Does Aphantasia Affect IQ? What the Studies Measured

Does aphantasia affect IQ? Nothing published shows that it does. Aphantasia is the lifelong absence of voluntary mental pictures, and we found no study that measured IQ in a representative sample of people with it. Only two small studies report IQ-style scores for people with aphantasia, and they point in opposite directions: 115.5 against 110.0 for typical imagers in a Wechsler study of 69 volunteers, and 91.7 against 95.7 on a three-minute screener in a study of 30. Neither gap was statistically significant. Studies that looked at other abilities found little or no difference. This article sets out what each study measured, how common aphantasia is, and why the "aphantasia IQ" figures that circulate online cannot be read as a result.

What is aphantasia, and how is it measured?

The word was proposed in 2015 by Adam Zeman, Michaela Dewar and Sergio Della Sala in Cortex "to refer to a condition of reduced or absent voluntary imagery". Their paper described 21 people who had contacted the team because they had always lacked visual imagery, 19 of them male, and the authors noted that the group was self-selected. The experience was not new: Francis Galton had reported in 1880 that some people described "no power of visualising". Researchers measure imagery with the Vividness of Visual Imagery Questionnaire (VVIQ), created by David Marks in 1973, in which people rate 16 items from "no image at all" to "perfectly clear and vivid as normal vision". In the reversed scoring that Zeman's group uses, totals run from 16 to 80.

Where to draw the line is not settled. Zeman's 2020 survey and Milton's 2021 study counted a score of 16 to 23 as aphantasia, Pounder's 2022 study used 25 or below, and Dance's 2022 prevalence study counted a score of exactly 16 separately from 17 to 32. A 2023 paper in Cortex by Blomkvist and Marks noted that these cut-offs are "arbitrary and differ between studies", which makes results hard to compare.

How common is aphantasia?

It depends on how you ask. The most recent international estimate on the standard questionnaire, by David Wright and colleagues in Frontiers in Psychology in 2024, found aphantasia in 1.2% of 3,049 people and in 0.9% of a combined 9,063, with about 3% weaker than typical (hypophantasia) and about 6% unusually vivid (hyperphantasia). Other studies give other numbers:

• Zeman and colleagues, 2020, a biobank cohort of 1,288 people: 0.7% scored the minimum of 16 out of 80 and 2.6% scored 16 to 23
• Dance, Ipser and Simner, 2022, two samples of about 500 (students, then an online crowd-sourcing site): 4.2% and 3.6%, 3.9% combined, with a cut-off that reached up to 32
• Beran and colleagues, 2023, 5,010 US adults: 8.9% when people simply said they had aphantasia, but 1.5% on the questionnaire scales
• New and colleagues, 2026, nearly 2,000 UK adults aged 18 to 30 (abstract only): 13.9% had aphantasia in at least one of eight senses, which is not comparable with visual-only estimates of 3 to 4%
• Faw, 2009, as reported by Zeman and colleagues in 2015 (we did not read the original): 2.1% to 2.7% of 2,500 people said they had no visual imagination

The pattern is consistent: one self-rating question produces the high numbers and a validated questionnaire the low ones. That matters for the IQ question. At roughly 1%, a sample of 10,000 people holds only about 90 to 120 with aphantasia, and finding 100 of them takes 8,300 to 11,000 people. Most studies instead recruit people who already know they have it, from forums, magazines and university mailing lists, so the people studied are not a random slice of everyone with the trait.

What do the two studies that report IQ-style scores show?

The first is Milton and colleagues (Cerebral Cortex Communications, 2021). They compared people with aphantasia, people with hyperphantasia and people with mid-range imagery on memory tests and brain scans, and recorded IQ as a background measure using the Wechsler Abbreviated Scale of Intelligence or the Wechsler Adult Intelligence Scale. There were 69 participants in three groups, recruited from email contacts to the research team, a biobank and the university community. The IQ results in the paper's Table 1:

• Aphantasia (VVIQ 16 to 23, 24 people): mean IQ 115.5, standard deviation 12.5
• Typical imagery (VVIQ 55 to 60, 20 people): mean IQ 110.0, standard deviation 11.4
• Hyperphantasia (VVIQ 75 to 80, 25 people): mean IQ 106.3, standard deviation 4.5

The difference across the three groups was significant (F(2,66) = 5.18, p = 0.008), but the paper reports that only one pair differed: aphantasia against hyperphantasia (p = 0.002). Aphantasia against typical imagery was not significant. Our own arithmetic from the printed means and standard deviations puts the aphantasia-versus-typical gap at 5.5 IQ points, a standardised difference (Cohen's d) of 0.45, with a 95% confidence interval of -1.9 to +12.8 points. That interval fits no difference at all and it fits a difference of nearly 13 points, which is what samples of 24 and 20 people can tell you. Five and a half points is about a third of a standard deviation on the usual scale; our IQ percentile calculator shows how any starting score translates into percentiles.

The second is Monzel and colleagues (eLife, 2024), a study of autobiographical memory and brain connectivity in Bonn. It matched 14 people with congenital aphantasia to 16 controls on age, sex, education and intelligence, the last "assessed with a short intelligence screening", the three-minute mini-q rather than a full battery. The scores were 91.7 (standard deviation 16.6) for aphantasia and 95.7 (10.0) for controls, t = 0.81, p = .425. This time the aphantasia group was lower. Our interval for the gap is -4.0 points, from -14.1 to +6.1. The authors' Bayes factor of 2.29 is only weak support for "no difference", and with 30 people that is all a result like this can offer.

Put side by side, the two studies disagree about the sign. Combining them with inverse-variance weights, our calculation and a rough one because the tests and samples differ, gives +2.1 points with a 95% interval of -3.6 to +7.9. That is compatible with a small deficit, with nothing, and with a modest advantage, which is the honest summary of two small, self-selected samples.

Read the Milton study as its authors do. IQ was a background variable, not the question under test. The volunteers had to attend two sessions that included a brain scan. The hyperphantasia group's narrow spread (standard deviation 4.5 against 11.4 and 12.5 in the other two) is unusual. The paper names two Wechsler tests without saying which subtests were used or whether the figure is full-scale IQ. The authors called the aphantasia-versus-hyperphantasia difference modest, said it "introduces a complexity into the interpretation" of their results, and did not conclude that aphantasia raises IQ.

Do other studies find a difference in reasoning or general ability?

Pounder and colleagues (Cortex, 2022) compared 20 people with congenital aphantasia and 20 typical imagers on memory, planning and mental rotation tasks. They did not give an IQ battery. They matched the groups on the Wechsler Test of Adult Reading, a word-reading test used as a proxy for intelligence, and the groups did not differ: a predicted full-scale IQ of 108 (standard deviation 3.2) for aphantasia and 106.6 (4.4) for typical imagers, p = .36. The only group differences were in response time, on a planning task and, among the subgroup scoring the minimum of 16, on mental rotation. Accuracy did not differ. The authors concluded that the cognitive profile of people without imagery "does not greatly differ" from typical.

A 2025 study in Neuropsychologia by Delem and colleagues tested 45 people with aphantasia and 51 controls on questionnaires and tasks that included verbal and non-verbal reasoning and verbal and spatial working memory, and reported that initial group comparisons "revealed minimal differences". We could read only the abstract, so we cannot give the tests or the effect sizes. Kay, Keogh and Pearson (2024) found that people with aphantasia were slower but more accurate on two mental rotation tasks, favouring analytic strategies where controls favoured rotating an object in the mind.

Outside the laboratory the picture is similar. Monzel, Dance, Azañón and Simner (2023) compared the school certificates of 115 people with their imagery vividness and found moderate to strong evidence that school performance in most subjects was not associated with vividness. Xing and Nicholas (Frontiers in Psychology, 2026) gave 332 undergraduates two 25-item midterms: vividness went with how clear people felt their recall was, but predicted neither answer correctness nor overall exam scores. Working memory is the one place the research disagrees. Two 2026 papers in Neuropsychologia reach opposite conclusions: Knight, Milton and Zeman found no direct evidence of impaired visual working memory, while Zhao, Vogel and Bainbridge found aphantasic participants significantly impaired.

Why can small studies not settle it?

With 20 people per group, a study has about a 15% chance of detecting a true difference of 0.3 standard deviations, which is 4.5 IQ points, and with 25 per group about 18%. It would take roughly 176 people per group to reach 80%. (Our calculation, for a two-sided test at the 5% level.) A study that small can rule out only large gaps, so "no significant difference" means "not found", not "absent". The big survey samples have the opposite problem: they are large enough, but they contain no IQ test. Our explainer on the IQ margin of error shows how much noise sits inside a single score before group differences are even considered.

Are people with aphantasia more likely to work in science and maths?

This is where the idea that aphantasia goes with high intelligence mostly comes from. Zeman and colleagues' 2020 paper in Cortex surveyed 2,000 people with aphantasia and 200 with hyperphantasia and reported that aphantasia was associated with scientific and mathematical occupations, and hyperphantasia with creative ones. People with aphantasia were more often in computing and mathematics or in life, physical and social science jobs, and less often in arts, design, entertainment, sports and media (chi-square 84.5, p < .001).

That is a finding about careers, not test scores, and it came from self-selected respondents, which the authors acknowledge. The two groups reported the same education (67% of the aphantasia group and 68% of the hyperphantasia group had a degree or more than 15 years of education), so schooling does not explain the gap between them. But the study did not test IQ in either group, and many things beyond ability steer people towards code or towards design. Compare how our pieces on quant trading firms and the average IQ of software engineers treat the same step from a job to an IQ.

Does aphantasia matter on an IQ test?

Several IQ items look as if they need a mental picture: rotating a shape, folding paper in the mind, completing a matrix. The studies above suggest they do not, at least for accuracy: Kay and colleagues concluded that visual imagery is not needed for success on classical mental rotation tasks, because other strategies work. Speed is another matter. On a timed test a slower strategy could cost points even when the answers are right, and we found no study that compared people with aphantasia and typical imagers on the subtests of a standard IQ battery or on a timed full-length test. Our guides to paper-folding and mental rotation questions, Raven's matrices and spatial IQ tests show what those items ask, and our look at photographic memory covers the better-studied question of imagery and memory.

How should you read a claim about aphantasia and IQ?

• Was IQ tested, or inferred from jobs, education or a self-reported "high IQ"? Only two small studies we found report IQ-style scores
• Who took part? Volunteers from forums, magazines and mailing lists are not a sample of everyone with aphantasia
• What cut-off defined aphantasia: 16 to 23, 25 or below, or up to 32? The label changes who counts
• What was the comparison group: typical imagers, people with hyperphantasia or the general population? Milton's aphantasia group scored higher than the hyperphantasia group and not significantly higher than typical imagers
• How big was the study? Twenty to 25 people per group can detect only large differences

The short answer is that aphantasia has not been shown to raise or lower IQ. The two small studies with IQ-style scores point in opposite directions, the studies that checked ability found the same accuracy with a different strategy, and the "high-IQ trait" idea rests on career surveys. A measured number for yourself comes from the IQ Metrics test, and the pattern-recognition piece explains what most of its items have in common.

Common questions