Are left-handed people smarter? What the meta-analyses actually found

0 The short answer

Left-handers do not score higher on IQ tests, and the best evidence points to a difference too small to matter in either direction. The 2018 systematic review covered 36 studies and 66,108 people, and its meta-analyses of 18 of those studies found right-handers ahead by a standardized difference of 0.07, which is about 1 IQ point by our arithmetic, and the difference lost statistical significance when the largest study was excluded. Handedness differs more clearly at the low end, where left-handedness is more common among people with intellectual disability, while the evidence on giftedness rests on only a handful of studies that disagree. None of these group results can tell you anything about the score of one left-handed person.

1 What Counts as Left-Handed in These Studies, and How Many People Are?

There is no single definition of a left-hander, and the studies of handedness and IQ use at least four, which is the first reason their results are hard to compare. Some studies ask which hand a person writes with. Some use a questionnaire that asks which hand is used for a list of tasks, of which the best known is the Edinburgh inventory described by Oldfield in 1971. Some measure how fast or accurately each hand performs a task, which is hand skill and not hand preference. Others rely on an observer: in the Australian cohort discussed below, the interviewer classified each child as right-handed, left-handed or mixed-handed after the child wrote and drew.

The 2020 meta-analysis by Papadatou-Pastou and colleagues in Psychological Bulletin brought together five sets of analyses of hand preference for different manual tasks. It reports that the prevalence of left-handedness lies between 9.3 percent, under the most stringent criterion of left-handedness, and 18.1 percent, under the most lenient criterion of non-right-handedness. The best overall estimate was 10.6 percent, or 10.4 percent when studies of elite athletes were excluded. The authors argue that this is a roughly one in ten ratio, and that it varies with study characteristics such as year of publication and the way handedness was measured, and with participant characteristics such as sex and ancestry. They also write that there is no universal agreement on how to assess handedness.

Sex matters for any comparison. A 2008 meta-analysis of 144 studies and 1,787,629 participants, whose correction notice in Psychological Bulletin reprints the abstract, found a male to female odds ratio of 1.23 for left-handedness, with a 95 percent confidence interval of 1.19 to 1.27. The notice corrects a statement about one moderator, weakening the evidence that writing hand differed from other instruments, and it says the broad conclusions are unchanged. Because men and women differ slightly in how often they are left-handed, and the page on average IQ by gender shows that their score distributions differ slightly in spread, a careless comparison could blur the two.

Three words recur in the research, and the rest of this page keeps them apart. Left-handed means the left hand is preferred for the criterion used. Non-right-handed lumps left-handers and mixed-handers together, which is why it yields the lenient 18.1 percent estimate. Mixed-handed means no consistent preference, which the studies define differently. A result for one of these groups does not transfer to another, and the sections below show that the mixed group in particular behaves differently from the left-handed group.

Two neighboring questions are handled elsewhere. The page on left brain versus right brain covers the hemisphere myth, including what handedness does and does not say about personality. The page on IQ and creativity covers creative achievement. This page is about measured IQ and the abilities a full test samples, as the page on what IQ measures and the page on the CHC model describe them.

2 What Did the 2018 Meta-Analysis Find About Handedness and IQ?

The 2018 systematic review and meta-analyses by Ntolka and Papadatou-Pastou found a statistically significant but marginal difference favoring right-handers, and it depended on a single study. The paper, titled Right-handers have negligibly higher IQ scores than left-handers, appeared in Neuroscience and Biobehavioral Reviews. According to its abstract, the systematic review covered 36 studies with 66,108 individuals that measured full IQ scores in different handedness groups. Eighteen of those studies, with 20,442 individuals, were further included in three sets of meta-analyses that compared standardized mean IQ scores in left-handers, in non-right-handers and in mixed-handers with those of right-handers.

The abstract reports that the bulk of the studies in the systematic review found no differences between left-handers and right-handers. In the meta-analyses, the difference between right-handers and left-handers was statistically significant but marginal in magnitude, with d of minus 0.07, and the data sets were homogeneous. Significance was lost when the largest study was excluded. No differences in mean IQ were found between right-handers and non-right-handers, or between right-handers and mixed-handers, and no sex differences were found. The authors conclude that the intelligence differences between handedness groups in the general population are negligible.

In a plain language summary for Atlas of Science, dated January 4, 2018, its second author, Marietta Papadatou-Pastou, wrote that 36 published studies had measured full IQ scores with standardized tests in healthy left-handers and right-handers, and that 18 of them provided the numbers needed for a meta-analysis. She illustrated the result with a case in which left-handers average 100 and right-handers 101.05, on a scale with a standard deviation of 15. She also wrote that the difference was no longer there when the study with the largest sample size was excluded. The same summary warns that the finding concerns healthy individuals and does not hold for atypical groups, such as people with intellectual disabilities and gifted people, which later sections of this page take up.

Three features of the abstract deserve attention before any conclusion. The first is the unit. A standardized mean difference of 0.07 is a difference of seven hundredths of one standard deviation, and the page on why the standard deviation is 15 explains why multiplying by 15 gives the number of points, about 1.05. That multiplication is our arithmetic, not a figure the authors report in the abstract, although their illustration of 101.05 is the same number. The second is the word homogeneous, which means the study results were statistically consistent with one common effect, so there was no sign of a scatter that a hidden moderator would create. The third is the loss of significance without the largest study, which is a statement about fragility and is the subject of a later section.

3 Every Study Side by Side

Read by design, the studies of handedness and IQ range from no difference in meta-analyses of healthy people to lower scores for left-handed and mixed-handed children in large cohorts, and each row measures something slightly different. The table below lists the studies this page relies on. Results are quoted as each source reports them, and the cell says where a figure is our arithmetic. A study that tests prevalence of handedness in a diagnostic group asks a different question from one that compares IQ means, so the third column states which.

The table has two kinds of rows. The first four test whether IQ or ability differs by handedness, in adults and in children, and they do not agree on the size of any gap. The next three ask about the tails of the distribution or about outcomes downstream of ability. The last two bear on mechanisms and on whether a related cognitive difficulty goes with hand preference. The sections that follow take them in that order, beginning with why the headline gap is so fragile.

4 Why Does the Gap Disappear When the Largest Study Is Removed?

A difference of 0.07 standard deviations that survives only while one study stays in the analysis is a weak signal, and the arithmetic of what it would mean even if real is small. In a meta-analysis each study is weighted by its precision, and a very large study carries most of the weight. If removing it moves the pooled estimate from significant to not significant, the pooled result is mostly that study's result. The abstract does not name the largest study or give its size, so this page does not either, and it does not claim to know why that study differed.

Our arithmetic on the published number shows how little is at stake. These are our calculations, assuming normal distributions with equal spread and a left-handed share of 10.6 percent, the best estimate from the 2020 prevalence meta-analysis, and they are not figures any author reports.

If left-handers averaged a point lower than right-handers, then roughly 47 out of 100 left-handers would still score above the right-handed average, against 50 out of 100 right-handers. Picking one person from each group, the right-hander would score higher about 52 times in 100. Handedness would explain well under one tenth of one percent of the differences between people in IQ. By comparison, the page on IQ and height and the page on what IQ correlates with place much larger correlations in perspective, and even those leave most individual variation unexplained.

A point is also small next to the noise in any single score. The page on reliability and validity explains that a person tested twice on different days will not get an identical number, and the page on IQ score versus percentile shows why a point near the middle of the scale moves the percentile only a little. Population averages can be statistically detectable with 20,442 people while still being irrelevant for predicting what any one person will score. That is the logic behind the word negligible in the paper's title.

Two cautions keep the interpretation honest. First, a nonsignificant result without the largest study does not prove a difference of exactly zero. It means the data are compatible with zero, with the existing difference, and with values on either side. Second, the abstract's statement about homogeneity means the studies were consistent with one effect size, and not that the studies were all large or well measured. A set of small studies can be homogeneous because each is too imprecise to disagree.

5 Do Hand Skill and Mixed-Handedness Tell More Than Left Versus Right?

Studies that measure how skillful each hand is, instead of which hand a person prefers, find that the weakest scores sit near the point where the two hands are equally skilled, which is not the same as being left-handed. This is the pattern in two large samples, and it is the main reason the mixed group behaves differently from the left-handed group in the literature.

Crow and colleagues used the National Child Development Study, a UK national cohort, to test whether relative hand skill predicts ability. They had 12,770 individuals and measured relative hand skill with a square checking task, then related it to verbal, non-verbal and mathematical ability and to reading comprehension at age 11. Their abstract reports some modest decrements in extreme right-handers, but the most substantial deficits near the point of equal hand skill, which they call hemispheric indecision. For verbal ability girls performed better than boys, yet the relationship to relative hand skill was closely similar for both sexes. For reading comprehension, boys close to the point of equal hand skill showed greater impairments than girls. When the analysis grouped children by writing hand, the relationship appeared symmetrical about the same point. The authors tie the pattern to a hypothesis about the persistence of handedness in the population, which this page reports without endorsing.

Nicholls and colleagues tested competing models in 895 participants who completed a general cognitive ability scale and a test of hand preference and hand performance. Their models range from higher ability in left-handers or moderate right-handers to lower achievement in left-handers or mixed-handers. As indexed by the performance measures, moderate right-handers had higher scores than strong left-handers or strong right-handers, and the performance measure also showed lower scores for left-handers than for right-handers. The preference data showed little or no association with cognitive ability. The authors suggest that preference measures cluster people at the extremes, and they describe the effects as subtle, identifiable perhaps only in large studies with sensitive performance measures. They say the data add support to a model that predicts a disadvantage for strong left-handers and strong right-handers alike, while also confirming recent research showing a disadvantage for left-handers.

These results are not in conflict with the 2018 meta-analysis, but they are not the same finding. The meta-analysis compared groups defined by preference, and found no difference between mixed-handers and right-handers. The two cohorts above used skill, or a mixture of skill and preference, and located the shortfall in the middle of the skill distribution and in the group that cannot be classified. A mixed group that is defined by a failure to show a clear preference, as in the Australian study of young children below, may contain children whose development is slower for reasons unrelated to handedness itself. The outcome measures also differ: the 2018 review used full IQ scores, Crow and colleagues used verbal, non-verbal, mathematical and reading measures, and Nicholls and colleagues used one general cognitive ability scale.

For a reader, the practical point is that "left-handed" is the wrong box for most of this signal. If the effect exists at all, the studies place it among people whose two hands are about equally skilled and among children classified as mixed-handed, a group that is a small share of the population. The page on dyslexia and IQ and the page on ADHD and IQ cover conditions that appear in some handedness papers, and they show that a condition and an IQ score are separate measurements.

6 What Do the Large Studies of Young Children Show?

The best known large study of children found that left-handed and mixed-handed children scored lower on school readiness measures at age 4 or 5, but those measures are not IQ tests and the study cannot say whether the gap lasts. Johnston and colleagues, in a paper titled Nature's experiment, used the first wave of the Longitudinal Study of Australian Children, collected in 2004, with 4,942 children aged 4 or 5.

The interviewer classified hand preference after the child completed a test that involved writing and drawing. Children were classified as mixed-handed if the interviewer was sure the child used both hands or was unsure which hand was dominant. By that definition 85.85 percent were right-handed, 9.85 percent left-handed and 4.30 percent mixed-handed, and the shares differed by sex: 10.82 percent of boys and 8.83 percent of girls were left-handed, and 5.62 percent of boys and 2.93 percent of girls were mixed-handed. The paper's development measures included a school readiness test designed to assess general cognitive abilities needed for beginning school, a short form of a vocabulary test, and teacher ratings in six areas such as fine motor and language skills.

The abstract reports three findings. First, the probability that a child is left-handed was not significantly related to child health at birth, family composition, parental employment or household income. Second, left-handed and mixed-handed children performed significantly worse in nearly all measures of development than right-handed children, and the relative disadvantage was larger for boys than for girls. Third, these differences could not be explained by the socioeconomic characteristics of the household, parental attitudes or investments in learning resources. The authors set out several theories of handedness in the paper, including a genetic model, a pathological theory in which some left-handedness comes from early brain damage, and an advantage theory that would predict more left-handers at the top of the distribution, and they test the data against them.

Four limits apply when reading this study next to the meta-analysis. The outcome is development at about age 4 or 5, not Full Scale IQ, and a preschool test samples a narrower range of abilities than the CHC domains of a full battery. The classification came from one interviewer's observation of one session, and the paper notes that some children do not show a strong preference until later in life. The sample is one country and one cohort. And the result is an association: the authors rule out the household explanations they measured, but they cannot rule out unmeasured ones, so the paper does not show that being left-handed causes a developmental gap.

One further reason for caution is selection into the mixed group. Children who show no preference at 4 may be children whose motor or language development is slightly delayed, and a gap for mixed-handed children fits that reading as well as it fits a handedness reading. The adult meta-analysis found no mixed-handed difference, which is what one would expect if a developmental delay evens out by adulthood, though no study reviewed here tests that. The page on whether IQ changes with age explains why a score at 4 or 5 is a weak guide to a score in adulthood.

7 Is Left-Handedness More Common in Intellectual Disability?

Yes, and this is the one place where the meta-analyses agree on a difference, but it is a statement about how often handedness occurs in a diagnostic group and not about the IQ of left-handed people. Papadatou-Pastou and Tomprou published five meta-analyses in 2015 of studies that assessed handedness prevalence in individuals with intellectual disability of unknown or idiopathic nature compared with typically developing individuals, and in intellectually gifted individuals compared with typically developing individuals. They included 19 data sets totaling 16,076 participants, of whom 5,795 had intellectual disability, 8,312 were typically developing and 1,969 were intellectually gifted. The abstract states that elevated levels of atypical handedness were robust only for the comparison of intellectual disability with typical development.

In an author summary published December 19, 2015, the authors put numbers on that result. People with intellectual disability were 1.98 times more likely to be left-handed and 2.66 times more likely to be non-right-handed than the general population. They illustrated the first figure with a 10 percent baseline, which gives a left-handed prevalence of 19.6 percent among people with intellectual disability, and described that as higher than 10 percent but not very much higher. They also explained that they used only studies that included a general population comparison group, so that differences in how studies measured handedness would not be mistaken for real group differences, and that the proposed explanations are speculative, for example that brain damage could have led to both left-handedness and intellectual disability.

Three points keep the result in proportion. First, the direction of the statement matters. If 19.6 percent of people with intellectual disability are left-handed, then about four in five are right-handed. Conversely, most left-handers do not have intellectual disability, because intellectual disability is uncommon in the first place. A statement about the share of left-handers in a diagnostic group cannot be inverted into a statement about the share of people with the diagnosis among left-handers without knowing the base rates. Second, the meta-analysis covered intellectual disability of unknown cause, so conditions with a known genetic or medical cause were not part of it. Third, it does not follow that left-handedness lowers anyone's IQ. The authors' own suggestion that a shared cause, such as early brain injury, could produce both traits is an example of why a prevalence difference is not a causal claim.

The result also sits comfortably beside the 2018 finding of a negligible difference in healthy people. If left-handedness is somewhat more common in a small clinical group at the low end, the average over the whole population moves only a little, and the 2018 summary says explicitly that its conclusion concerns healthy individuals. The page on what a low IQ score means explains how intellectual disability is defined and why an IQ score is only one part of that definition.

8 Are Left-Handers Overrepresented Among Gifted People?

The evidence is thin and mixed: one study of extremely precocious students found left-handedness and mixed-handedness at high frequencies, and the meta-analysis of five giftedness studies found at most a marginal difference in the opposite direction. The popular claim that left-handers dominate the gifted rests on a very small base.

Benbow studied extremely mathematically or verbally precocious students, in the top 1 in 10,000 in reasoning ability. The abstract reports three physiological characteristics found at high frequencies: left-handedness or mixed-handedness, asthma and other allergies, and myopia. It suggests that the first two may reflect a common influence, testosterone, on the nervous and immune systems during fetal development, and that such students may exhibit bihemispheric representation of cognitive functions. The abstract gives no percentages. The selection also matters: on a normal curve one in 10,000 lies about 3.7 standard deviations above the mean, by our arithmetic, but the students were selected by a test of reasoning ability, so no IQ is implied.

The 2015 meta-analyses by Papadatou-Pastou and Tomprou included 1,969 intellectually gifted participants, and the abstract says elevated atypical handedness was not robust for giftedness. In their author summary, the researchers reported slightly fewer left-handers among gifted people than in the general population, only marginally significant, and no difference for non-right-handedness. They called it not a very sure bet. In the January 2018 summary of the IQ meta-analysis, Papadatou-Pastou added that only five published studies had compared handedness prevalence between intellectually gifted and typical individuals, which calls for further research.

How can a study of the top 1 in 10,000 and a pooled analysis of gifted groups disagree? There are at least three candidate reasons, none of them tested here. Giftedness was defined differently, with different cutoffs and different tests, and a talent search that selects the extreme tail is not a random draw from the gifted. Handedness was classified differently, and Benbow lumped left and mixed. And a feature of a very small tail can be real without moving a group average: a doubling in a group that makes up one in 10,000 of the population would be nearly invisible in a population mean, which is also consistent with the 2018 finding of a negligible average difference. That is speculation, and the sources do not test it.

What can be said is narrow. No study reviewed here shows that left-handers are more likely than right-handers to reach a gifted score, and no study shows the reverse with confidence. The page on the gifted IQ range explains where thresholds such as 130 sit, the page on twice exceptional covers giftedness alongside a disability, and the page on gifted adults covers what is known about them. For a feel for tail sizes, the rarity calculator turns any score into a one in N figure.

9 Do Left-Handers Do Worse in School, in Mathematics or in Pay?

Economists have found that left-handers score about a tenth of a standard deviation lower on cognitive skill measures and earn less, but these are observational comparisons, and a related study of mathematical difficulty found no link with hand preference. The studies in this section move from ability to outcomes, and the further the outcome is from a test score, the more other factors can enter.

Goodman, in the Journal of Economic Perspectives in 2014, used multiple datasets from the United States and the United Kingdom. The abstract reports that left-handers score about 0.1 standard deviations lower on cognitive skill measures, which on a scale with a standard deviation of 15 would be about 1.5 points by our arithmetic. It reports that left-handers show more behavioral problems and learning disabilities such as dyslexia, complete less schooling and work in less cognitively demanding jobs, and that their annual earnings are 10 to 12 percent lower. According to the abstract, much of the earnings gap can be explained by observable differences in cognitive skills and behavioral problems. The title frames the difference in terms of brain structure, an interpretation that this page reports and does not test.

The size of the cognitive gap in that paper, 0.1 standard deviations, is of the same order as the 0.07 in the 2018 meta-analysis, and it points the same way. The two are not the same measure. The economic datasets use cognitive skill measures, which are not necessarily full IQ tests, and they include children and adults in samples that were not built to compare handedness. The meta-analysis restricted itself to studies that measured full IQ, in healthy people. That the sign agrees and the size is similar is reassuring about the existence of a small gap, but the meta-analysis found that the gap disappeared without the largest study, so agreement on a point or two is also within what chance and measurement differences could produce.

The larger earnings gap shows why downstream outcomes cannot be read back into ability. A cognitive difference of 0.1 standard deviations looks small beside a 10 to 12 percent earnings difference, and the page on IQ and income reviews how much of income a test score accounts for. The authors attribute much of the gap to cognitive skills and behavioral problems together, which means behavior, schooling and occupation carry part of it. Other explanations, such as the designs of tools, classrooms and workplaces for right-handers, are not tested in the abstract and are not claimed here. The page on IQ and academic achievement reviews how much of schooling a test score accounts for.

On mathematics, Papadatou-Pastou and colleagues reported in 2021 in Laterality three new studies, with 134, 1,893 and 153 participants, and two sets of meta-analyses covering 22 studies and 3,667 participants, of whether atypical hand preference is more common in people with mathematical learning difficulties. No difference in atypical hand preference was found when handedness was assessed with self report questionnaires. There was weak evidence of a difference, with a p value of .049, when writing hand was the criterion in the first study, and when the data were combined meta-analytically no hand preference differences were detected. The authors suggest that handedness effects may need larger samples, that direction of preference may not be a sensitive enough measure, or that atypical hand preference is not associated with mathematical difficulty. The latter, they note, would suggest that handedness is specifically linked to language-related conditions rather than conditions related to cognitive abilities at large. The page on the Mathematical Achievement subtest describes what a quantitative measure samples.

10 Why Might Handedness and Ability Be Linked at All?

Handedness is partly heritable and is used as a behavioral marker of how the brain organizes language, and every proposed link to ability is a hypothesis that the studies above do not settle. The 2015 meta-analysis opens by calling handedness a widely used behavioral indicator for cerebral laterality, mainly of language, so the interest in a link with cognitive ability is not arbitrary.

The genetic evidence is the firmest. Medland and colleagues analyzed handedness, assessed by writing or drawing or by self report, in Australian and Dutch twins and their siblings, 54,270 individuals from 25,732 families. Additive genetic effects accounted for about a quarter of the variance, 23.64 percent with a 95 percent confidence interval of 20.17 to 27.09 percent, and the remainder was accounted for by non-shared environmental influences. There were no differences in prevalence between zygosity groups or between twins and their singleton siblings. In plain terms, the authors attribute roughly three quarters of the variation to environmental influences that family members do not share, so most of it is not a matter of inherited genes. The page on what heritability actually means explains how to read a figure like a quarter.

Four hypotheses recur in the papers reviewed here, and each is a different account of why a link, if any, would exist.

• A balanced genetic polymorphism. Nicholls and colleagues describe a model in which a cognitive disadvantage falls on strong left-handers and strong right-handers, and Crow and colleagues propose that the persistence of handedness variation reflects continuing selection on a gene tied to language.
• A shared cause, such as early brain injury. The authors of the intellectual disability meta-analyses suggest that brain damage could produce both left-handedness and intellectual disability, and the Australian study considers a pathological theory in which a subgroup of left-handers acquired their handedness from brain damage.
• A hormonal influence before birth. Benbow suggests that testosterone during fetal development may affect both handedness and the immune system in very precocious students.
• Different brain organization for language. Goodman frames the gap in terms of brain structure, and Benbow suggests that very precocious students may have bihemispheric representation of cognitive functions.

None of these is shown by the studies in the table. The evidence is correlational in every case, the strongest measure of ability in the group, the full IQ meta-analysis, found almost no difference to explain, and the places where a difference does show up, the middle of the hand skill distribution, mixed-handed young children and a small clinical group, are exactly the places where a shared developmental cause is hardest to rule out. The page on intelligence and the brain reviews what has survived replication about brain structure and IQ, and the page on left brain versus right brain explains why asymmetry in the brain does not translate into two kinds of people.

One point about prevalence also matters for mechanisms. The 2020 meta-analysis argues that cultural factors, such as pressure against left-hand use, moderate how large the left-handed share is. In samples from times or places where children were pushed to write with the right hand, some people classified as right-handed by writing hand may have a left-hand preference for other tasks. That is a reason why the definition problem in the first section is not a technicality, and why the studies that used questionnaires about many tasks and the studies that used writing hand alone can disagree.

11 What This Does Not Say About You, Your Child or a Left-Handed Friend

A group difference of about one point, a prevalence difference in a clinical group and a thin literature on giftedness say nothing about the score of any particular left-handed person. The studies compare averages and shares of groups, and the people inside each group spread across nearly the whole range of scores. With a gap of about 1.05 points and a standard deviation of 15, the two distributions overlap almost completely, as the table in an earlier section shows.

Several mistaken readings follow from group results and are worth naming. A left-handed adult is not likely to score higher or lower for that reason, and the evidence does not support presenting left-handedness as a sign of high ability. A left-handed child is not at an identifiable risk of lower ability on the strength of these studies, which found differences in some measures of young children and not in the meta-analysis of healthy people. A child who shows no clear hand preference at 4 or 5 is in the group in which one study found the largest development gap, and the study cannot say what happens afterward or why. And a parent, teacher or employer who attaches meaning to handedness is applying a group average to an individual, which is not what any of these papers supports.

There is also a measurement point. Any speeded task that requires a hand response could in principle favor one hand, and the studies here did not examine whether handedness affects performance on particular subtests or response formats. That is our observation and not a finding. The way to learn what a person can do is to measure it directly, with a standardized instrument, and to read the result with its confidence interval. The page on online IQ tests explains what a normed online test can and cannot tell a person, and the page on what IQ scores mean shows how to read a result without over-reading it.

For a left-handed reader who sees this topic raised as a curiosity, the more useful questions are the usual ones about any score: which abilities the test samples, what the confidence interval is, and whether the conditions of testing were comparable to those of the norm group. None of the studies reviewed here suggests that being left-handed should change any of those answers.

12 What the Evidence Supports, Stated Narrowly

The evidence supports four narrow statements, and each stops short of the claim that left-handers are smarter or less smart. The first is that in healthy people the average difference in full IQ between left-handers and right-handers is negligible. The 2018 systematic review and meta-analyses of 18 studies, with 20,442 people, found d of minus 0.07, homogeneous and not significant without the largest study, which is about 1 point by our arithmetic. Larger gaps reported elsewhere, such as the 0.1 standard deviations in the economic datasets, are of the same small order.

The second is that left-handedness and non-right-handedness are somewhat more common among people with intellectual disability of unknown cause. The meta-analyses found this robust, with the authors summarizing it as 1.98 times as likely to be left-handed, and they describe the mechanism as speculative. The third is that the signal, where there is one, concentrates in people with little difference in skill between the hands and in young children classified as mixed-handed, and it is not a feature of left-handedness as such. The fourth is that the claim of a left-handed advantage in giftedness has support from one study of extremely precocious students and is not supported by the pooled analysis of five studies, so it is unresolved.

What the evidence does not support is any use of handedness to estimate an individual's ability, any claim that left-handers are more creative or more intelligent as a group, and any causal statement about what handedness does to development. The designs are observational, the definitions of handedness differ across studies, and the effects are small. The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) define validity as the degree to which evidence and theory support the interpretations of test scores for proposed uses. By that standard, using a person's handedness as an indicator of their ability is an interpretation the evidence does not support, and reading a single IQ score as a statement about a handedness group is another. The American Psychological Association's Ethical Principles of Psychologists and Code of Conduct, in the version effective January 1, 2017, addresses the interpretation of scores in Standard 9.06. It requires psychologists to take into account test factors, test-taking abilities and other characteristics of the person assessed that might reduce the accuracy of their interpretations, and to indicate any significant limitations of their interpretations.

Two neighboring questions remain open on this page. Whether left-handers are overrepresented in particular creative or athletic fields is not an IQ question, and the page on IQ and creativity covers the creative side. Whether any IQ difference between groups reflects the groups or the tests is a measurement question covered by the pages on reliability and validity and how IQ scores are normed. The sibling pages on what can lower your IQ, what IQ correlates with, IQ of politicians and conspiracy beliefs and IQ apply the same discipline to other exposures and traits.

13 Sources Behind This Page

Every figure above is traceable to one of the following, and each is linked at the point where it is used. Sample sizes, standardized differences, odds ratios and percentages are quoted as the papers and their authors report them, and any conversion into IQ points, shares of a group or variance explained is labeled as our arithmetic. Publisher pages that block automated readers are cited by DOI after the record was confirmed with Crossref and PubMed on October 13, 2026, and the abstracts were read through PubMed. The two summaries on Atlas of Science are plain language texts by the authors of the papers, opened on October 13, 2026.

• Ntolka E and Papadatou-Pastou M. Right-handers have negligibly higher IQ scores than left-handers: Systematic review and meta-analyses. Neuroscience and Biobehavioral Reviews, 2018, volume 84, pages 376 to 393; and Papadatou-Pastou M, IQ differences between right-handers and left-handers are negligible, Atlas of Science, January 4, 2018.
• Papadatou-Pastou M and Tomprou D. Intelligence and handedness: Meta-analyses of studies on intellectually disabled, typically developing, and gifted individuals. Neuroscience and Biobehavioral Reviews, 2015, volume 56, pages 151 to 165; and Papadatou-Pastou M and Tomprou D, Left-handers, you are not smarter after all, Atlas of Science, December 19, 2015.
• Papadatou-Pastou M, Ntolka E, Schmitz J, Martin M, Munafò M R, Ocklenburg S and Paracchini S. Human handedness: A meta-analysis. Psychological Bulletin, 2020, volume 146, issue 6, pages 481 to 524.
• Correction to Papadatou-Pastou M, Martin M, Munafò M R and Jones G V (2008), Sex differences in left-handedness: A meta-analysis of 144 studies. Psychological Bulletin, 2017, volume 143, issue 10, page 1115.
• Nicholls M E R, Chapman H L, Loetscher T and Grimshaw G M. The relationship between hand preference, hand performance, and general cognitive ability. Journal of the International Neuropsychological Society, 2010, volume 16, issue 4, pages 585 to 592.
• Crow T J, Crow L R, Done D J and Leask S. Relative hand skill predicts academic ability: global deficits at the point of hemispheric indecision. Neuropsychologia, 1998, volume 36, issue 12, pages 1275 to 1282.
• Johnston D W, Nicholls M E R, Shah M and Shields M A. Nature's experiment? Handedness and early childhood development. Demography, 2009, volume 46, issue 2, pages 281 to 301.
• Benbow C P. Physiological correlates of extreme intellectual precocity. Neuropsychologia, 1986, volume 24, issue 5, pages 719 to 725.
• Goodman J. The wages of sinistrality: Handedness, brain structure, and human capital accumulation. Journal of Economic Perspectives, 2014, volume 28, issue 4, pages 193 to 212.
• Papadatou-Pastou M, Panagiotidou D, Abbondanza F, Fischer U, Paracchini S and Karagiannakis G. Hand preference and mathematical learning difficulties: New data from Greece, the United Kingdom, and Germany and two meta-analyses of the literature. Laterality, 2021, volume 26, issue 5, pages 485 to 538.
• Medland S E, Duffy D L, Wright M J, Geffen G M, Hay D A, Levy F, van-Beijsterveldt C E M, Willemsen G, Townsend G C, White V, Hewitt A W, Mackey D A, Bailey J M, Slutske W S, Nyholt D R, Treloar S A, Martin N G and Boomsma D I. Genetic influences on handedness: Data from 25,732 Australian and Dutch twin families. Neuropsychologia, 2009, volume 47, issue 2, pages 330 to 337.
• Oldfield R C. The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 1971, volume 9, issue 1, pages 97 to 113.
• American Educational Research Association, American Psychological Association and National Council on Measurement in Education. Standards for Educational and Psychological Testing, 2014; and American Psychological Association, Ethical Principles of Psychologists and Code of Conduct, Standard 9.06, version effective January 1, 2017.

14 Frequently Asked Questions