Social Cognition
Mental State Stories Task
Also called: advanced theory of mind task, nonliteral language stories, mental state attribution stories
Measure how people work out why someone says something not meant literally: participants read short everyday stories and choose why, with matched physical stories as controls.
10-13 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
- 174
- Evidence grade
- C
- Registered clinical studies
- 10
Ana gave Tom a scarf he did not like. “I love it,” Tom said.
Ana gave Tom a scarf he did not like. “I love it,” Tom said.
Why did Tom say that?What it measures
Measures how well people work out why someone says something not meant literally. Participants read short everyday stories in which a person tells a white lie, jokes, pretends, speaks with irony, bluffs or persuades, and matched stories about physical events, and choose the best of four answers to why. The main score is percent right on the mental-state stories minus percent right on the physical stories. The stories are in English.
Constructs
- Theory of mind
- Mental state attribution
- Nonliteral language
- Social cognition
Research uses
- Theory of mind research in adults
- Autism research
- Schizophrenia and social cognition research
- Aging and social cognition research
How the paradigm works
The Mental State Stories Task measures how well people work out why someone says something not meant literally. Happé (1994) wrote short everyday stories that end in such a remark and asked why the person said it: understanding the answer needs the speaker's thoughts and intentions, not only the words. Fletcher and colleagues (1995) compared them with physical stories, in which something happens for a physical reason and no one's mind needs explaining. Jolliffe and Baron-Cohen (1999) gave the stories to adults, and White, Hill, Happé and Frith (2009), adapting them for children, matched the control stories for difficulty. Story tests of this kind are used in autism research (Happé 1994; Jolliffe and Baron-Cohen 1999) and in aging research, where Maylor and colleagues (2002) compared younger and older adults on mental-state and control stories.
More about the paradigm
The 24 stories here were written for this test; no published story is used. In 12 of them a person tells a white lie, makes a joke, pretends, speaks with irony, tells the truth to be disbelieved (a double bluff) or says something to persuade, two stories of each. The other 12 turn on a physical cause, such as freezing, wind or rust, and use no word for a mental state. Stories are 44 to 50 words, with the same mean length in both kinds. Under each story, a question asks why the person said it, or why the thing happened, and the participant chooses from four answers scored by rule, where the published versions asked for a spoken or written answer that a rater coded. The answers are the best answer; the words taken at face value, or in a physical story someone's wish; a reason that does not fit the story; and a true detail that explains nothing. The story stays on the screen, with no time limit. Two practice stories, one of each kind, have feedback. The session's seed orders the stories and the answers, with the best answer in each place equally often. A setting gives 12 stories, 6 of each kind, in about half the time.
The main score is percent right on the mental-state stories minus percent right on the physical stories, so reading and choosing count the same on both sides. The test also reports each kind's percent right and how often each wrong kind of answer was chosen: the words taken at face value and a reason that does not fit the story, on the mental-state stories, and someone's wish, on the physical stories. A session needs at least half of the physical stories right. The stories are English; sessions run in other languages are flagged for review. Scores support research and do not diagnose.
Scores
- Mental State Minus Physical%
- Percent right on the mental-state stories minus percent right on the physical stories.
- Below zero means the remarks were harder to explain than the physical events
- Mental State Stories%
- Percent of mental-state stories answered with the best answer.
- Higher is better
- Physical Stories%
- Percent of physical stories answered with the best answer.
- Below 50% invalidates
- Literal Answers%
- Percent of mental-state stories answered with the remark taken at face value.
- Informational
- Answers That Do Not Fit%
- Percent of mental-state stories answered with a reason about the mind that does not fit the story.
- Informational
- Wishes in Physical Stories%
- Percent of physical stories answered with someone's wish or plan.
- Informational
- Detail Answers%
- Percent of all stories answered with a true detail that explains nothing.
- Informational
- Median Answer Time, Mental Statems
- Median time from the story's screen to a best answer on mental-state stories.
- Informational; there is no time limit
- Median Answer Time, Physicalms
- Median time from the story's screen to a best answer on physical stories.
- Informational; there is no time limit
- Stories Answereditems
- Stories answered in the scored run.
- Fewer than set is flagged for review
- Stories Setitems
- Stories the session set out to show.
- Informational
- Anticipationscount
- Answers under 100 ms.
- Lower supports interpretation
Use and evidence
174 papers from 2021 to 2025 name this task in their title or abstract, down 14% compared with 2016 to 2020. That ranks it 155 of 495 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 40 |
| 2024 | 34 |
| 2023 | 33 |
| 2022 | 35 |
| 2021 | 32 |
| 2020 | 48 |
| 2019 | 42 |
| 2018 | 43 |
| 2017 | 31 |
| 2016 | 38 |
| 2015 | 33 |
| 2014 | 26 |
| 2013 | 33 |
| 2012 | 32 |
| 2011 | 38 |
| 2010 | 41 |
| 2009 | 40 |
| 2008 | 27 |
| 2007 | 22 |
| 2006 | 31 |
| 2005 | 23 |
| 2004 | 21 |
| 2003 | 15 |
| 2002 | 6 |
| 2001 | 15 |
| 2000 | 3 |
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
- 6
- Papers on reliability or validity
- 35
- Registered clinical studies using it as an outcome
- 10
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 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Stories
- 24 stories written for this test: 12 in which a person says something not meant word for word, two each of a white lie, a joke, pretending, irony, a double bluff and persuasion, and 12 in which something happens for a physical reason.
- Length
- 44 to 50 words each, with the same mean length within a word in both kinds.
- Question
- Why the person said it, or why the thing happened, with four answers: the best answer; the words taken at face value, or in a physical story someone's wish; a reason that does not fit the story; and a true detail that explains nothing.
- Session
- 24 stories by default, or 12 as a setting, 6 of each kind with one of each mental-state kind. The session's seed orders them, with no more than 3 of one kind in a row and no two of the same mental-state kind together, and puts the best answer in each of the four places equally often within each kind.
- Time
- No time limit. The story stays on the screen while the participant chooses.
- Practice
- 2 stories, one of each kind, with feedback. The test has no feedback.
The stories were written for this test; no published story is used, so published scores and cutoffs do not apply to it. The stories are English; sessions run in other languages are flagged for review.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: mental_state_story from the scored run, one choice of four per story, the answer in each place equally often, so chance is 25%. Accuracy significantly below chance invalidates. One place on at least 90% of answers invalidates unless accuracy is above chance. Answers under 200 ms on at least half of 4 or more answers invalidate. Fewer than half of the physical stories right invalidates. A run short of its stories and a session run in another language are flagged for review. A hidden tab is flagged.
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). 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 physical stories ask what made something happen, which needs no reasoning about a mind and is answered right nearly always by a person who reads the story; a guesser gets half or more of 12 of them right about 5 times in 100. Below half, the answers do not show that the stories were read, and the main score, a difference from the physical stories, would rise for the wrong reason. The stories are English; a session in another language shows them to a participant who may not read English fluently, so it is flagged for review. The story stays on the screen with no time limit, so a hidden tab changes nothing in the procedure and is only flagged.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Stories | 24 stories (standard) | 12 stories, 24 stories (standard) |
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
- Instructions and screens: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. Words and sentences: English.
- 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 Mental State Stories Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Happé FGE, 1994; Fletcher PC, Happé F, Frith U, et al., 1995; Jolliffe T, Baron-Cohen S, 1999; White S, Hill E, Happé F, Frith U, 2009), with the standard settings (stories: 24 stories). The main score was mental state minus physical (percent right on the mental-state stories minus percent right on the physical stories).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Happé, F. G. E. (1994). An advanced test of theory of mind: Understanding of story characters' thoughts and feelings by able autistic, mentally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24(2), 129-154. https://doi.org/10.1007/BF02172093Stories that end in a remark not meant literally, and the question why it was said.
- Fletcher, P. C., Happé, F., Frith, U., Baker, S. C., Dolan, R. J., Frackowiak, R. S. J., & Frith, C. D. (1995). Other minds in the brain: A functional imaging study of "theory of mind" in story comprehension. Cognition, 57(2), 109-128. https://doi.org/10.1016/0010-0277(95)00692-RPhysical stories as the control.
- Jolliffe, T., & Baron-Cohen, S. (1999). The Strange Stories Test: A replication with high-functioning adults with autism or Asperger syndrome. Journal of Autism and Developmental Disorders, 29(5), 395-406. https://doi.org/10.1023/A:1023082928366The stories in adults, with a control task that needs no mental states.
- White, S., Hill, E., Happé, F., & Frith, U. (2009). Revisiting the strange stories: Revealing mentalizing impairments in autism. Child Development, 80(4), 1097-1117. https://doi.org/10.1111/j.1467-8624.2009.01319.xControl stories matched for difficulty.
- Maylor, E. A., Moulson, J. M., Muncer, A.-M., & Taylor, L. A. (2002). Does performance on theory of mind tasks decline in old age? British Journal of Psychology, 93(4), 465-485. https://doi.org/10.1348/000712602761381358Story tests with control stories in aging research.
Source papers
- Happé FGE (1994). An advanced test of theory of mind: Understanding of story characters' thoughts and feelings by able autistic, mentally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24(2):129-154.Paradigm source, Children and adults
Questions
How long does the Mental State Stories Task take?
About 10 to 13 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?
Mental State Minus Physical, Mental State Stories, Physical Stories, Literal Answers, Answers That Do Not Fit, Wishes in Physical Stories, Detail Answers, Median Answer Time, Mental State, Median Answer Time, Physical, Stories Answered, Stories Set and Anticipations. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Mental State Stories Task used in research?
174 papers from 2021 to 2025 name it in their title or abstract, down 14% compared with 2016 to 2020. That ranks it 155 of 495 test families in the ConductCognition test census (use grade C). Source: ConductCognition test census 0.22.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Mental State Stories Task have?
In the published research, 6 papers report its test-retest reliability and 10 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?
Stories (12 stories, 24 stories (standard)). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
The instructions and screens are in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. The words and sentences stay in English, so sessions in other languages are flagged for review. 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.