Attention
Latent Inhibition Task
Also called: Stimulus preexposure task, unreinforced preexposure task, latent inhibition paradigm
Measure how prior exposure affects learning: participants watch shapes, then learn which familiar and new shapes predict a light.
3-5 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 3
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how prior exposure affects learning. Participants watch shapes, then learn which familiar and new shapes predict a light. Scores compare early learning for the two kinds of cue.
Constructs
- Latent inhibition
- Selective attention
- Associative learning
- Stimulus preexposure
Research uses
- Research on attention and associative learning
- Group comparisons of learning after unreinforced exposure
How the paradigm works
Latent inhibition is slower learning about a cue after it has been seen without consequence. This task compares learning about familiar and new shapes in the same session. Participants first watch two shapes with no outcome, prediction or response. Each appears 20 times for 1 second, separated by a 250 ms blank. The exposure count can be set from 10 to 40 per shape.
More about the paradigm
After four button-practice trials with a separate star, participants predict whether each of four shapes will be followed by a light. One familiar and one new shape always predict the light; one familiar and one new shape never do. Each appears 12 times in a seeded sequence, with one of each cue in every block of four. Predictions have no time limit. The outcome and answer feedback appear for 1 second, followed by a 250 ms blank. Which shape takes each role comes from the session seed.
The primary score is early prediction accuracy for the new light cue minus that for the familiar light cue, over the first six presentations of each. A positive difference means slower early acquisition for the familiar cue. The accompanying scores show later learning, no-light predictions, and how many presentations each light cue needed for three consecutive correct answers. Unlike Cue Blocking, no cue is paired with an already predictive cue. Unlike Probabilistic Reversal Learning, acquisition rules never switch.
Scores
- Early Exposure Effect%
- Novel-positive minus preexposed-positive accuracy over the first 6 presentations of each cue. Both complete early cells are required.
- Positive means slower early acquisition for the familiar cue; compare groups
- Familiar Cue Early Accuracy%
- Correct predictions on the first 6 presentations of the familiar light cue.
- Informational
- New Cue Early Accuracy%
- Correct predictions on the first 6 presentations of the new light cue.
- Informational
- Familiar Cue Accuracy%
- Correct predictions across the 12 presentations of the familiar light cue.
- Informational
- New Cue Accuracy%
- Correct predictions across the 12 presentations of the new light cue.
- Informational
- No-Light Cue Accuracy%
- Correct predictions for the two shapes that never predict a light.
- Learning check
- Final Prediction Accuracy%
- Correct predictions on the last 3 presentations of each of the 4 cues.
- At most half correct is flagged for review
- Familiar Cue Trials to Criterioncount
- Presentation of the familiar light cue ending the first run of 3 consecutive correct predictions; null if not reached.
- Informational
- New Cue Trials to Criterioncount
- Presentation of the new light cue ending the first run of 3 consecutive correct predictions; null if not reached.
- Informational
- Practice Accuracy%
- Share of practice answers that follow the explicit instruction.
- Quality check
- Anticipationscount
- Presses within 100 ms of a choice screen appearing; ignored by the task. Imported answers under 100 ms are excluded from learning scores.
- Quality check
- Predictions Scoredcount
- Usable acquisition answers in the selected run.
- Fewer than planned is flagged for review
- Planned Predictionscount
- Acquisition trials planned: 48.
- Informational
- Opening Presentationscount
- Unique preexposure presentations recorded in the selected run.
- Fewer than planned is flagged for review
- Planned Opening Presentationscount
- Two shapes, each shown 20 times by default, or 10 to 40 times in the settings.
- Informational
Use and evidence
3 papers from 2021 to 2025 name this task in their title or abstract, down 77% compared with 2016 to 2020. That ranks it 482 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 1 |
| 2024 | 0 |
| 2023 | 0 |
| 2022 | 0 |
| 2021 | 2 |
| 2020 | 3 |
| 2019 | 0 |
| 2018 | 2 |
| 2017 | 4 |
| 2016 | 4 |
| 2015 | 2 |
| 2014 | 1 |
| 2013 | 4 |
| 2012 | 3 |
| 2011 | 2 |
| 2010 | 1 |
| 2009 | 3 |
| 2008 | 1 |
| 2007 | 4 |
| 2006 | 4 |
| 2005 | 2 |
| 2004 | 5 |
| 2003 | 5 |
| 2002 | 2 |
| 2001 | 1 |
| 2000 | 4 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 4
- Registered clinical studies using it as an outcome
- 0
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 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Shapes
- 4 original monochrome shapes, assigned to the four cue roles from the session seed. A separate star appears only in practice. No licensed pictures or published items.
- Opening sequence
- 2 shapes, each shown 20 times by default (10 to 40 as a setting), for 1,000 ms each with a 250 ms blank. Participants watch without responding. No light or prediction instruction appears in this part.
- Practice
- 4 button-practice trials after the opening sequence. Each tells the participant which answer to choose, then shows the outcome and feedback. Practice uses a separate star.
- Learning
- 48 trials: 12 presentations each of a familiar light cue, a new light cue, a familiar no-light cue and a new no-light cue. Each block of 4 contains all four cues in seeded order. The light follows both positive cues on every trial and neither negative cue.
- Each prediction
- Click, tap, or press 1 for No light or 2 for Light, with no time limit. Feedback lasts 1,000 ms, followed by a 250 ms blank. Presses under 100 ms are ignored and counted.
- Scores
- Early exposure effect: new minus familiar light-cue accuracy over the first 6 presentations of each cue. Whole-session cue accuracy, final accuracy over the last 3 presentations of each cue, trials to 3 consecutive correct predictions, practice accuracy and recorded counts are also reported.
An original research version using shapes and explicit predictions. Published studies used other stimuli and response measures. Compare groups; the exposure effect has no diagnostic cutoff or percentile.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: latent_inhibition_learning from the run with the most unique acquisition trials. Predictions are learned from feedback, so early accuracy is not tested against chance. At most half correct in the last three presentations per cue, missed practice instructions, missing opening presentations and incomplete acquisition are flagged for review. Choices under 200 ms on at least half the trials invalidate. A hidden tab withholds the percentile.
Why these rules
The measure compares acquisition, so early errors are part of the effect. The final predictions check whether any stable cue-outcome learning occurred. Passive preexposure needs the shapes to be watched; the browser cannot verify gaze. Quality thresholds are ConductScience choices, not clinical cutoffs.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Presentations of each familiar shape | 20 | 10 to 40 |
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
- Untimed test: runs on desktops, laptops, tablets and phones. 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 Latent Inhibition Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Byrom, Msetfi & Murphy, 2018; Aranzubia-Olasolo, Nelson & Sanjuán Artegain, 2024), with the standard settings (presentations of each familiar shape: 20). The main score was early exposure effect (novel-positive minus preexposed-positive accuracy over the first 6 presentations of each cue).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Byrom, N. C., Msetfi, R. M., & Murphy, R. A. (2018). Human latent inhibition: Problems with the stimulus exposure effect. Psychonomic Bulletin & Review, 25(6), 2102-2118. https://doi.org/10.3758/s13423-018-1455-4Unreinforced exposure, response requirements and novelty controls in human latent inhibition.
- Aranzubia-Olasolo, M., Nelson, J. B., & Sanjuán Artegain, M. del C. (2024). Latent inhibition in humans from simple stimulus exposure. Journal of Experimental Psychology: Animal Learning and Cognition, 50(2), 118-130. https://doi.org/10.1037/xan0000374Evidence for an exposure effect without masking. This task uses original shapes and prediction choices, not their video procedure.
Source papers
- No normative data for this task.Paradigm source, N/A
Questions
How long does the Latent Inhibition Task take?
About 3 to 5 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Early Exposure Effect, Familiar Cue Early Accuracy, New Cue Early Accuracy, Familiar Cue Accuracy, New Cue Accuracy, No-Light Cue Accuracy, Final Prediction Accuracy, Familiar Cue Trials to Criterion, New Cue Trials to Criterion, Practice Accuracy, Anticipations, Predictions Scored, Planned Predictions, Opening Presentations and Planned Opening Presentations. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Latent Inhibition Task used in research?
3 papers from 2021 to 2025 name it in their title or abstract, down 77% compared with 2016 to 2020. That ranks it 482 of 495 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.22.0, counted 3 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Latent Inhibition Task have?
In the published research, 0 papers report its test-retest reliability and 0 registered clinical studies use it as an outcome (evidence grade D in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Presentations of each familiar shape (10 to 40). 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.