What Is Intelligence? A Clear, Science-Based Definition

What is intelligence, in plain terms?

Intelligence is the mental ability to learn from experience, reason through problems, and adapt to new situations. That's the short version. Psychologists have argued over the fine print for over a century, but almost every serious definition circles back to three things: learning quickly, reasoning well, and adjusting when circumstances change.

It's not the same as knowledge. Someone can memorise every capital city in the world and still struggle to solve a novel logic puzzle — that's knowledge without much reasoning flexibility. Intelligence is closer to the engine than the fuel; it's what lets you pick up new skills, spot patterns, and figure out what to do when you've never faced a situation before.

This matters because the word gets thrown around loosely — "street smart," "book smart," "emotionally intelligent" — and beginners often assume scientists have one tidy answer. They don't, entirely. But there is a well-supported core model, plus useful add-ons, and understanding both gives you a much clearer picture than any single soundbite.

The core scientific model: general intelligence (the "g factor")

In the early 1900s, a psychologist named Charles Spearman noticed something curious: people who scored well on one type of mental test (say, vocabulary) also tended to score well on totally different tests (say, spatial puzzles). Scores across unrelated tasks kept correlating with each other. He proposed that a single underlying factor was driving this overlap, which he called the g factor (general intelligence, a statistical measure of the shared ability behind performance on many different cognitive tasks).

The g factor isn't a specific skill you can point to — it's more like the common thread running through reasoning, memory, verbal ability, and processing speed tests. Modern cognitive science has repeated Spearman's work many times with much larger samples, and the overlap he found keeps showing up. It's one of the most replicated findings in psychology.

This is the concept an IQ score (intelligence quotient, a number that ranks someone's reasoning performance against the general population) is trying to estimate. IQ tests use a mix of different task types precisely because no single task captures g cleanly — you need several angles to triangulate it.

Why "general" doesn't mean "everything"

A high g factor score predicts someone will tend to do reasonably well across a broad range of cognitive tasks — but "tend to" is doing real work in that sentence. It doesn't mean someone excels at literally everything, and it says nothing about creativity in the artistic sense, physical coordination, or how someone treats other people.

CHC theory: the modern, more detailed map

Spearman's g factor was the starting point, not the finish line. Over the following decades, researchers such as Raymond Cattell, John Horn, and John Carroll built out a more layered picture, eventually merged into what's now called CHC theory (Cattell-Horn-Carroll theory, a framework that breaks general intelligence down into several broad abilities).

CHC theory keeps g at the top but adds a second layer of broad abilities underneath it, including:

• Fluid reasoning — solving new problems without relying on prior knowledge (spotting the pattern in a sequence you've never seen before)
• Crystallised knowledge — accumulated facts, vocabulary, and learned information
• Working memory — holding and manipulating information in your head for a few seconds at a time
• Processing speed — how quickly you can perform simple mental tasks accurately
• Visual-spatial ability — reasoning about shapes, rotations, and spatial relationships

This is the theoretical backbone behind most modern IQ tests, including the structure of tests like the WAIS (Wechsler Adult Intelligence Scale) or Raven's Progressive Matrices, which are referenced here only as well-known examples, not as tests you'd take on this site. Each subtest is designed to tap into one or more of these broad abilities, and the overall score aggregates them back toward an estimate of g.

Multiple intelligences: a different, more popular idea

In 1983, Howard Gardner proposed something that resonated widely outside academic psychology: the theory of multiple intelligences, arguing that "intelligence" isn't one thing but several relatively independent types — musical, bodily-kinesthetic, interpersonal, logical-mathematical, linguistic, and more.

It's an appealing idea because it validates strengths that traditional testing doesn't capture well, like musical talent or people skills. But it's worth knowing that Gardner's theory has much weaker empirical support than g factor and CHC theory. Studies looking for these "independent" intelligences generally still find they correlate with each other and with general cognitive ability — which is the opposite of what full independence would predict.

That doesn't make the idea worthless. It's a useful reminder that being "smart" comes in more flavours than a single number can express — practical skill, creativity, social insight. It's just not the model the mainstream science of cognitive ability is built on.

Comparing the three models at a glance

Model	Core idea	Scientific support	Best used for
g factor (Spearman)	One general ability underlies performance across cognitive tasks	Very strong, highly replicated	Explaining why cognitive test scores correlate
CHC theory	g sits above several broad abilities (memory, speed, reasoning, etc.)	Strong, basis for modern testing	Structuring actual IQ test batteries
Multiple intelligences (Gardner)	Several independent, unrelated types of intelligence	Weak, disputed	Recognising strengths beyond test-taking

What intelligence, as scientists define it, doesn't measure

This is where a lot of confusion creeps in, so it's worth being blunt about it.

• It's not a measure of your worth or effort. IQ measures a snapshot of cognitive ability, not character, kindness, or how hard someone works.
• It's not fixed knowledge. Fluid reasoning tasks are deliberately designed so prior learning doesn't help much — that's what makes them a measure of reasoning rather than memorisation.
• It's not emotional or social skill. Concepts like "emotional intelligence" borrow the word but describe a different, much less standardised area of research.
• It's not destiny. It correlates with things like academic performance, but correlation isn't a guarantee for any one individual — plenty of exceptions exist in both directions.
• It's not culture-free. Every test is built by people, in a language, with assumptions baked in. Good tests try to minimise cultural bias, but "completely neutral" is an aspiration, not a fact.

How this connects to an IQ score

When you see an IQ score, you're looking at an estimate of where someone's general cognitive ability sits compared with other people of the same age, expressed on a scale centred on 100. If you want the mechanics behind that number — why 100 is "average," what a standard deviation is, and how the bell curve shapes the whole scale — the IQ scale explained breaks it down step by step, and what is the average IQ goes deeper into how that centre point gets calculated.

If you're wondering what counts as strong versus typical, what is a good IQ score walks through how to read a number in context rather than in isolation. And for a full breakdown of every band on the scale, from low scores to the rarely-seen top end, the IQ levels chart lays it out plainly.

Key takeaways

• Intelligence, at its scientific core, means the ability to learn, reason, and adapt — not accumulated knowledge on its own.
• The g factor is the most replicated finding behind the concept: performance across different cognitive tasks tends to correlate.
• CHC theory is the modern, detailed model most real IQ tests are built on, breaking g into abilities like fluid reasoning, working memory, and processing speed.
• Gardner's multiple intelligences theory is popular and intuitive but has much weaker scientific backing than g factor or CHC theory.
• IQ scores estimate general cognitive ability relative to a population — they don't measure character, creativity, or emotional skill.

One honest note before you go: any IQ test you take here, or anywhere, gives you an orientative estimate for self-knowledge and curiosity — not a clinical or medical diagnosis. Treat your score as one useful data point about how you think, not a final verdict on who you are.