Identical twins raised in different families differ by 15 IQ points on average when their schooling was very different, compared with under 6 points when it was similar. That is the striking figure from a reanalysis published in 2025. Yet it rests on about ten pairs and on old cases. In 2026, a Swedish study of 440 German twin pairs concluded that genes explain most of the link between IQ and success in education and work. Do you have to pick a side? No, as long as you understand what "heritable" means.
- Twins raised apart: an average gap of 5.8 points with similar schooling, 15.1 points with very different schooling (87 pairs in total, about ten in the most contrasted group).
- German twins: IQ measured around age 23 predicts education and occupation around age 27, and genetic factors explain 69 to 98% of that link. Neither income nor wealth was measured.
- Key point: genes weigh heavily on average, but large differences in environment can still shift one person's score.
Why Identical Twins Raised Apart Interest Researchers
Identical (monozygotic) twins share the same DNA. If they are separated early and raised by two families, the differences between them cannot come from genes: they come from the environment, and partly from the imprecision of the tests. According to the 2025 reanalysis, the large studies of these twins found a gap of about 8 IQ points and an intraclass correlation of about 0.75 (1 if both twins always get the same score, 0 if they are no more alike than two strangers). For comparison, on a scale with a mean of 100 and a standard deviation of 15, two people picked at random differ by about 17 points on average (a simple statistical calculation). This resemblance long served as an argument for a strong genetic influence on IQ.
The 2025 Reanalysis: Sorting Twins by Schooling
Jared Horvath and Katie Fabricant (Acta Psychologica, May 2025) argue that classic studies averaged dozens of pairs without accounting for life experiences such as education. They gathered every case published over the last century that gave each twin's IQ score and life history: 87 pairs, grouped by whether their schooling had been similar, somewhat different or very different.
| Schooling of the two twins | Average IQ gap | Intraclass correlation |
|---|---|---|
| Similar | 5.8 points | 0.87 |
| Somewhat different | 12.1 points | 0.80 |
| Very different | 15.1 points | 0.56 |
With very different schooling, the gap between two people with the same DNA comes close to the gap between two strangers. The authors say the use of the twin correlation as a measure of genetic influence needs to be re-examined, and that teams who only published averages should share their individual data.
What This Study Does Not Prove
- A tiny group: the "very different" group has only about ten pairs. A single unusual pair can move the average.
- Old data: cases published over a century, in different eras, with different IQ tests.
- An association, not proof of cause: nobody assigned the twins' schooling at random. The twin who had little schooling may also have grown up poorer. And a child who does well in class is more likely to stay in school.
- Not the end of genes: even in the most contrasted group, the twins' scores remain correlated (0.56).
Does School Really Raise IQ?
Yes, and stronger studies show it. In 2018, Stuart Ritchie and Elliot Tucker-Drob pooled 142 estimates from 42 data sets (over 600,000 participants), using methods that get around the chicken-and-egg problem: accounting for IQ measured before the extra schooling, using reforms that lengthened compulsory schooling for a whole generation, or comparing children born just before or just after the school-entry cut-off date. The result: about 1 to 5 IQ points per extra year of education, an effect that persists later in life. For the authors, education is the most robust method identified so far for raising intelligence. The twin finding is therefore plausible, even if its exact size remains uncertain.
The German Twins: IQ at 23, Status at 27
On 2 February 2026, Petri Kajonius (Lund University, Sweden) published in Scientific Reports an analysis of the TwinLife panel, which has followed more than 4,000 German families since 2014: 228 identical twin pairs and 212 same-sex fraternal twin pairs. IQ was measured around age 23 with a computer-based reasoning test, which assesses fluid intelligence rather than knowledge (two facets whose genetics partly differ, see Two Kinds of Smart). Around age 27, socioeconomic status was measured through educational level and occupation (job prestige, socioeconomic class).
- IQ at 23 correlates 0.27 to 0.43 with these measures. It explains only 7 to 18% of the differences in status, the author points out.
- IQ heritability is estimated at about 75% (correlation of 0.76 between identical twins, 0.32 between fraternal twins), a high figure for young adults, the author notes.
- Genetic factors explain 69 to 98% of the link between IQ and status at 27.
Watch the headlines. One English-language headline spoke of "wealth", and the Lund University press release (31 March 2026) mentions income. The paper measures neither: only education and occupation. Acknowledged limits: parents' IQ and social background were not taken into account, follow-up lasted only four years, some occupation data are missing, and the model is simplified. Funding: Lund University, no competing interests declared.
Finally, "explained by genes" does not mean "independent of school". The genetic share includes what genes do through the environment: as they grow up, people partly choose their experiences according to their inclinations, which, according to the authors of a study of 11,000 twin pairs, could explain why the heritability of IQ rises with age.
Genes or School: Why Both Findings Fit Together
Heritability is the share of the differences between people in a population, at a given time, that is linked to genetic differences. It describes a group, not an individual: 75% heritability does not mean that 75% of your IQ comes from your genes.
It also depends on the environments present: the more alike they are, the more the remaining differences come from genes. In today's Germany, schooling is long and compulsory for everyone, and differences in schooling are likely much smaller than for some separated twins of the past. For those twins, the environment varied a lot, and it shows. A classic image sums it up: in the same field, seeds grow taller or shorter depending on their genes, but the gap between a fertile field and a poor one comes from the soil.
A 2018 review describes heritability and malleability as coexisting, and notes that standard measures poorly capture interactions between genes and environment, which can inflate the share attributed to genes. For more on what DNA can and cannot explain, see 1 Million Genomes, 1,260 Variants, and DNA Still Explains Under 10% of Your Personality.
What This Means for You
- Your IQ is not fixed: high heritability is not destiny, and years of education have a measurable effect. Twin studies describe populations, not your "potential".
- An online score is an estimate: our IQ test (20 questions) and our full IQ test (80 questions, four domains) compare your score with the people who took them on the site, not with a representative sample. It is not a clinical IQ: only an assessment with a psychologist gives one.
- Giftedness is something else: our giftedness questionnaire is a self-assessment questionnaire of traits often linked with high intellectual potential, not an IQ test. None of these results is a diagnosis.
Frequently asked questions
Is IQ inherited?
Largely, at the level of a population: depending on age, twin studies attribute about 20% of IQ differences to genetic differences in infancy and perhaps 80% in later adulthood. Heritable does not mean fixed: school matters too.
Can identical twins have very different IQs?
Yes. Raised apart with very different schooling, they differed by 15 points on average in the 2025 reanalysis. But that figure comes from about ten pairs, and part of any gap reflects the imprecision of tests.
Does IQ at 23 predict wealth at 27?
The TwinLife study measured neither wealth nor income, only education and occupation. IQ explained only 7 to 18% of the differences in status.
References: Horvath JC, Fabricant K (2025). IQ differences of identical twins reared apart are significantly influenced by educational differences. Acta Psychologica 257:105072. doi:10.1016/j.actpsy.2025.105072 (reanalysis of published cases, 87 pairs of identical twins reared apart) | Kajonius PJ (2026). Longitudinal associations between cognitive ability and socioeconomic status are partially genetic in nature. Scientific Reports 16:4315. doi:10.1038/s41598-026-37786-3 (longitudinal twin study, German TwinLife panel, 228 identical and 212 same-sex fraternal twin pairs) | Lund University (31 March 2026). New twin study highlights the power of genetics (press release) | Ritchie SJ, Tucker-Drob EM (2018). How much does education improve intelligence? A meta-analysis. Psychol Sci 29(8):1358-1369. doi:10.1177/0956797618774253 (meta-analysis of quasi-experimental studies, 142 effect sizes, 42 data sets, over 600,000 participants) | Haworth CMA, Wright MJ, Luciano M et al. (2010). The heritability of general cognitive ability increases linearly from childhood to young adulthood. Mol Psychiatry 15(11):1112-1120. doi:10.1038/mp.2009.55 (11,000 twin pairs, 4 countries) | Plomin R, Deary IJ (2015). Genetics and intelligence differences: five special findings. Mol Psychiatry 20(1):98-108. doi:10.1038/mp.2014.105 (review) | Sauce B, Matzel LD (2018). The paradox of intelligence: heritability and malleability coexist in hidden gene-environment interplay. Psychol Bull 144(1):26-47. doi:10.1037/bul0000131 (review)