Reasoning
Property Induction Task
Also called: category-based induction task, inductive generalization task, premise diversity task, blank property task
Measure how people generalize from examples: participants rate how likely other animals, or all birds or all mammals, share a new property, with diverse, typical and similar examples set in matched pairs.
5-8 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
- 32
- Evidence grade
- D
- Registered clinical studies
- 0
What it measures
Measures how people generalize from examples. Participants learn that one or two kinds of animal have a made-up kind of cell, then rate how likely another animal, or all birds or all mammals, is to have it, on 34 arguments: 16 matched pairs and 2 checks. The main score is the diversity effect: how much more two different animals than two similar ones persuade that a whole category shares a property. The arguments are written in each of the eight languages.
Constructs
- Inductive reasoning
- Category-based induction
- Generalization
- Conceptual knowledge
Research uses
- Reasoning and concept research
- Development of reasoning
- Cross-cultural comparison
- Within-study group comparison
How the paradigm works
The property induction task measures how people generalize a new property from examples to other members of a category (Rips 1975; Osherson, Smith, Wilkie, López and Shafir 1990). Each argument says that one or two kinds of animal have a made-up kind of cell, shown with their pictures, and asks how likely another animal, or all birds or all mammals, is to have it. Participants rate each argument on a bar from 0 (very unlikely) to 10 (very likely).
More about the paradigm
The 34 arguments are 16 matched pairs, 8 about birds and 8 about mammals, and 2 checks. The two arguments of a pair differ in one thing. Typicality pairs have a typical or an atypical premise animal (blackbirds or penguins) and conclude about the whole category. Diversity pairs add a similar or a dissimilar second animal to the same first one (eagles and owls, or eagles and swans). Similarity pairs ask about a named animal from a premise animal like it or unlike it (ducks or parrots, then geese). The checks go from a whole category to one of its members. Every argument has its own kind of cell, named by a code, so nothing carries over. The animal names and sentences were written for this task in each of the eight languages, and the pictures come from Noto Emoji. The session's random seed deals the codes and orders the arguments, never the two of a pair in a row; both are recorded. One practice argument comes first. The test has no feedback and no time limit.
The main score is the diversity effect: the mean, over the diversity pairs, of the rating with the dissimilar second animal minus the rating with the similar one. The typicality and similarity effects are computed the same way. Adults usually rate diverse premises as stronger, and in children the diversity effect appears later than the effects of similarity (López, Gelman, Gutheil and Smith 1992) and typicality (Rhodes, Brickman and Gelman 2008). Scores support research and do not diagnose.
Scores
- Diversity Effectpoints
- Mean, over the 6 diversity pairs, of the rating with a dissimilar second animal minus the rating with a similar one, on the 0 to 10 scale.
- Positive means two different animals persuaded more than two similar ones that the whole category shares the property
- Typicality Effectpoints
- Mean, over the 6 typicality pairs, of the rating from a typical animal minus the rating from an atypical one.
- Positive means typical animals persuaded more
- Similarity Effectpoints
- Mean, over the 4 similarity pairs, of the rating from an animal like the conclusion animal minus the rating from one unlike it.
- Positive means similar animals persuaded more
- Mean Rating, Diverse Premisespoints
- Mean rating of the diversity arguments with a dissimilar second animal.
- Higher means stronger
- Mean Rating, Similar Premisespoints
- Mean rating of the diversity arguments with a similar second animal.
- Higher means stronger
- Mean Rating, Typical Premisepoints
- Mean rating of the typicality arguments from a typical animal.
- Higher means stronger
- Mean Rating, Atypical Premisepoints
- Mean rating of the typicality arguments from an atypical animal.
- Higher means stronger
- Mean Rating, Similar Premisepoints
- Mean rating of the similarity arguments from an animal like the conclusion animal.
- Higher means stronger
- Mean Rating, Dissimilar Premisepoints
- Mean rating of the similarity arguments from an animal unlike the conclusion animal.
- Higher means stronger
- Mean Ratingpoints
- Mean rating of the 32 paired arguments.
- Describes how readily properties were generalized overall
- Check Ratingpoints
- Mean rating of the 2 checks, which go from a whole category to one of its members.
- Expected high; under 5 is flagged for review
- Median Response Timems
- Median time from an argument's appearance to the confirmed rating, leaving out times under 100 ms.
- Describes pace; the test has no time limit
- Arguments Rateditems
- Arguments rated in the scored run, of 34.
- A session with fewer than 26 is invalid
Use and evidence
32 papers from 2021 to 2025 name this task in their title or abstract, down 41% compared with 2016 to 2020. That ranks it 550 of 711 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 3 |
| 2024 | 4 |
| 2023 | 9 |
| 2022 | 10 |
| 2021 | 6 |
| 2020 | 8 |
| 2019 | 11 |
| 2018 | 12 |
| 2017 | 13 |
| 2016 | 10 |
| 2015 | 17 |
| 2014 | 11 |
| 2013 | 18 |
| 2012 | 6 |
| 2011 | 13 |
| 2010 | 18 |
| 2009 | 8 |
| 2008 | 4 |
| 2007 | 14 |
| 2006 | 10 |
| 2005 | 4 |
| 2004 | 9 |
| 2003 | 3 |
| 2002 | 1 |
| 2001 | 10 |
| 2000 | 5 |
Use grade D
30 to 99 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 1
- Registered clinical studies using it as an outcome
- 0
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
Note on the counts: Original arguments in each language; typical and atypical animals chosen by the author, not rated; no population norms..
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 0.58.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Arguments
- 34 arguments: 16 matched pairs, 8 about birds and 8 about mammals, and 2 checks from a whole category to one of its members.
- Pairs
- 6 typicality pairs (a typical or an atypical animal, such as blackbirds or penguins, toward all birds or all mammals), 6 diversity pairs (the same first animal with a similar or a dissimilar second one, toward the whole category) and 4 similarity pairs (an animal like or unlike a named conclusion animal).
- Properties
- Each argument has its own made-up kind of cell, named by a code such as Q56, so no property carries over and none can be known already.
- Answer
- A rating on a bar from 0 (very unlikely) to 10 (very likely), then Confirm, with no time limit and no feedback.
- Pictures and languages
- Each animal is shown with its Noto Emoji picture beside the sentence that names it. The names and sentences are written in each of the eight languages.
- Order
- The session's seed deals the codes and orders the arguments with no more than 3 of one category in a row and never the two of a pair in a row. All of it is saved.
- Practice
- 1 argument outside the set, with no feedback.
The arguments, animal pairs and sentences were written for this task; no published argument set is used. Typical and atypical animals were chosen by their features, not from a rating study.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: property_induction_argument from the scored run, one rating from 0 to 10 per argument. Ratings have no right answer, so accuracy is not judged. One rating on at least 90% of arguments is flagged for review. Ratings confirmed under 200 ms on at least half of 4 or more arguments invalidate. Fewer than 26 of the 34 arguments rated invalidates. A mean check rating under 5 is flagged for review. A hidden tab is flagged.
Why these rules
Inductive strength is a judgment with no correct value, so no answer is right or wrong. The same rating on nearly every argument can come from a person who sees no difference between them, so it flags and never excludes. An answer under 200 ms is faster than a simple visual reaction, so it cannot reflect the item. Following response-time effort screening for untimed tests (Wise and Kong 2005), the session is invalid only when at least half the answers are that fast. The effects compare matched pairs, which need most of the arguments rated. A fact about a whole category makes each member very likely to share it, so low check ratings suggest the facts were not read. Nothing runs while the tab is hidden, so a hidden tab is flagged.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
This test runs with fixed settings.
Need a different version of this test?
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 Property Induction Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Rips, 1975; Osherson, Smith, Wilkie, López & Shafir, 1990). The main score was diversity effect (mean, over the 6 diversity pairs, of the rating with a dissimilar second animal minus the rating with a similar one, on the 0 to 10 scale).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Rips, L. J. (1975). Inductive judgments about natural categories. Journal of Verbal Learning and Verbal Behavior, 14(6), 665-681. https://doi.org/10.1016/S0022-5371(75)80055-7
- Osherson, D. N., Smith, E. E., Wilkie, O., López, A., & Shafir, E. (1990). Category-based induction. Psychological Review, 97(2), 185-200. https://doi.org/10.1037/0033-295X.97.2.185
- Gelman, S. A., & Markman, E. M. (1986). Categories and induction in young children. Cognition, 23(3), 183-209. https://doi.org/10.1016/0010-0277(86)90034-XProperty induction in young children.
- Heit, E. (2000). Properties of inductive reasoning. Psychonomic Bulletin & Review, 7(4), 569-592. https://doi.org/10.3758/BF03212996Review of the similarity, typicality and diversity effects.
- Rhodes, M., Brickman, D., & Gelman, S. A. (2008). Sample diversity and premise typicality in inductive reasoning: Evidence for developmental change. Cognition, 108(2), 543-556. https://doi.org/10.1016/j.cognition.2008.03.002Diversity and typicality compared from 6 years to adulthood.
Source papers
- Rips LJ (1975). Inductive judgments about natural categories. Journal of Verbal Learning and Verbal Behavior, 14(6):665-681.Paradigm source, Adults
- Osherson DN, Smith EE, Wilkie O, López A, Shafir E (1990). Category-based induction. Psychological Review, 97(2):185-200.Paradigm source, Adults
- Rhodes M, Brickman D, Gelman SA (2008). Sample diversity and premise typicality in inductive reasoning: Evidence for developmental change. Cognition, 108(2):543-556.Paradigm source, 6 years to adults
Questions
How long does the Property Induction Task take?
About 5 to 8 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?
Diversity Effect, Typicality Effect, Similarity Effect, Mean Rating, Diverse Premises, Mean Rating, Similar Premises, Mean Rating, Typical Premise, Mean Rating, Atypical Premise, Mean Rating, Similar Premise, Mean Rating, Dissimilar Premise, Mean Rating, Check Rating, Median Response Time and Arguments Rated. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Property Induction Task used in research?
32 papers from 2021 to 2025 name it in their title or abstract, down 41% compared with 2016 to 2020. That ranks it 550 of 711 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.58.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Property Induction 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 languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese, with its words in the same language. 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.