The Stroop effect: 90 years of one elegant task

In 1935, a graduate student named John Ridley Stroop published a paper in Journal of Experimental Psychology on a task he had constructed almost as an aside. Show people a list of color words printed in mismatched colored ink — the word "red" printed in blue ink, "yellow" in green — and ask them to name the ink color, not the word. The task is simple. It is also remarkably difficult, and the difficulty is reliable.

Ninety years later, the Stroop effect remains the most-used laboratory measure of cognitive control, with thousands of published variations. The reason it works tells us something specific about how attention is built.

1. The basic effect

Naming the ink color of a congruent word (RED in red ink) is fast. Naming the ink color of an incongruent word (RED in blue ink) is slower, with more errors. The slowdown — typically 150-400 milliseconds — is the Stroop interference.

The asymmetry is informative: reading the word is faster and more automatic than naming the color. Asking participants to name the word (regardless of ink color) produces almost no interference from the color. The slower process can't interfere with the faster one. The faster one routinely interferes with the slower (MacLeod, 1991).

2. What it measures

The Stroop task is the canonical measure of three things, depending on which paper you read:

Selective attention — the ability to focus on one feature while ignoring another. Response inhibition — the ability to suppress the prepotent (automatic) response. Cognitive control — the broader top-down management of attention.

These overlap. The Stroop task is robust because it forces all three to operate simultaneously. The interference indexes how well that orchestrated control works in a given individual or condition.

3. The neural locus

Functional imaging consistently localizes Stroop performance to the anterior cingulate cortex, dorsolateral prefrontal cortex, and posterior parietal regions. ACC activation is particularly tied to conflict detection — registering that the word and color disagree. DLPFC handles resolving that conflict by amplifying the relevant signal (Botvinick et al., 2001).

Lesions to these regions produce predictable Stroop deficits. Conditions that compromise prefrontal function — ADHD, schizophrenia, OCD, normal aging — all show increased Stroop interference. The task is sensitive enough to detect subtle frontal-lobe changes that other measures miss.

4. The bilingualism connection

A repeated finding in the bilingualism literature: bilinguals show smaller Stroop interference than matched monolinguals on average. The proposed mechanism is the one we keep returning to — bilinguals' constant practice in suppressing the non-target language transfers to other interference-suppression tasks (Bialystok et al., 2008).

This advantage is not as robust as some earlier claims suggested. Multiple replication failures have appeared since 2015. The current state of evidence: a small bilingual advantage on Stroop in some populations, absent in others, and the conditions that distinguish them are not yet fully clear.

5. The lesson of a small task

The Stroop effect has outlasted most of twentieth-century psychology not because it's surprising — it isn't — but because it operationalizes something real that's hard to measure otherwise. Cognitive control is invisible from the outside; the Stroop task makes it visible by setting up a conflict where the controlled response and the automatic response disagree.

The trick is the measurement. The thing being measured is one of the most consequential features of how human minds work.