Reaction time studies measure what questionnaires can't reach — not what you say you think, but how your mind is wired underneath. Read how they work, then try them yourself — all twelve have playable demos.
So…here's the logic. The mind is a network of associations — ideas that have been experienced together, over and over, end up linked. And linked ideas activate each other automatically: think doctor and nurse is already stirring, whether you asked it to or not. So when a task flashes a word and demands a response within a fraction of a second, your speed is never just speed. Pairings your mind treats as natural produce fast, accurate responses; pairings it treats as strange produce hesitation and errors. The lag is small — tens of milliseconds — but it is measurable, repeatable, and surprisingly hard to fake. The speed with which one concept calls up another becomes a proxy for the strength of the connection between them.
That's the assumption every task on this page shares. Where a questionnaire asks you to report on your mind — with all the self-presentation and blind spots that invites — a reaction time task watches your mind operate in real time, before deliberation has a chance to tidy things up. Researchers call the family “implicit measures” for exactly that reason.
Well... almost every implicit task is built on the same arithmetic: a difference score. The task makes you respond under one pairing of concepts, then under the opposite pairing, and subtracts. If you were faster (or more accurate) when SELF shared a response with WARM than when SELF shared a response with HARSH, the difference is your score — the rest is statistics. And the statistics are where it gets cool…I teach statistics at the university, so trust me on this! Four flavors appear in my studies:
From signal detection theory. In go/no-go tasks (the GNAT), how cleanly you separate “press” items from “don't press” items under a deadline. Higher d′ when two categories share the go-key means that pairing was easier for your mind to hold.
The IAT family's metric: the reaction-time difference between the two pairings, divided by your own variability. Dividing by your SD makes scores comparable across slow and fast responders.
The AMP doesn't time you at all — it counts judgments. If ambiguous symbols are rated pleasant less often right after one kind of prime than another, the affect leaking off the prime is doing the work.
In approach–avoidance tasks, how much faster you move toward a stimulus than away from it. Negative values mean the thing pulls you in; positive values mean it pushes you back.
So if all these tasks measure associations, why does my lab run twelve of them? Because “the strength of an association” turns out to be several different things wearing one coat. Some tasks measure pure association — how tightly two concepts are wired together (the GNATs and IATs). Some measure evaluation — the gut feeling a concept triggers (the AMP). Some measure motivation — whether your body wants to move toward or away (the AAT). Others capture attention (the emotional Stroop: does a threatening word grab your eyes?), conflict (mouse-tracking: does your hand waver between two answers?), beliefs (the RRT: can you affirm a proposition quickly?), and identity (the identification GNAT: how tightly is a group wired to your sense of me?). Same millisecond logic, aimed at different layers of the mind.
Where did all this machinery come from? Well…it's a good story — one that runs from the founders of the field straight down to this lab. It all began with… (a history lesson!)
One honest caveat before you play. Implicit measures are research instruments, designed to detect differences across groups of people; any single person's score on any single day bounces around. The demos below use shortened versions — fewer trials, more noise. To the extent that your score surprises you, treat it as a conversation starter, not a diagnosis.
The lab's implicit program measures the same constructs — self-compassion, vegan stigma, masculinity, the moral status of animals — layer by layer: association, evaluation, motivation, attention, conflict, propositions, and identity. Every study below has a playable demo. Each one runs in your browser, takes a few minutes, and calculates your score the same way the research version does — your result appears at the end, with a plain-language read of what it means. Use a computer with a keyboard (these need real keys, not a touchscreen). Anonymous responses are recorded so we can keep testing the tasks, and each demo asks three optional demographic questions before showing your score.
Measures implicit meat–masculine and vegan–feminine associations. You press the spacebar for category pairs under a 650-ms deadline; signal detection separates real sensitivity from button-mashing. The question: does the link between meat and manhood live at a level self-report can't reach?
Paradigm: Nosek & Banaji (2001)
Try the demo →Pairs SELF words (I, me, my, myself...) with WARM versus HARSH attributes. The index is d′(SELF+WARM) − d′(SELF+HARSH) — to our knowledge the first implicit measure of self-compassion in the literature. Does the inner critic show up in your reaction times?
Paradigm: Nosek & Banaji (2001)
Try the demo →Crosses VEGAN with good/bad and strong/weak attributes across four blocks. Because the single-category design never forces vegan against omnivore, it can separate vegans' internalized negativity from omnivores' outgroup prejudice — two different things the usual designs blur together.
Paradigm: Nosek & Banaji (2001)
Try the demo →A vegan or neutral prime flashes for 83 ms before a Chinese pictograph you judge as pleasant or unpleasant. You're told to ignore the prime; reliably, people can't. The score is the gap in “pleasant” judgments after vegan versus neutral primes — affect leaking from one stimulus onto the next.
Paradigm: Payne, Cheng, Govorun, & Stewart (2005)
Try the demo →Arrow keys move a stick figure toward or away from words; approach bias is the speed advantage for moving toward. One arm uses food words, the other self-kind versus self-critical words — approach and avoidance of one's own inner voice as an implicit measure of fear of self-compassion.
Paradigm: Krieglmeyer & Deutsch (2010)
Try the demo →Categorize farm, companion, and wild animals as FOOD or NOT FOOD while the cursor's full trajectory records. The measure isn't your answer — it's how much your hand curved toward the other answer on the way. A real-time motor signature of moral conflict.
Paradigm: Freeman & Ambady (2010)
Try the demo →The two-key classic: sort SELF, WARM, and HARSH words left and right while the pairings switch between rounds. Scored with the Greenwald D-algorithm. Run alongside the GNAT, it asks whether two different paradigms agree about the same mind — the heart of a validation argument.
Paradigm: Karpinski & Steinman (2006); scoring: Greenwald, Nosek, & Banaji (2003)
Try the demo →Name the ink color of stigma words (preachy, militant, smug...) versus neutral words. If identity-threatening content grabs attention, color-naming slows — an attentional measure that requires no evaluation at all. Stigma theory predicts the slowdown is largest for those who carry the stigmatized identity.
Paradigm: Stroop (1935); emotional variant: Williams, Mathews, & MacLeod (1996)
Try the demo →Pairs FARM animals and COMPANION animals with moral-worth attributes (worthy vs. commodity). The index of interest is the moral discount: how much less moralized farm animals are than dogs and cats in implicit memory — the cognitive substrate of the meat paradox.
Construct: Caviola, Everett, & Kahane (2019)
Try the demo →Speeded TRUE/FALSE responding under two induced rules: answer as if “vegans are weak” were true, then as if “vegans are strong” were true. Unlike association tasks, this measures speeded endorsement of the proposition itself — the difference between ideas being linked and ideas being believed.
Paradigm: relational responding (De Houwer and colleagues)
Try the demo →Pairs VEGAN with ME versus NOT-ME pronoun poles. Where every other task taps evaluation of vegan content, this one taps identification — how tightly the group is wired to the self. Theory predicts internalized stigma bites hardest in those who implicitly identify most.
Paradigm: Nosek & Banaji (2001)
Try the demo →A category prime flashes for a quarter second, then a letter string: word or non-word? If VEGAN primes speed up stereotype words specifically, the stereotype is active in memory — semantic priming as a measure of which ideas a concept automatically wakes up.
Paradigm: semantic priming lexical decision
Try the demo →Caviola, L., Everett, J. A. C., & Kahane, G. (2019). The moral standing of animals: Towards a psychology of speciesism. Journal of Personality and Social Psychology, 116(6), 1011–1029. https://doi.org/10.1037/pspp0000182
Freeman, J. B., & Ambady, N. (2010). MouseTracker: Software for studying real-time mental processing using a computer mouse-tracking method. Behavior Research Methods, 42(1), 226–241. https://doi.org/10.3758/BRM.42.1.226
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. K. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74(6), 1464–1480. https://doi.org/10.1037/0022-3514.74.6.1464
Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the Implicit Association Test: I. An improved scoring algorithm. Journal of Personality and Social Psychology, 85(2), 197–216. https://doi.org/10.1037/0022-3514.85.2.197
Karpinski, A., & Steinman, R. B. (2006). The Single Category Implicit Association Test as a measure of implicit social cognition. Journal of Personality and Social Psychology, 91(1), 16–32. https://doi.org/10.1037/0022-3514.91.1.16
Krieglmeyer, R., & Deutsch, R. (2010). Comparing measures of approach–avoidance behaviour: The manikin task vs. two versions of the joystick task. Cognition and Emotion, 24(5), 810–828. https://doi.org/10.1080/02699930903047298
Nosek, B. A., & Banaji, M. R. (2001). The go/no-go association task. Social Cognition, 19(6), 625–666. https://doi.org/10.1521/soco.19.6.625.20886
Payne, B. K., Cheng, C. M., Govorun, O., & Stewart, B. D. (2005). An inkblot for attitudes: Affect misattribution as implicit measurement. Journal of Personality and Social Psychology, 89(3), 277–293. https://doi.org/10.1037/0022-3514.89.3.277
Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662. https://doi.org/10.1037/h0054651
Williams, J. M. G., Mathews, A., & MacLeod, C. (1996). The emotional Stroop task and psychopathology. Psychological Bulletin, 120(1), 3–24. https://doi.org/10.1037/0033-2909.120.1.3