There is no agreed number of types of intelligence, and the pages that give you a confident count are giving you one person's list. The famous list is Howard Gardner's theory of multiple intelligences — eight, sometimes counted as nine. Psychometric research answers the question in a different shape entirely: not separate intelligences, but broad ability areas that correlate with one another. This page sets out both, and where the evidence actually stands.
Quick answer
The best-known answer is eight — Gardner's linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal and naturalist intelligences — with existential intelligence often counted as a ninth. The count varies because the theory admits an intelligence on judgement against a set of criteria, not by measurement. Mainstream psychometrics describes the same ground as a hierarchy of broad abilities that overlap rather than as separate intelligences, which is what an IQ test actually samples.
The eight (or nine) types of intelligence on Gardner's list
Howard Gardner set out the theory in Frames of Mind (1983) with seven intelligences, added naturalist intelligence in the 1990s, and has treated existential intelligence as a candidate rather than a settled member. The list below is the one most sources mean by “the types of intelligence”.
| Intelligence | What it covers | Where you meet it |
|---|---|---|
| Linguistic | Sensitivity to words, meaning and the uses of language | Writing, argument, learning languages |
| Logical-mathematical | Reasoning, abstraction, number and formal problem-solving | Mathematics, programming, formal logic |
| Spatial | Mental representation and manipulation of space and form | Navigation, design, engineering drawing |
| Musical | Pitch, rhythm, timbre and musical structure | Performance, composition, close listening |
| Bodily-kinesthetic | Skilled control of the body and of objects | Sport, dance, surgery, crafts |
| Interpersonal | Reading and responding to other people | Teaching, negotiation, care work |
| Intrapersonal | Accurate self-knowledge and self-regulation | Planning, judgement about one's own limits |
| Naturalist | Recognising and classifying patterns in the natural world | Biology, agriculture, field identification |
| Existential (proposed) | Posing and pursuing questions about meaning and existence | Philosophy, theology — treated as a candidate, not a member |
Why the number keeps changing
Seven, eight, nine, and on some sites twelve or thirteen. The count moves for a reason worth understanding, because it explains why you will never find a page that settles it.
Gardner admits an ability to the list when it satisfies a set of criteria he set out himself — among them that the ability can be selectively damaged by brain injury, that it has a plausible evolutionary history, that it shows up in exceptional individuals, and that it has its own symbol system. Those are reasoned criteria, but applying them is an act of judgement. Nothing measures an intelligence into existence, so nothing prevents someone else applying the criteria differently — or, in the case of the ten-, twelve- and thirteen-item lists that circulate on blogs, ignoring them entirely and adding “financial” or “spiritual” intelligence. If a page offers you twelve types of intelligence, that is not newer research; it is a longer list.
What the research actually says about multiple intelligences
The theory is enormously influential in schools and genuinely contested in research, and the two facts are usually reported separately. Here they are together.
The case against. Lynn Waterhouse reviewed the evidence in Educational Psychologist in 2006 and concluded that multiple intelligences theory, the Mozart effect and emotional intelligence theory alike “lack adequate empirical support and should not be the basis for educational practice”, arguing that each has better-supported counterparts in cognitive psychology and cognitive neuroscience.
The reply, in the same issue. Gardner and Moran answered that the review “misunderstands and oversimplifies MI theory” and that Waterhouse's own argument undermines her conclusion. They did not present a new dataset; their case is that the theory rests on a synthesis of evidence across disciplines rather than on the kind of single validation study a critic asks for.
That is the real state of play, and it is why this page will not tell you the theory is proven or debunked. The best-known attempt to test it with conventional psychometric methods is Visser, Ashton and Vernon's Beyond g: Putting multiple intelligences theory to the test (Intelligence, 2006); all three papers are linked in the sources below, and they are short enough to read.
The other answer: general ability and the CHC model
Psychometrics never asked how many intelligences there are. It asked how mental abilities relate to each other, and got an awkward answer very early: they correlate positively. Charles Spearman reported it in 1904 — people who do well on one kind of mental task tend to do better than average on others — and named the shared component g.
That does not mean there is only one ability. The model most test publishers work from today, Cattell-Horn-Carroll theory, describes a hierarchy: a general factor at the top, broad abilities beneath it — fluid reasoning, comprehension-knowledge, working memory, processing speed, visual processing and others — and narrow abilities beneath those. Kevin McGrew's 2009 account in Intelligence traces how it was assembled from decades of factor-analytic work.
Two of those broad abilities are the pair most people have actually met under other names: fluid reasoning and comprehension-knowledge, which we cover in fluid vs crystallized intelligence. The practical difference between the two frameworks is that CHC's abilities are defined by how test scores actually cluster, while Gardner's intelligences are defined by criteria applied to a body of evidence. One is a description of measurements; the other is an argument about what deserves the name.
Types of intelligence are not learning styles
This is the most common mistake made with the theory, and Gardner has objected to it for years. Multiple intelligences is a claim about kinds of ability. Learning styles is a claim about instruction: that a person learns better when teaching matches their preferred format — visual, auditory, kinesthetic. The second claim is the one with a clear evidential problem.
Pashler, McDaniel, Rohrer and Bjork reviewed it for Psychological Science in the Public Interest in 2008. The specific claim they examined is what they call the meshing hypothesis — “according to which instruction is best provided in a format that matches the preferences of the learner” — and they found the evidence base for it inadequate, despite the concept's “great influence within the education field” and a thriving industry selling assessments of it.
So “I am a visual learner” and “I have strong spatial intelligence” are not the same statement, and the first is on weaker ground than the second.
Where emotional intelligence fits
Emotional intelligence is a separate framework again, and it maps loosely onto the interpersonal and intrapersonal entries above. It has the same difficulty as the rest of this field — several competing definitions and instruments, some of which measure personality more than ability — and it overlaps with IQ less than most people expect. We set out what is measurable there in IQ and EQ compared.
What an IQ test does and does not cover
An IQ test samples reasoning, verbal and numerical problem-solving, working memory and processing speed, and reports a single score with subscores — see how IQ tests work for the mechanics and the IQ scale for what the number means. It does not measure musical skill, physical control, self-knowledge or social warmth, and it was never built to.
That is a limit worth stating plainly rather than working around. A score describes performance on one family of reasoning tasks under time pressure. It is not a summary of a person, and a low or middling score is not evidence that the other abilities on this page are missing. The honest use of a test result is as one piece of self-knowledge among several — which is also the only thing our own test claims to be, as our methodology sets out.
Trust and scope notes
This page is educational. It does not provide diagnosis, educational placement advice or individualized professional guidance, and no online test — including ours — establishes which abilities a person has.
IQ Revealed is not affiliated with, endorsed by or connected to Howard Gardner, Harvard University, the American Psychological Association or any publisher of the tests and theories referenced on this page. They are named so that the frameworks described above can be traced to the people and papers that produced them. The quoted phrases are taken from the published abstracts of the papers linked in the sources.