Social Cognition
Brief Implicit Association Test
Also called: BIAT, brief IAT
Measure how strongly plants and buildings are each linked with good in a few minutes: participants sort pictures with one key for two named categories and another for anything else, scored as D.
4-6 minTimed: desktop or laptop with a mouse or trackpad
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 143
- Evidence grade
- C
- Registered clinical studies
- 12
XWhat it measures
Measures how strongly plants and buildings are each linked with good from how fast people sort pictures when one of them shares a key with happy faces. Two categories are named in each short block; participants press I for pictures of either and E for anything else.
Constructs
- Implicit association strength
- Attitudes
- Speeded categorization
Research uses
- Social cognition and attitude research at the group level
- Short implicit measures in longer study batteries
- Method studies of implicit measures
How the paradigm works
The Brief Implicit Association Test measures how strongly two concepts are each linked with good from how fast people sort pictures when one concept shares a response key with good. Each short block names two categories at the top right of the screen: a target, plants or buildings, and good, shown by happy faces. Pictures of either named category are answered with the I key and anything else with the E key, so the other target and bad, shown by sad faces, are never named. A wrong key shows a red X until the right key is pressed, and the response time runs to the right key.
More about the paradigm
The task follows Sriram and Greenwald (2009) with pictures in place of words, so it runs in every language: five Noto Emoji pictures each of plants, buildings, happy faces and sad faces. Good is named in every block, as the source advises for attitudes. Four blocks of 20 trials (32 in the settings) alternate the named target, in an order drawn from the session's seed, and a third pair of blocks can be added. Each block opens with 4 trials of plants and buildings only, which are not scored, then alternates a face on odd trials with a plant or building on even trials, 400 ms after each right key. Two practice parts of 16 trials come first: the faces alone, then drawn curved and angled shapes with the faces, after the source's practice with nonsocial categories. Sorting the faces alone is a step the source does not have.
The D score follows the procedure Nosek and colleagues (2014) recommend: latencies over 10,000 ms are dropped, the 4 opening trials of each block left out, error trials kept with their time to the right key, latencies recoded into 400 to 2,000 ms, and D computed for each pair of blocks, the mean latency with buildings named minus with plants named over the standard deviation of both blocks, then averaged. Positive D means plants are linked with good more strongly than buildings are. The task also reports each pair's D, the mean latency with each target named, the error rate and the share of latencies under 300 ms, which sets a session aside when above 10%. A D score is a research measure for group comparisons and says little about one person. Scores support research and do not diagnose.
Scores
- D Scorescore
- The mean over pairs of blocks of the mean latency with buildings named minus with plants named, over the standard deviation of both blocks' latencies.
- Positive values mean faster sorting when plants share a key with good; not better or worse
- D, First Pairscore
- D from the first two blocks.
- Part of the D score
- D, Second Pairscore
- D from the third and fourth blocks.
- Part of the D score
- Plants and Good RTms
- Mean recoded latency in the blocks naming plants and good.
- Lower is faster
- Buildings and Good RTms
- Mean recoded latency in the blocks naming buildings and good.
- Lower is faster
- Error Rate%
- Share of scored trials whose first key was wrong.
- Above 30% is flagged for review
- Named-Category RTms
- Mean latency of pictures from a named category, up to 10,000 ms.
- Usually faster than other pictures
- Anything-Else RTms
- Mean latency of pictures from the categories not named, up to 10,000 ms.
- Informational
- Fast Responses%
- Share of scored latencies under 300 ms.
- Above 10% is flagged for review
- Anticipationscount
- First keys pressed within 100 ms of a picture.
- Quality check
- Slow Trials Droppedcount
- Scored latencies over 10,000 ms, left out of every latency score.
- Quality check
- Block Ordercount
- 1 when plants were named first, 2 when buildings were.
- Informational
- Scored Trialstrials
- Trials after each block's 4 opening trials.
- Quality check
Use and evidence
143 papers from 2021 to 2025 name this task in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 154 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 43 |
| 2024 | 28 |
| 2023 | 21 |
| 2022 | 24 |
| 2021 | 27 |
| 2020 | 18 |
| 2019 | 26 |
| 2018 | 26 |
| 2017 | 26 |
| 2016 | 40 |
| 2015 | 26 |
| 2014 | 35 |
| 2013 | 19 |
| 2012 | 18 |
| 2011 | 16 |
| 2010 | 20 |
| 2009 | 13 |
| 2008 | 5 |
| 2007 | 2 |
| 2006 | 5 |
| 2005 | 0 |
| 2004 | 1 |
| 2003 | 0 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 0 |
Use grade C
100 to 299 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 7
- Papers on reliability or validity
- 78
- Registered clinical studies using it as an outcome
- 12
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Pictures
- Noto Emoji pictures, 5 each of plants, buildings, happy faces (good) and sad faces (bad). No words, so the task runs in every app language.
- Blocks
- 4 blocks (standard; 6 with a third pair) naming plants and good, then buildings and good, alternating, in an order drawn from the seed. Pictures of a named category take I; anything else takes E.
- Trials
- 20 per block (standard; 20 or 32): 4 opening trials of plants and buildings only, not scored, then a face on odd trials and a plant or building on even trials. As many I as E trials, and no more than 4 in a row on one key.
- Errors
- A red X below the picture until the right key is pressed; the time runs to the right key. The next picture comes 400 ms later.
- Practice
- 2 parts of 16 trials: the faces alone, then drawn curved and angled shapes with the faces.
- Scoring
- Latencies over 10,000 ms dropped, the 4 opening trials left out, the rest recoded into 400 to 2,000 ms; D for each pair of blocks, averaged. More than 10% of latencies under 300 ms is flagged.
Sriram and Greenwald (2009) used words and photographs and had no practice that sorts one kind of item alone. Here the items are pictures and the faces are practiced alone first, so effects may differ in size from theirs.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: every brief_implicit_association_trial of the scored run, I for a named category and E for anything else (chance 50%: each block has as many I as E trials), correctness from the first key pressed. Below chance invalidates and not above chance is flagged. One key on 90% of first responses invalidates when accuracy is not above chance. More than 10% of scored latencies under 300 ms, or more than 30% of first responses wrong, is flagged and withholds the percentile. There is no response window, so nothing is missed.
Why these rules
Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). Nosek, Bar-Anan, Sriram, Axt and Greenwald (2014) set aside sessions with more than 10% of latencies under 300 ms, which are too fast to reflect sorting, and found a 30% error-rate exclusion of some use in small samples.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials per block | 20 (standard) | 20 (standard), 32 |
| Pairs of blocks | 2 (standard) | 2 (standard), 3 |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Timed test: participants need a desktop or laptop with a mouse or trackpad. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Brief Implicit Association Test in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Sriram & Greenwald, 2009; Nosek et al., 2014), with the standard settings (trials per block: 20; pairs of blocks: 2). The main score was D score (the mean over pairs of blocks of the mean latency with buildings named minus with plants named, over the standard deviation of both blocks' latencies).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Sriram, N., & Greenwald, A. G. (2009). The Brief Implicit Association Test. Experimental Psychology, 56(4), 283-294. https://doi.org/10.1027/1618-3169.56.4.283
- Nosek, B. A., Bar-Anan, Y., Sriram, N., Axt, J., & Greenwald, A. G. (2014). Understanding and using the Brief Implicit Association Test: Recommended scoring procedures. PLoS ONE, 9(12), e110938. https://doi.org/10.1371/journal.pone.0110938The scoring procedure.
Source papers
- Sriram N, Greenwald AG (2009). The Brief Implicit Association Test. Experimental Psychology, 56(4):283-294.Paradigm source, Adults
- No normative data for this task.Paradigm source, N/A
Questions
How long does the Brief Implicit Association Test take?
About 4 to 6 minutes, as listed in the test library.
What device does a participant need?
Timed test: participants need a desktop or laptop with a mouse or trackpad. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
D Score, D, First Pair, D, Second Pair, Plants and Good RT, Buildings and Good RT, Error Rate, Named-Category RT, Anything-Else RT, Fast Responses, Anticipations, Slow Trials Dropped, Block Order and Scored Trials. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Brief Implicit Association Test used in research?
143 papers from 2021 to 2025 name it in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 154 of 395 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Brief Implicit Association Test have?
In the published research, 7 papers report its test-retest reliability and 12 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Trials per block (20 (standard), 32) and pairs of blocks (2 (standard), 3). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.