Can the Stroop test diagnose ADHD?

The effect-size values come fae Lansbergen an colleagues' review. Its main methodological point is that commonly uised interference formulas are sensitive tae the outcome measure an baseline correction. A group comparison disnae turn intae a diagnostic classifier jist acause it's statistically significant. Lansbergen et al., 2007

Whit did the review o child diagnosis find?

Homack an Riccio's review assesses sensitivity an specificity in bairns an adolescents. It concludes that the Stroop test has limited uiss for tellin ADHD apart fae ither clinical groups. That's directly relevant tae diagnosis: a task can differ fae the average for healthy controls withoot identifyin whit condition explains ane person's result. Homack, 2004

The population matters an aw. A paper-an-pencil test for a bairn, a verbal task for an adult, an an online button-response gemme dinnae share ae set o norms or ae threshold. The clinical Stroop review documents muckle variation in versions an scorin. Scarpina et al., 2017

Why is the task no specific tae ADHD?

Stroop interference is an experimental effect in ordinar readers. The original experiments an later review explain hou clashing word an colour information changes responses withoot needin a diagnosis. Sae experiencin interference isnae evidence o a disorder by itsel. Stroop, 1935 MacLeod, 1991

Why disnae a normal-lookin score rule ADHD oot?

Group averages overlap, an the diagnostic question is aboot a person's wider history an how they get on, no jist the result o ae task. The limited discrimination in the child meta-analysis, an the strong dependence on scorin in the ither review, baith say no tae uisin ae lone score as an answer. A result is neither a confirmation nor a rule-oot by itsel. Homack, 2004 Lansbergen et al., 2007

Reliability brings anither limit. Strong experimental effects can still produce fairly noisy scores for differences atween individuals. The research on the reliability paradox matters ony time a short cognitive task is uised tae mak an important claim aboot ane person. Hedge et al., 2017

Online measurin brings mair uncertainty

Equipment an browser timin affect absolute response times, while task instructions an baselines affect the difference score. A consumer gemme disnae automatically gain the validation o a named clinical test jist acause baith uise clashing colour words. Anwyl-Irvine et al., 2020 Scarpina et al., 2017

International guidance on uisin tests stresses the need for evidence that backs the intended interpretation o a test, in the richt context. A clinically meaningful claim needs validation for that uiss an population, no jist an attractive chart or a familiar task name. International Test Commission, 2001

Whit's a sensible uiss for the test?

Uise it tae explore interference an compare speed an accuracy on the same task in similar conditions. If ye're concerned aboot attention in ilka-day life, a gemme score shouldnae decide gin thae concerns merit a professional assessment. This page explains the limits o measurin; it disnae gie a self-diagnosis rule or advice on treatment. Homack, 2004 International Test Commission, 2001

Nane o the papers cited establishes that practisin this browser gemme treats ADHD, prevents it, or maks fowk generally smarter. The supported interpretation is still performance on ae particular conflict task. MacLeod, 1991 Scarpina et al., 2017