Fluid Intelligence Test

Where fluid reasoning sits

In the Cattell–Horn–Carroll (CHC) model of cognitive abilities, fluid reasoning (usually written Gf) is a broad ability near the top of the hierarchy. It feeds narrower skills such as inductive reasoning, sequential reasoning, and quantitative reasoning, and it correlates strongly with the general factor that runs through almost every mental test. Because Gf is about handling novelty rather than recalling content, tasks that load on it present unfamiliar material with a discoverable structure — figural matrices, letter and number series, and analogies — so that prior knowledge gives little advantage.

Open, research-oriented item sets built on this idea — the International Cognitive Ability Resource (ICAR) is a well-known public example — lean heavily on matrix reasoning and number-and-letter series for exactly this reason. The formats below follow the same logic without reusing any published item.

Why fluid reasoning matters

Fluid reasoning is the part of thinking you fall back on whenever the map runs out: a tool with no manual, a task you have never done, a puzzle whose rules you must infer. It predicts how quickly people pick up genuinely new skills, which is why it shows up as a strong correlate of learning and problem-solving across very different settings.

It is also the ability most people mean, loosely, when they call someone “quick.” The value of measuring it deliberately is that raw speed and confidence are poor guides — a careful reasoner who checks every rule usually beats a fast one who commits to the first pattern that looks right.

How to read a matrix

A figural matrix usually hides more than one rule. A reliable strategy is to check what changes across each row and, separately, down each column. When a feature is constant along one direction and cycles along the other, you have found a rule; combine every rule you find, then read off the missing cell. If several options still survive, look for the one feature you have not yet accounted for.

How MOSAIC-48 measures fluid reasoning

In the full assessment, fluid-reasoning items are not shown as one isolated block. Six of them are interleaved with four other reasoning domains across two mixed rounds, arranged by a round-robin so you never meet two items from the same domain back to back. The items are untimed, so the score reflects how well you find the structure rather than how fast you click.

Those six items sample all three of the subskills the demo illustrates — rule discovery (finding the pattern in a series), rule integration (combining two or more simultaneous rules in a matrix), and anomaly detection (spotting the figure that breaks a rule). Fluid reasoning is one of eight domains that combine into your Composite Cognitive Index.

Reading your fluid-reasoning score

Your fluid-reasoning result is best read against your own other domains, not as a rank against other people. It tells you how comfortably you handle unfamiliar structure compared with, say, your verbal or spatial performance — a relative map, not a label.

Treat any single session as a snapshot. Fluid reasoning is sensitive to sleep, focus, and the particular items you happen to see, so a tired session understates you and a lucky run flatters you. It is an internal-reference estimate for self-insight, not a standardized or official score, and it is not meant for diagnosis or selection.