Social Cognition
Mind Perception Task
Also called: mind perception survey, dimensions of mind perception, experience and agency, mind attribution task, perceived mind
Measure how people see the minds of others. Participants compare a baby, an adult, a dog, a fish, a robot and a computer on capacities such as feeling pain and making plans.
6-9 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
- 305
- Evidence grade
- D
- Registered clinical studies
- 1
Which one is more able to feel pain?
The baby
The robotWhat it measures
Measures how people see the minds of others. Participants compare 6 characters, a baby, an adult, a dog, a fish, a robot and a computer, two at a time on capacities such as feeling pain or making plans. The scores place each character on experience and agency, the two dimensions of mind perception.
Constructs
- Mind perception
- Experience and agency
- Theory of mind
- Anthropomorphism
Research uses
- Social cognition research
- Moral psychology research
- Human-robot and human-computer interaction research
- Within-study group comparison
How the paradigm works
The mind perception task measures how people see the minds of others. Gray, Gray and Wegner (2007) asked 2,040 people on a web survey to compare 13 characters, among them a five-month-old infant, an adult woman and man, a family dog, a frog, a wild chimpanzee and a sociable robot, two at a time on one of 18 mental capacities, such as feeling pain or planning, on five-point scales. A principal components analysis of the characters' means found two dimensions rather than one. Experience, the capacity to feel (hunger, fear, pain, pleasure, rage, desire, personality, consciousness, pride, embarrassment and joy), accounted for 88% of the variance, and agency, the capacity to act and plan (self-control, morality, memory, emotion recognition, planning, communication and thought), for 8%. Liking a character, wanting to save it and seeing it as having a soul rose with both dimensions, while punishment for a death went more with agency and reluctance to harm more with experience.
More about the paradigm
Here, participants see six characters, drawn as pictures and named in their language: a baby five months old, an adult, a pet dog, a fish, a robot that can move and talk, and a laptop computer. They compare them two at a time on six capacities, three of each dimension: feeling pain, fear and hunger, and resisting temptation, remembering and making plans. A ten-capacity form adds joy, embarrassment, telling right from wrong and understanding how others feel. Each capacity is one block of all 15 pairs, with five answers from much more the one on the left to much more the one on the right and about the same in the middle. The order of the blocks and of the pairs, and which character stands on the left, are drawn from the session's seed, and each pair stands each way round equally often. The characters are introduced first, two practice pairs ask which one is more able to swim, and there is no time limit. No real person, religious figure, or dead or ill person appears, and a fish takes the place of the frog.
Each comparison gives the character on the left 2 to -2 and the other the opposite. A character's score on a capacity is the mean of what its comparisons gave it, and its experience and agency scores average the capacities of each dimension, from -2 to 2, where 0 is the middle of the six characters. The main score is the correlation across the six characters between experience and agency: near 1, a person sees more or less mind on one scale; lower, the two dimensions come apart, as in Gray and colleagues' study. The saved comparisons also let a study run the principal components analysis across its participants. Scores describe how people see the minds of these characters; they are research measures for comparing groups, and they do not describe a person's character or diagnose.
Scores
- Experience-Agency Correlationscore
- The Pearson correlation across the six characters between their experience and agency scores. Null with fewer than three characters placed or no spread on either dimension.
- Near 1 means the minds were seen on one scale of more or less mind; lower values mean the two dimensions came apart, as Gray, Gray and Wegner (2007) found
- Baby: Experiencescore
- The baby's experience score, from -2 to 2: for each capacity to feel (pain, fear and hunger, with joy and embarrassment in the 10-capacity form), the mean of what its comparisons gave it, averaged over those capacities.
- Positive means the baby was seen as more able to feel than the middle of the six characters
- Baby: Agencyscore
- The baby's agency score, from -2 to 2: for each capacity to act and plan (resisting temptation, remembering and making plans, with telling right from wrong and understanding how others feel in the 10-capacity form), the mean of what its comparisons gave it, averaged over those capacities.
- Positive means the baby was seen as more able to act and plan than the middle of the six characters
- Adult: Experiencescore
- The adult's experience score, from -2 to 2, computed as for the baby.
- Positive means the adult was seen as more able to feel than the middle of the six characters
- Adult: Agencyscore
- The adult's agency score, from -2 to 2, computed as for the baby.
- Positive means the adult was seen as more able to act and plan than the middle of the six characters
- Dog: Experiencescore
- The dog's experience score, from -2 to 2, computed as for the baby.
- Positive means the dog was seen as more able to feel than the middle of the six characters
- Dog: Agencyscore
- The dog's agency score, from -2 to 2, computed as for the baby.
- Positive means the dog was seen as more able to act and plan than the middle of the six characters
- Fish: Experiencescore
- The fish's experience score, from -2 to 2, computed as for the baby.
- Positive means the fish was seen as more able to feel than the middle of the six characters
- Fish: Agencyscore
- The fish's agency score, from -2 to 2, computed as for the baby.
- Positive means the fish was seen as more able to act and plan than the middle of the six characters
- Robot: Experiencescore
- The robot's experience score, from -2 to 2, computed as for the baby.
- Positive means the robot was seen as more able to feel than the middle of the six characters
- Robot: Agencyscore
- The robot's agency score, from -2 to 2, computed as for the baby.
- Positive means the robot was seen as more able to act and plan than the middle of the six characters
- Computer: Experiencescore
- The computer's experience score, from -2 to 2, computed as for the baby.
- Positive means the computer was seen as more able to feel than the middle of the six characters
- Computer: Agencyscore
- The computer's agency score, from -2 to 2, computed as for the baby.
- Positive means the computer was seen as more able to act and plan than the middle of the six characters
- Answers About the Same%
- The share of comparisons answered "about the same".
- Informational
- Comparisons Answeredcount
- Comparisons answered, out of 90 (150 in the 10-capacity form).
- Informational
- Median Response Timems
- Median time from a comparison appearing to an answer, over answers at or after 100 ms.
- Informational
- Answers Under 100 mscount
- Answers made less than 100 ms after their comparison appeared.
- 0 expected
Use and evidence
305 papers from 2021 to 2025 name this task in their title or abstract, up 82% compared with 2016 to 2020. That ranks it 100 of 670 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 105 |
| 2024 | 62 |
| 2023 | 58 |
| 2022 | 37 |
| 2021 | 43 |
| 2020 | 43 |
| 2019 | 35 |
| 2018 | 31 |
| 2017 | 32 |
| 2016 | 27 |
| 2015 | 24 |
| 2014 | 15 |
| 2013 | 22 |
| 2012 | 19 |
| 2011 | 14 |
| 2010 | 5 |
| 2009 | 7 |
| 2008 | 3 |
| 2007 | 7 |
| 2006 | 1 |
| 2005 | 1 |
| 2004 | 5 |
| 2003 | 2 |
| 2002 | 3 |
| 2001 | 2 |
| 2000 | 1 |
Use grade B
300 to 999 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 29
- Registered clinical studies using it as an outcome
- 1
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: includes studies that rate the minds of robots, brands and AI with questionnaires.
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 0.48.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Characters
- 6 characters, each a Noto Emoji picture with its name: a baby five months old, an adult, a pet dog, a fish, a robot that can move and talk, and a laptop computer. No real person, religious figure, or dead or ill person appears.
- Capacities
- 6 by default, 3 for each dimension: feeling pain, fear and hunger (experience), and resisting temptation, remembering and making plans (agency). A 10-capacity form adds joy, embarrassment, telling right from wrong and understanding how others feel.
- Comparisons
- Every pair of characters once on each capacity, 15 pairs a capacity: 90 comparisons, or 150 in the 10-capacity form. Each asks which one is more able, with five answers from much more the one on the left to much more the one on the right, and about the same in the middle.
- Order
- One capacity at a time, each introduced by its question. The order of the capacities and of the pairs, and which character stands on the left, are drawn from the session's seed; each pair stands each way round equally often.
- Practice
- The six characters are introduced first, then 2 practice pairs ask which one is more able to swim, with the answer shown back. There is no time limit.
- Scores
- Each character's experience and agency, from -2 to 2: for each capacity, the mean of what its comparisons gave it, averaged over the capacities of each dimension. The main score is the correlation across the six characters between experience and agency: near 1, a person sees more or less mind on one scale; lower, the two dimensions come apart.
- Languages
- Questions, names and instructions in English, Italian, German, Spanish, French, Japanese, Portuguese and Chinese.
Gray, Gray and Wegner (2007) found the two dimensions by a principal components analysis across 13 characters, each respondent comparing every pair on one of 18 capacities; here each person compares six characters on several capacities, and each dimension averages the capacities their analysis placed in it. The saved comparisons let a study run the principal components analysis across its own participants. A fish takes the place of their frog, and their God, dead woman, fetus, man in a persistent vegetative state and the respondent are left out. Scores describe how people see these characters' minds and are best compared between groups; they do not describe a person's character and do not diagnose.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: the scored run's comparisons. No answer is correct, so accuracy is not judged against chance. A run with no comparison answered, or with fewer than half the planned comparisons answered, is invalid, and any other shortfall is flagged; one answer chosen for at least 90% of at least 20 comparisons is flagged. Answers under 200 ms on at least half the screens invalidate.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. The scores describe how a person compares the minds of six characters, so every answer is possible, "about the same" too. One answer for nearly every pair suggests the answers were not read; it is flagged for review because it can also be a real view. 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.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Capacities | 6 capacities, 90 comparisons (standard) | 6 capacities, 90 comparisons (standard), 10 capacities, 150 comparisons |
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 Mind Perception Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Gray HM, Gray K, Wegner DM, 2007; Waytz A, Gray K, Epley N, Wegner DM, 2010), with the standard settings (capacities: 6 capacities, 90 comparisons). The main score was experience-agency correlation (the Pearson correlation across the six characters between their experience and agency scores).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Gray, H. M., Gray, K., & Wegner, D. M. (2007). Dimensions of mind perception. Science, 315(5812), 619. https://doi.org/10.1126/science.1134475
- Waytz, A., Gray, K., Epley, N., & Wegner, D. M. (2010). Causes and consequences of mind perception. Trends in Cognitive Sciences, 14(8), 383-388. https://doi.org/10.1016/j.tics.2010.05.006A review of when people see minds in others, why, and what follows, such as moral rights.
Source papers
- Gray, Gray and Wegner (2007): 2,399 web surveys completed by 2,040 people, averaging 30 years of age; experience accounted for 88% and agency for 8% of the variance in the characters' means.N 2040, Adults (mean age 30)
Questions
How long does the Mind Perception Task take?
About 6 to 9 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?
Experience-Agency Correlation, Baby: Experience, Baby: Agency, Adult: Experience, Adult: Agency, Dog: Experience, Dog: Agency, Fish: Experience, Fish: Agency, Robot: Experience, Robot: Agency, Computer: Experience, Computer: Agency, Answers About the Same, Comparisons Answered, Median Response Time and Answers Under 100 ms. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Mind Perception Task used in research?
305 papers from 2021 to 2025 name it in their title or abstract, up 82% compared with 2016 to 2020. That ranks it 100 of 670 test families in the ConductCognition test census (use grade B). Source: ConductCognition test census 0.48.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Mind Perception Task have?
In the published research, 0 papers report its test-retest reliability and 1 registered clinical study uses 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?
Capacities (6 capacities, 90 comparisons (standard), 10 capacities, 150 comparisons). 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, 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.