto, M.E. (2015): Prader-Willi syndrome: a review of clinical, genetic, and endocrine findings, Journal of Endocrinological Investigation.
Birth incidence: Lionti, T. et al. (2015), a population-based profile of 160 Australians; Butler, J.V. et al. (2002), a population-based study; Vogels, A. et al. (2004), Flanders. ICD-10 bands as set out in our IQ of 65 guide.

Share this article

Know someone who keeps seeing these numbers quoted without the scale they were measured on? Send it to them.

Keep reading

All articles →

Understanding IQ

Down Syndrome and IQ

September 16, 202610 min read

Understanding IQ

Williams Syndrome and IQ

October 10, 20269 min read

Scores & Scales

What Does an IQ of 65 Mean?

An IQ of 65 is the 1st percentile: about 1 person in 102 scores it or lower. It sits inside the ICD-10 mild range, is labelled Extremely Low on Wechsler tests and has a likely true score of about 60 to 73. See its mental age, what it means for daily life and how courts treat it.

October 23, 202611 min read