John Ridley Stroop

Introduction

John Ridley Stroop (March 21, 1897 – September 1, 1973) was an American experimental psychologist whose dissertation produced one of the most cited results in the history of the field. In 1935 he reported that naming the ink colour of a word takes measurably longer, and produces more errors, when the word itself names a different colour — say the word RED printed in blue. The Stroop effect that carries his name is now the standard laboratory demonstration that a highly practised process can run off on its own and interfere with what a person is trying to do.

The oddity of Stroop's position is that this is almost the whole of his scientific legacy. He published four papers in experimental psychology, left the field within a few years of his doctorate, and spent the rest of a long career teaching biblical studies at a small Christian college in Nashville. As his biographer Colin MacLeod put it, "the Bible, not psychology, was his life's work."

Early Life and Education

Stroop was born on a farm at Hall's Hill, in Rutherford County, Tennessee, about seven miles from Murfreesboro, the second youngest of six children. He was a frail child who was not expected to live long, and he was shielded from the heavy work of the family farm — a circumstance he later credited for giving him the time to read. He finished first in his class at Kitrell County School, graduated from David Lipscomb High School in Nashville in 1919, and took a diploma from David Lipscomb Junior College in 1921.

He married Zelma Dunn on December 23, 1921, while still an undergraduate, and had three sons over the next seven years. To support them he taught at David Lipscomb College and worked as a janitor and librarian at the same time as doing his own coursework — a combination that delayed his degrees but that colleagues remembered as characteristic.

At George Peabody College in Nashville he took a B.Sc. in 1924 and an M.A. in 1925, then stayed on for doctoral work in experimental psychology with a minor in educational psychology, conducted in the Jesup Psychological Laboratory under Joseph Peterson. His early work already concerned the time people take to recognise and name colour words. He received the PhD in 1933 and published the dissertation two years later.

Academic and Professional Career

Stroop did not pursue an academic career in the way the discovery might have made possible. He intended to continue the line of research after his doctorate, but Peterson died in 1935 and the thread was dropped. He worked briefly for the Tennessee Educational Commission and at Tennessee Polytechnic Institute, then returned to David Lipscomb College, where he would spend the rest of his working life.

There he served as registrar for eleven years, then as chair of the psychology department from 1948 to 1964. He retired in 1967 and returned as Emeritus Professor of Biblical Studies, teaching and writing until his death. He also spent a year as dean of Ohio Valley College in Parkersburg, West Virginia — which is where he died, on September 1, 1973, at the age of 76.

Alongside his teaching he preached most Sundays and taught Bible classes for decades, keeping meticulous records of where he had preached and on what text. The work produced seven books on scripture, of which the trilogy God's Plan and Me was the principal one and was used in Christian-school teaching for years afterwards.

Major Contributions to Psychology

Studies of Interference in Serial Verbal Reactions (1935)

The paper appeared in the Journal of Experimental Psychology (18, 643–662) and reported three experiments. Stroop had subjects read colour words printed in black ink, read the same words printed in conflicting colours, name the colours of coloured squares, and finally name the ink colour of words that spelt a different colour. The last condition is the one that matters: it was substantially slower than naming the colours of plain squares, and it produced errors in which the subject read the word instead of naming the colour.

Two features of the design deserve attention. First, Stroop used each subject as their own control across conditions, which was not standard practice at the time and which made the interference effect hard to attribute to anything but the conflict itself. Second, he ran extended practice — subjects came back for many sessions over several days — and showed that the interference shrank but did not disappear, which he took as evidence that the reading response had become automatic in a way colour naming had not, and that practice reorganises the relative strength of the two habits rather than switching one off.

He also built a control for a boring alternative explanation. If the slowdown were merely the result of an unfamiliar or visually odd stimulus, then naming the colour of a nonsense string or of a set of coloured shapes should be slow too. It was not. The interference required a readable word that named a different colour, which is why the result is about meaning and habit rather than about visual complexity.

The third experiment pushed further: Stroop had subjects name the word rather than the colour while the ink colour conflicted, and found interference in that direction too, though far smaller. That asymmetry — colour naming suffers much more from word reading than word reading suffers from colour — is still the central constraint that any theory of the effect has to explain. The mirror-image slowdown is now called reverse Stroop, and the size difference between the two directions is treated as a finding in its own right rather than a curiosity.

An odd footnote: the group judgment paper

Stroop's first publication, in 1932, has nothing to do with attention. "Is the judgment of the group better than that of the average member of the group?" asked whether averaging the estimates of several people beats taking a typical person's estimate — one of the earliest experimental treatments of what is now called the wisdom of crowds. It is a reminder that the four papers are not a programme but the residue of a few active years, and that the colour-word work was never meant to be a career.

Automaticity and the fate of the finding

Stroop's own explanation was in terms of habit strength: adults have read words hundreds of thousands of times and named colours almost never, so the word reading response is set off involuntarily and competes with the required one. That framing — an automatic process versus a controlled one — became the standard vocabulary for the phenomenon, and it is why the effect is now taught as the basic demonstration of automaticity.

The timing, however, was wrong. In the mid-1930s American psychology was behaviourist, and a paper about verbal habits and interference attracted little notice. Stroop's result was essentially dormant until the arrival of information-processing psychology in the 1960s, when a generation of researchers looking for ways to study attention inside the laboratory found it waiting. Since then it has been replicated in well over seven hundred publications and adapted into emotional, spatial, numerical and cross-modal variants.

From the 1990s it was taken up again, this time as a probe of the brain rather than of the mind. The Stroop became one of the standard tasks in functional imaging of cognitive control, and the finding that conflict between competing responses recruits midline frontal cortex — together with theories that cast that region as a conflict monitor rather than as the seat of attention — rests substantially on Stroop-style designs. The effect has thus had three separate careers: a behaviourist study of verbal habits, an information-processing measure of attention, and a neuroscientific assay of control.

Major Publications

  • Stroop, J. R. (1932). Is the judgment of the group better than that of the average member of the group? Journal of Experimental Psychology, 15(4), 550–562.
  • Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662.
  • Stroop, J. R. (1935). The basis of Ligon's theory. American Journal of Psychology, 47(3), 499–504.
  • Stroop, J. R. (1938). Factors affecting speed in serial verbal reactions. Psychological Monographs, 50(1), 38–48.

These four papers, together with his religious writings, are the whole of his published output. The standard biographical account is Colin M. MacLeod (1991), "John Ridley Stroop: Creator of a landmark cognitive task," Canadian Psychology, 32(3), 521.

Impact and Legacy

Few psychologists have their name attached to so robust and so widely used an effect. The Stroop task appears in undergraduate textbooks, in clinical neuropsychological batteries, in functional imaging studies of cognitive control, and in the growing literature on individual differences in executive function. It is the default instrument for asking whether someone can suppress a prepotent response.

Four lines of criticism qualify that legacy, and none of them concerns the reliability of the basic finding, which is not in doubt.

Stroop was not first. James McKeen Cattell reported in Mind in 1886 that naming objects and colours takes longer than reading their names — the same asymmetry, fifty years earlier. Stroop knew this literature and addressed a contemporary version of it directly in his 1935 note on Ligon's theory. The eponym is a textbook instance of Stigler's law: the effect is named for the person who measured it carefully, not the person who noticed it.

The mechanism is contested. Stroop's automaticity account was superseded, or at least supplemented, by the parallel distributed processing model of Cohen, Dunbar and McClelland (1990), which explains the effect and its asymmetry in terms of the relative strength of two processing pathways rather than in terms of a race between responses. Later work has shown that different Stroop variants are not all explained by the same mechanism: the emotional Stroop and the classic colour-word Stroop dissociate in ways that suggest they are not one phenomenon.

The clinical use is weaker than the reputation. The test is widely used in the assessment of attention and executive function, but performance on it is impaired in an extraordinarily broad set of conditions — schizophrenia, depression, anxiety, ADHD, traumatic brain injury, dementia, and the ordinary effects of ageing. A deficit on the Stroop tells you that something is wrong with attentional control; it does not tell you what. Colin MacLeod's own integrative review (1991) is candid about this, and reviewers of the clinical literature have repeatedly concluded that the test lacks the specificity to be used diagnostically on its own.

And it is a difference score, which is where the real problem lies. The quantity of interest — the interference effect — is the difference between an incongruent and a neutral condition, and difference scores inherit the unreliability of both components. This has turned out to be more damaging than it first appears. The Stroop is the paradigm case of what is now called the reliability paradox: the task produces large, effortlessly replicable group effects and simultaneously shows test–retest reliability too low to support the individual-differences and clinical uses it is now put to. A task can be a superb instrument for comparing conditions and a poor instrument for measuring a person, and the same data can support both verdicts.

The size of that problem has since been measured in the setting where the clinical use is most plausible. Hachenberger and colleagues (2025, Behavior Research Methods, 57, article 29) embedded the Stroop and a go/no-go task in ecological momentary assessment, with participants completing both several times a day, and decomposed the variance using dynamic structural equation modelling. The Stroop difference scores — error rate and reaction time alike — showed poor within-subject reliability, while the go/no-go scores were moderate to good. Occasion specificity, the share of variance belonging to the moment rather than the person, was between 1.4 and 11.1 per cent for the Stroop and between 16.1 and 37.2 per cent for the go/no-go. Within-subject correlations between Stroop performance and self-reported self-control were very small and mostly non-significant. The finding is awkward in a specific way: the task that is most often carried into the field, on phones, as a momentary assay of self-control, is the one that behaves least like a measurement of anything stable when it gets there.

Two further complications follow from the same place. The size of the measured interference depends on what is used as the neutral baseline — colour patches, rows of letters, or neutral words each give a different answer — so "the" Stroop effect is in part a property of the scoring convention. And the interference is not a fixed quantity of the person: it shrinks as the proportion of incongruent trials in the block rises, both at the level of the whole list and at the level of the individual item, because participants adjust how much attention they allocate to word reading. The effect is real, robust, and routinely treated as a trait; it is also substantially a response to the context in which it is measured.

Learn More

References

  • Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643-662.

Last updated: 2026-10-20

Menu
Loading