Types of
Intelligence

1 Types of Intelligence: The Short Answer

Updated July 9, 2026 by Structural. The famous "types of intelligence" come from Howard Gardner's theory of multiple intelligences, and while the idea is popular and humane, its central claim is not supported by the evidence. Gardner proposed that we have eight or nine largely independent intelligences, from linguistic and logical to musical and interpersonal. The trouble is that when these abilities are measured, they do not behave as separate things. They correlate positively with one another, which is the signature of a single underlying factor, not of independent intelligences.

It helps to name the two extremes this debate swings between, because the truth is in neither. At one pole is the caricature that intelligence is a single number that sums a person up, which almost no serious researcher actually holds. At the other is Gardner's picture of many independent intelligences, which the data does not support. The accurate position is less catchy than either: one general capacity, expressed through several distinct but related abilities. Most public argument about types of intelligence is really a fight between the two caricatures, while the evidence quietly sits in the middle.

This does not mean intelligence is just one number. The scientifically supported picture sits between the two extremes: there is a strong general factor, called g, that runs through all cognitive abilities, and there are several broad abilities, such as fluid reasoning, verbal knowledge, and spatial ability, that are distinct but still correlated under it. That validated structure, known as the CHC model, is the honest answer to "how many kinds of intelligence are there." It is more than one, but nothing like eight or nine independent ones.

So this page takes the popular theory seriously and fairly, gives it real credit for what it gets right, and then lays out what the science actually says. For the validated structure, see our guides to the g factor and the CHC model, and for a close cousin of this debate, see EQ vs IQ.

2 What People Mean by "Types of Intelligence"

When people talk about types of intelligence, they are almost always referring to the theory of multiple intelligences, put forward by the Harvard psychologist Howard Gardner in his 1983 book Frames of Mind. Gardner argued that the traditional view of intelligence as a single general ability was too narrow, and that the mind is better understood as a set of relatively separate intelligences, each with its own strengths, its own developmental path, and even its own basis in the brain.

It is worth appreciating how Gardner arrived at his list, because his method explains a lot about the theory's strengths and weaknesses. Rather than gathering test data and analyzing it statistically, he reasoned from a set of criteria, such as whether an ability could be isolated by brain damage, whether prodigies or savants showed it on its own, and whether it had a distinct developmental course. That is a thoughtful, scholarly approach, and it produces intuitively appealing categories. But it is a very different enterprise from measuring abilities in thousands of people and seeing how they actually relate, which is where the theory runs into trouble.

The theory landed at a receptive moment and spread rapidly, especially through schools. It offered teachers a hopeful message: a child who struggled with reading or math might be gifted in music, movement, or understanding people, and deserved to have that gift recognized as a genuine intelligence rather than a mere hobby. That message was, and is, emotionally powerful, which is a large part of why the idea became a fixture of education while remaining a minority position among intelligence researchers.

It is important to be clear about what is and is not in dispute. Nobody serious denies that people have different profiles of strengths, that a brilliant novelist may be hopeless at spatial tasks, or that talents in music and athletics are real and valuable. The dispute is narrower and more technical: whether these are separate intelligences in the scientific sense, independent of one another, or whether they are better understood as facets of a more unified cognitive ability plus a range of talents. That is the question the evidence speaks to.

3 Gardner's Intelligences, Listed

Gardner's original 1983 framework described seven intelligences, and he later added an eighth, with a ninth sometimes discussed as a candidate. Laid out plainly, they are:

• Linguistic intelligence: facility with words, language, reading, and writing.
• Logical-mathematical intelligence: reasoning, numbers, logic, and abstract problems.
• Spatial intelligence: picturing and manipulating objects and space.
• Musical intelligence: sensitivity to rhythm, pitch, tone, and musical structure.
• Bodily-kinesthetic intelligence: control of the body and skilled physical movement.
• Interpersonal intelligence: understanding and relating to other people.
• Intrapersonal intelligence: understanding oneself, one's feelings and motivations.
• Naturalistic intelligence: recognizing and categorizing plants, animals, and features of the natural world, added in the 1990s.

An existential intelligence, a sensitivity to deep questions about life and meaning, is sometimes floated as a ninth, which Gardner has treated cautiously. Notice already that some of these, linguistic, logical-mathematical, and spatial, line up closely with abilities that mainstream intelligence tests have always measured, while others, musical and bodily-kinesthetic in particular, are the ones most people would ordinarily call talents. That distinction turns out to matter a great deal.

The uneven nature of the list is itself telling. Three of the items map neatly onto abilities that intelligence tests already measured decades before Gardner wrote, while others sit far outside anything psychologists had called intelligence. That mix is not necessarily fatal on its own, but it does mean the theory is doing two jobs at once: relabeling some well-studied cognitive abilities and, at the same time, promoting talents like music and movement into the same category. Judging the theory fairly requires keeping those two jobs separate, because they succeed and fail for different reasons.

4 Why the Theory Is So Appealing

Before criticizing the theory, it is worth understanding honestly why it captured education so completely, because the appeal is real and partly deserved. Multiple intelligences pushes back against a genuine problem: the tendency to reduce a whole person to a single IQ number and to treat that number as a verdict on their worth or potential. That reduction is dehumanizing, and MI offers a generous alternative in which everyone has areas of genuine strength.

There is also an institutional reason the theory spread so fast, beyond its emotional appeal. It arrived with a ready-made application for schools, a rationale for varied, engaging, strengths-based teaching that many good educators already believed in. Multiple intelligences gave that instinct a respectable-sounding scientific banner to march under, and once a framework is embedded in teacher training and curriculum materials, it becomes self-sustaining regardless of how the underlying science fares. Popularity in education and support among researchers are two different currencies, and this theory has always been rich in the first and poor in the second.

The theory is also democratic in spirit. It says that the child who cannot sit still but moves with grace, or who struggles academically but reads a room instantly, is not simply less intelligent, but intelligent in a different way. For teachers facing a room of very different children, that is a humane and motivating framework, and it encourages attention to strengths that a narrow academic focus can crush. These are good instincts, and any honest critique has to acknowledge that MI is popular for reasons that are partly admirable.

The problem is not the humane instinct. It is that a comforting message and a correct scientific theory are two different things, and the evidence has to be judged on its own terms. You can fully embrace the value of diverse talents and still ask whether Gardner's specific claim, that these are separate and independent intelligences, holds up when tested. It largely does not, and pretending otherwise would be its own kind of condescension.

5 The Central Problem: The Intelligences Correlate

Here is the finding that undercuts the theory's core claim. Gardner's central proposal is that the intelligences are independent, so that being strong in one tells you little about the others. But when researchers actually measure abilities across many people, they find the opposite: performance on different mental tasks is positively correlated. People who do well on verbal tasks tend, on average, to do better on spatial and numerical ones too. This pervasive positive correlation is so consistent it has a name, the positive manifold, and it is one of the most replicated findings in all of psychology.

That positive correlation is precisely the evidence for a general factor of intelligence. When many abilities rise and fall together, statisticians extract a common factor, g, that accounts for the shared variance, typically a large share of it. In other words, the very data Gardner's theory would need to refute, ability measures being independent, comes out the other way: the abilities cohere. A theory built on the independence of intelligences runs directly into the single most robust structural fact about them, which is that they are not independent at all.

It is worth being precise about what the correlations do and do not prove. They do not prove that everyone good at one thing is good at everything, or that specific abilities do not exist; clearly they do. What they show is that the abilities are not independent, which is the specific thing Gardner's theory asserts. A modest but reliable positive correlation between, say, verbal and spatial ability is entirely compatible with people having distinct strengths, and it is flatly incompatible with those strengths being separate intelligences that vary with no relation to one another. The theory made a strong claim, and the strong claim is the one the data refutes.

Gardner's response has generally been to argue that his intelligences cannot or should not be measured with standard tests, which sidesteps the correlation problem but at a heavy cost. A theory that cannot be tested against data is not really competing on the same field as one that can, and it forfeits the ability to be confirmed as well as the risk of being refuted. That move keeps MI safe from disproof, but it also keeps it out of science.

6 Talents, Abilities, and the Word "Intelligence"

The second major criticism is about definitions, and it is more than a quibble. Several of Gardner's intelligences, especially musical and bodily-kinesthetic, are things most people and most psychologists would call talents or abilities rather than intelligences. Renaming athletic or musical skill an "intelligence" does not, by itself, make it one; it mostly changes the label. Critics argue this broadens the word intelligence so far that it loses its meaning and becomes a synonym for "anything a person can be good at."

The definitional worry has a practical edge, not just a philosophical one. If intelligence expands to cover every human capacity, the word stops doing useful work, because a term that describes everything distinguishes nothing. Psychologists guard the word carefully precisely so that statements about intelligence, that it predicts this or develops like that, stay meaningful. Calling athletic grace an intelligence feels generous, but it quietly empties the concept, and a concept that means anything a person is good at cannot support the kind of specific, testable claims that make a science.

Revealingly, Gardner himself has acknowledged that his choice of the word intelligence was partly strategic. He has said that if he had called them talents, no one would have paid attention, whereas calling them intelligences was provocative and forced a debate. That is a candid admission, but it also concedes the critics' point: the reframing was rhetorical as much as empirical. Calling dance an intelligence wins an argument about respect; it does not establish a new scientific category.

This matters because science depends on constructs that are carefully defined and measured. Intelligence, as researchers use it, refers to reasoning, learning, and problem-solving ability, and it has been operationalized into reliable, validated tests. Gardner's intelligences, by contrast, were derived from reasoning over case studies and neuropsychology rather than from data, and crucially they have never been turned into a validated, reliable measure. There is no psychometrically sound "multiple intelligences test," which is why the popular online quizzes with that name are just self-report questionnaires.

7 The Learning-Styles Myth It Gets Confused With

One of the most important things to know about multiple intelligences is what it is not, because it is constantly confused with a genuinely debunked idea: learning styles. The learning-styles claim holds that each person has a preferred mode of learning, often labeled visual, auditory, or kinesthetic, and that teaching to that preferred style improves learning. It is enormously popular among educators and almost entirely unsupported by evidence.

The persistence of learning styles despite the evidence is a cautionary tale worth keeping in mind here. Surveys find that the great majority of teachers believe in matching instruction to learning styles, and a whole market of materials supports it, yet the careful experiments simply do not back the central claim. It is a vivid example of how an idea can be near-universal in practice and still be wrong, which is exactly the pattern to watch for with multiple intelligences: enormous popularity is not evidence, and the two questions, is it beloved and is it true, have to be kept apart.

When researchers have tested the crucial prediction, that matching instruction to a student's preferred style produces better learning than mismatching it, the effect essentially does not appear. People do have preferences, but teaching to those preferences does not reliably help them learn, and what matters far more is matching the method to the material, a diagram for spatial content, practice for a skill, regardless of anyone's supposed style. The learning-styles industry rests on a claim that repeated experiments have failed to confirm.

Gardner has been clear that his theory is not a theory of learning styles, and he has expressed frustration at the conflation. Multiple intelligences is a claim about the structure of the mind; learning styles is a claim about how to teach, and the two are logically separate. But in practice they travel together in schools, and both draw on the same appealing intuition that people are fundamentally different kinds of learners. Keeping them distinct is important: you can reject the learning-styles myth outright while still treating MI as a serious, if flawed, theory of mind.

8 What the Evidence Actually Supports: The g Factor

If the intelligences are not independent, what is the real structure? The most established finding is the general factor of intelligence, g. It emerges directly from the positive manifold: because performance across diverse cognitive tasks correlates, a general factor can be extracted that captures what they share, and it typically accounts for a substantial portion of the differences between people across all the tasks. Whatever g is, it is not an artifact; it shows up in essentially every well-constructed battery of mental tests ever built.

It is worth heading off a common misreading of g at this point. Saying there is a general factor does not mean intelligence is a single thing sitting in one place in the brain, nor that g is fixed, nor that it captures everything valuable about a person. g is a statistical regularity, the shared variance across abilities, and it is compatible with those abilities having their own distinct character. Researchers insist on it not out of ideology but because of the stubbornness of the data: extract a general factor from almost any diverse test battery and it reappears, which is not something a serious theory of intelligence can simply ignore.

What makes g scientifically serious is not just that it exists but that it predicts. General cognitive ability is among the best single predictors psychology has of academic achievement, job performance, and a range of life outcomes, as our pages on those topics document. A theory of intelligence has to reckon with g, because it is the most reliable and most predictive construct in the field. Multiple intelligences, by declining to be measured, never engages with it, which is a large part of why intelligence researchers do not regard MI as a serious competitor.

None of this means g is the whole story or that a single number captures everything about a mind. It means that any accurate account of intelligence has to include a strong general factor, because the data force it. The interesting scientific question is not whether g exists, but how it relates to the more specific abilities beneath it, which is exactly what the next section is about.

9 The Real "Multiple Abilities": The CHC Model

Here is the part that often surprises people: mainstream science does recognize multiple cognitive abilities. It just organizes them very differently from Gardner. The most widely accepted framework, the Cattell-Horn-Carroll model, describes intelligence as a hierarchy. At the top sits g. Beneath it are several broad abilities, and beneath those are dozens of narrow, specific skills.

The history behind this model is worth a sentence, because it shows how it earned its authority. It grew out of decades of work by researchers who gathered enormous amounts of test data and analyzed the structure of abilities statistically, gradually converging on the same hierarchical picture from independent lines of research. That convergence, many investigators arriving at g over broad abilities over narrow skills from different data, is the kind of evidence that makes a model durable. It was not decreed from an armchair; it was assembled from data and repeatedly reproduced, which is precisely the test multiple intelligences never submitted to.

The broad abilities in the CHC model include fluid reasoning (solving novel problems), crystallized intelligence (acquired verbal knowledge), visual-spatial processing, short-term and working memory, processing speed, quantitative ability, and a few others such as auditory processing and long-term retrieval. These are genuinely distinct, which is why a good test reports a profile across them rather than a single number, as our page on cognitive domains explains. In that sense, the intuition behind multiple intelligences, that there is more than one kind of ability, is correct.

The crucial difference is that the CHC broad abilities are correlated and organized under g, not independent. Fluid reasoning and verbal ability are distinct, but people strong in one tend to be at least somewhat strong in the other, which is why they sit under a common general factor. This is the honest resolution of the whole debate: intelligence is neither a single undifferentiated number nor a set of unrelated intelligences, but a general capacity expressed through several correlated, measurable abilities. That structure is supported by a century of data, and it is what modern tests are built on.

10 Other Models: Sternberg and Beyond

Gardner is not the only person to propose a structure of intelligences, and it is worth briefly noting the main alternative for a fuller picture. The psychologist Robert Sternberg proposed a triarchic theory, dividing intelligence into three parts: analytical (the reasoning that standard tests measure), creative (generating novel ideas), and practical (the "street smarts" of getting things done in the real world). Sternberg's theory is more empirically oriented than Gardner's and has generated actual research and measures.

Sternberg's work also illustrates a healthier way for a challenger theory to behave. Rather than declaring his components immune to measurement, he built instruments and ran studies, which let the field evaluate the claims and accept some while setting others aside. That is how science is supposed to absorb a new idea: the useful parts survive testing and get incorporated, the rest is dropped. Multiple intelligences, by staying unmeasured, never allowed this process, so it has neither been confirmed nor productively refined in the decades since, remaining frozen as a popular idea rather than a developing science.

Even so, the triarchic theory is contested. Its most distinctive component, practical intelligence, based on tacit knowledge, has produced mixed results, and critics argue that once you account for general ability, the added predictive value of a separate practical intelligence is modest. It is a more testable theory than multiple intelligences, and it has contributed useful ideas about creativity and real-world problem-solving, but it has not displaced the g-plus-broad-abilities picture as the mainstream view.

The pattern across these alternatives is instructive. Each captures a real intuition, that reasoning is not everything, that creativity and practical skill matter, that people have diverse strengths, and each struggles at the same point: showing that its proposed components are truly separate from general ability rather than expressions of it. The intuitions are valuable; the claim of independence is where the science keeps pushing back.

11 Is Emotional Intelligence a "Type"?

Emotional intelligence is often folded into lists of intelligence types, so it deserves a brief, honest placement. Like Gardner's interpersonal and intrapersonal intelligences, emotional intelligence points at a real and valuable set of skills, perceiving, understanding, and managing emotions. And like multiple intelligences, it is genuinely useful but oversold, and its most popular measures largely overlap with personality rather than capturing a distinct ability.

The recurring pattern across emotional intelligence, multiple intelligences, and the others is worth naming directly, because it is the real lesson of this whole area. Each begins with a true and humane observation, that reasoning is not the only thing that matters and that people have varied gifts, and each then overreaches by packaging that observation as a set of new, independent intelligences on a par with cognitive ability. The observation deserves to be kept; the overreach deserves to be trimmed. Learning to separate the valuable insight from the inflated claim is most of what it takes to think clearly here.

The honest placement is that emotional skill is real and matters, but it is not a "type of intelligence" in the same validated sense as the cognitive abilities g is built from. It relates only weakly to reasoning ability, and much of what popular emotional-intelligence tests measure is personality traits under a new name. We treat this in depth on our dedicated page comparing EQ and IQ, which reaches the same balanced verdict this page does about multiple intelligences: a real and humane insight, wrapped in claims the evidence does not fully support.

Grouping emotional intelligence with Gardner's interpersonal intelligence is therefore reasonable, but it inherits the same caution. Both name something worth valuing; both stretch the word intelligence beyond what the data cleanly supports; and both are best understood as important human capacities that sit alongside cognitive ability rather than as additional, independent intelligences of the same kind.

12 What the Multiple-Intelligences Idea Gets Right

It would be as dishonest to dismiss multiple intelligences entirely as to accept it uncritically, so it is worth stating clearly what the idea gets right. First, and most importantly, people genuinely do differ in their profiles of ability. A person can be far stronger in verbal reasoning than in spatial ability, or vice versa, and a single overall number hides that texture. On this point, MI and mainstream science actually agree, they just describe the abilities differently.

This is why the disagreement is narrower than the culture war around it suggests. Defenders of multiple intelligences and mainstream researchers are not really arguing about whether people have diverse strengths, which both accept, but about how those strengths are organized and whether they warrant the word intelligence. Framed that way, a sensible reader does not have to pick a tribe. You can wholeheartedly value musical, physical, and interpersonal gifts, insist that no one be reduced to a single score, and still follow the evidence to a general factor with correlated broad abilities. Honoring people and honoring data are not in conflict.

Second, the theory's insistence that a person is more than one number is not just kind, it is correct in an important sense. Talents in music, movement, art, and human relationships are real, valuable, and life-shaping, even if calling them separate intelligences overstates the case. A good education and a fair society should recognize and develop those strengths, and MI has done real good by pushing schools in that direction, whatever its scientific flaws.

The accurate synthesis, then, is to keep the humane insight and drop the specific false claim. Yes, people have diverse strengths and are not captured by a single label; no, those strengths are not eight or nine independent intelligences. The scientifically grounded way to honor the insight is the profile of correlated broad abilities that the CHC model describes and that good tests actually measure, which delivers the "you are more than one number" message without abandoning the evidence.

13 Common Myths About Types of Intelligence

• Myth: science recognizes eight or nine separate intelligences. It does not. That is Gardner's contested theory, not the consensus. The evidence supports a general factor plus several correlated broad abilities.
• Myth: the intelligences are independent. Measured abilities correlate positively, the positive manifold, which is the core evidence for g and against independence.
• Myth: multiple intelligences and learning styles are the same thing. They are separate ideas, and Gardner rejects the conflation. Learning styles specifically is a debunked myth.
• Myth: there is a validated multiple-intelligences test. There is not. Popular "MI tests" are self-report questionnaires, not validated ability measures.
• Myth: believing in g means reducing people to one number. The mainstream model includes a whole profile of broad abilities under g; it is not a single-number view.
• Myth: emotional intelligence is a proven separate type of intelligence. It names real skills but relates weakly to cognitive ability, and its popular measures overlap heavily with personality.

14 What This Means for You, and Where ACIS Fits

The useful takeaway is a balanced one. Hold on to the humane core of the types-of-intelligence idea, that people have diverse strengths and are not summed up by one label, and let go of the specific claim that there are many independent intelligences, which the evidence does not support. If you want to understand your own mind, the goal is not to find which of nine intelligences you have, but to see your actual profile of correlated cognitive abilities, alongside the talents and interests that a cognitive test does not measure at all.

A word on where our own test fits, in the same honest spirit as the rest of this page. ACIS is built on the validated structure, not on multiple-intelligences theory. It measures your reasoning across the CHC broad abilities, fluid reasoning, verbal ability, visual-spatial processing, working memory, and speed, and returns both a full-scale score, reflecting the general factor, and a domain profile showing your specific strengths and weaknesses. That is the scientifically grounded version of "you have more than one kind of ability": a real profile of measurable, correlated abilities rather than a self-reported guess at which of Gardner's intelligences fits you. It will not tell you your musical or athletic talent, and it does not pretend to. What it will give you is an honest, validated map of how you reason.

Sources and Further Reading

• Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligences. The founding text of the theory.
• On the positive manifold and the general factor, see our guide to the g factor and standard treatments of factor analysis in intelligence research.
• On the validated hierarchical structure of abilities, see the Cattell-Horn-Carroll model and our page on cognitive domains.
• On learning styles, see the research reviews finding no reliable benefit from matching instruction to a learner's preferred style.
• Related on-site reading: EQ vs IQ, IQ and Creativity, and Fluid vs Crystallized Intelligence.

15 FAQ: Types of Intelligence