What Is Raven's Progressive Matrices? History, Format & Uses

Raven's Progressive Matrices is a non-verbal reasoning test in which you look at a visual pattern with one piece missing and choose the option that correctly completes it. First published by the psychologist John C. Raven in 1938, it is one of the most widely used and most researched measures of abstract reasoning in the world — and it is the format our own free online IQ test is based on.

Because the puzzles contain no words and no arithmetic, the test aims to measure reasoning ability as directly as possible, independent of language, vocabulary or formal education. This article explains where the test came from, how it works, what it does and doesn't measure, and where it is used today.

A short history

John C. Raven introduced the Progressive Matrices in 1938, drawing on theories of intelligence developed with his mentor Charles Spearman — in particular Spearman's idea of a general intelligence factor (often written as g). Raven wanted a clean way to measure what Spearman called "eductive" ability: the capacity to make sense of complexity and to work out relationships that aren't explicitly stated.

The test proved durable. Over more than eight decades it has been repeatedly revised and re-standardised for new populations, and it remains a standard tool in psychology. It is published today in three main forms for different ability levels and ages:

• Standard Progressive Matrices (SPM) — the original, for the general population.
• Coloured Progressive Matrices (CPM) — designed for young children and older adults.
• Advanced Progressive Matrices (APM) — harder items that spread out scores among high-ability adults.

How the matrices work

Each item shows a matrix — usually a 3×3 grid — of abstract figures that change according to one or more hidden rules. One cell, normally the bottom-right, is left blank. Your task is to identify the rule governing the pattern and select, from a set of options, the figure that belongs in the empty cell.

The "progressive" in the name is important: the items are ordered so that they start easy and become steadily harder. Early problems involve a single, obvious rule; later ones combine several rules at once — for example, a shape might rotate, gain elements, and change shading all across the same grid. Solving the hardest items requires holding multiple relationships in mind at the same time, which is exactly the kind of processing the test is meant to capture.

What the test actually measures

Raven's matrices are widely regarded as one of the best single-format measures of fluid reasoning — the ability to solve novel problems, independent of acquired knowledge. In a influential cognitive analysis, Carpenter, Just and Shell (1990) examined what makes the hardest items hard, and concluded that performance depends heavily on the ability to generate and manage abstract rules in working memory. That is a core component of what intelligence researchers mean by general ability.

It is just as important to be clear about what the test does not do. A matrix-reasoning score is not a complete picture of a person's mind: it says little about verbal ability, acquired knowledge, creativity, practical skills or emotional understanding. Intelligence is multi-faceted, and any single test captures only part of it — a point made carefully in the American Psychological Association's well-known review, "Intelligence: Knowns and Unknowns" (Neisser et al., 1996). If you want to understand how a single number is interpreted, see what a good IQ score actually means.

Why non-verbal, "culture-fair" reasoning matters

Many traditional intelligence tests rely on vocabulary, general knowledge or arithmetic — all of which are shaped by language and schooling. That makes it hard to compare people from different educational or cultural backgrounds fairly. Because Raven's items use only abstract shapes, they reduce that dependence, which is why the test is often described as relatively culture-fair.

"Culture-fair" does not mean "culture-free." Research, including Raven's own long-term data (Raven, 2000), shows that average scores can shift over time and differ across groups for environmental reasons — a phenomenon connected to the Flynn effect. Non-verbal matrices minimise cultural bias; they don't eliminate it entirely.

Where Raven's Progressive Matrices is used

• Clinical and educational psychology — as one component of a broader cognitive assessment, administered and interpreted by qualified professionals.
• Research — as a reliable, language-light measure of reasoning in studies across many countries.
• Occupational and military selection — historically used where quick, non-verbal reasoning is relevant.
• High-IQ societies — matrix-style, culture-fair tests are among the formats used in high-IQ admissions. If that interests you, read how to join Mensa.

How our test relates to Raven's Progressive Matrices

The test on this site uses the same core idea that Raven pioneered: a sequence of progressively harder visual matrices, each missing a piece, with six answer options. It contains 60 items, it is timed, and your raw score is converted into an IQ estimate using age-normed tables — the same general principle professional tests rely on, because reasoning performance varies with age.

To keep expectations honest: this is an online self-assessment for educational and entertainment purposes. It is not the official, copyrighted Raven's Progressive Matrices, it is not a clinical or diagnostic instrument, and it is not affiliated with Mensa. If you are curious about how reliable online testing can be in general, we cover that honestly in how accurate online IQ tests are.

With that context, the best way to understand the format is simply to try it.