Theory of Mind
Definition, everyday examples, false-belief tests, development, autism research, and the limits of measuring how people infer mental states.
Published June 12, 2026 · Updated July 12, 2026 · Evidence links checked July 12, 2026
Theory of mind (ToM) is the capacity to represent or infer mental states—including beliefs, knowledge, desires, intentions, and emotions—in yourself and other people, and to use those inferences to explain or anticipate behavior.
It is an umbrella construct, not literal mind reading. Mental states are not directly observable, so an inference can be useful and still be mistaken.
What theory of mind means in psychology
Theory of mind is a name for reasoning about mental states. If you explain a person’s action by saying that they believe the shop is open, want a quiet evening, or do not know the meeting moved, you are using a theory-of-mind inference. The key move is separating another person’s representation of a situation from both reality and your own perspective.
Premack and Woodruff introduced the label in a 1978 paper asking whether a chimpanzee could attribute goals and intentions. “Theory” here does not mean that people consciously build a scientific theory. It means that beliefs, desires, and intentions are hidden causes inferred from behavior and context. The underlying human capacities existed before researchers gave them this name.
Researchers often distinguish cognitive theory of mind—inferences about beliefs, knowledge, goals, and intentions—from affective theory of mind—inferences about feelings. A lesion study found partial separation between performance on cognitive and affective tasks, but these labels should not be treated as two perfectly isolated mental modules.
Theory of mind examples
Everyday theory of mind is usually an interpretation held with some uncertainty, not a test with one guaranteed answer. Context and direct questions can confirm or overturn the first inference.
A coworker missed the schedule update, so you expect them to arrive at the old time even though you know the new one.
Your partner was absent when the keys moved, so you predict they will first search the place where they left them.
You would choose a party, but your friend prefers a quiet evening. You recommend the option that fits their preference.
A flat reply might signal disappointment, fatigue, concentration, or nothing unusual. You keep alternatives open and check rather than assume.
Ana thinks that Ben missed an announcement. You represent Ana’s belief about what Ben knows.
The words “great timing” may be sincere or sarcastic. Tone, situation, and shared history shape the inference.
How theory of mind develops
Theory of mind is not one ability that switches on at a universal age. Children show components of social understanding before they can answer an explicit false-belief question, and increasingly complex reasoning continues after they pass one. A 2001 meta-analysis of 178 studies found a marked age-related change in false-belief performance, while country and task features also explained meaningful variation.
Preschool research often finds a progression across tasks involving diverse desires, diverse beliefs, access to knowledge, false beliefs, and hidden emotion. In a 2004 scale study, 75 children showed a reliable ordering across seven tasks. The order is a population-level pattern, not a timetable that every child must follow.
Language and executive function matter to task performance. A meta-analysis covering 104 studies and 8,891 children found a moderate-to-large association between language ability and false-belief understanding after accounting for age. Another meta-analysis, covering 102 studies and 9,994 children, found a moderate association between executive function and false-belief understanding, with results varying by task type. These are associations and task contributions; they do not show that language or executive function alone causes theory of mind.
False-belief tests: what they show
False-belief tasks ask whether someone can predict behavior from another person’s mistaken belief rather than from reality. Wimmer and Perner’s 1983 “Maxi” study established the classic structure. The familiar Sally–Anne task was a later adaptation used in a 1985 autism study.
The theory-of-mind answer is the old location: the character acts from what they believe, not from where the object really is. Good task design also checks whether the participant remembers the original and current locations.
A pass or fail samples one kind of explicit reasoning under one set of language, memory, inhibition, and attention demands. It is not a clinical diagnosis, a complete score of social ability, or proof that a person always interprets others accurately in daily life.
How theory of mind is measured—and why measures disagree
Researchers use first- and second-order false-belief stories, unexpected-contents tasks, Strange Stories, faux-pas tasks, the Reading the Mind in the Eyes Test (RMET), and naturalistic film-based measures such as the MASC. These tasks differ in whether they emphasize belief representation, language, emotion recognition, memory, or real-time interpretation. They should not be treated as interchangeable versions of one pure test.
A 2020 conceptual review argued that many tasks commonly labeled theory of mind do not uniquely isolate mental-state reasoning. The RMET is a useful example: Oakley and colleagues found that its performance may reflect emotion recognition and alexithymia as much as a distinct theory-of-mind capacity. A result should therefore be interpreted in relation to the exact task, comparison group, and question being studied.
Theory of mind and autism: a more accurate view
Theory-of-mind tasks have played a major role in autism research since Baron-Cohen, Leslie, and Frith used a Sally–Anne false-belief task in 1985. That historical study found a group difference, but it used a small sample and did not show that every autistic person lacks theory of mind.
Research since then has found substantial variation across autistic people, tasks, language demands, and contexts. Some autistic people find particular mental-state inferences difficult, while others pass standard tasks; different theory-of-mind measures often do not converge cleanly. Gernsbacher and Yergeau’s 2019 review documents empirical problems with treating a global absence of theory of mind as a defining fact about autism. A score on one task cannot establish a global absence of social understanding.
Social understanding is also reciprocal. In a specific adult study, information moved effectively through autistic-to-autistic and non-autistic-to-non-autistic communication chains, while mixed-neurotype chains showed lower transfer. Related rapport research also found effects of neurotype matching. These findings support interactional mismatch as one contributor; they do not prove that neurotype matching explains every communication difficulty.
Theory-of-mind task performance is not the same as empathy, compassion, caring, or moral character. No theory-of-mind task can diagnose or rule out autism by itself.
Theory of mind vs empathy, perspective-taking, and mentalizing
| Concept | Core question | Important distinction |
|---|---|---|
| Theory of mind | What might this person believe, know, want, feel, or intend? | A representational inference; it can be wrong and does not guarantee caring. |
| Perspective-taking | How does the situation look from this point of view? | Often a deliberate strategy and sometimes narrower than the full ToM umbrella. |
| Cognitive empathy | Can I identify or understand another person’s emotional perspective? | Overlaps with affective ToM, but terminology varies across research traditions. |
| Affective empathy | Do I emotionally resonate with another person’s state? | Emotional sharing is different from inferring a belief or intention. |
| Mentalizing | How do mental states in myself and others make behavior meaningful in this context? | A broader, context-sensitive clinical framework associated with reflective functioning, attachment, and mentalization-based treatment. |
Researchers do not always use these labels consistently, so the operational definition and task matter more than the label alone.
Common questions
What is theory of mind in simple terms?
It is the ability to reason about what you or another person may think, know, want, intend, or feel—especially when that mental state differs from reality or from your own. It is inference, not literal access to another mind.
At what age does theory of mind develop?
Components develop gradually. Explicit false-belief performance often changes markedly between about ages three and five, but age, language, executive demands, culture, and task design all affect performance. Passing one task is not an endpoint.
Is theory of mind the same as empathy?
No. Theory of mind concerns inferring mental states. Affective empathy concerns emotional resonance, and compassion concerns a caring response. A person can be strong in one and less consistent in another.
Do autistic people lack theory of mind?
No universal claim is supported. Group differences occur on some tasks, but autistic people vary widely, measures disagree, and communication fit matters. One task cannot establish a global lack of social understanding, empathy, or care.
Can a theory-of-mind test diagnose autism?
No. False-belief and related tasks are research measures with language, memory, attention, and executive demands. Autism diagnosis requires comprehensive clinical evaluation and developmental context.
Primary research and evidence reviews
- Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences. DOI 10.1017/S0140525X00076512.
- Wimmer, H., & Perner, J. (1983). Beliefs about beliefs. Cognition. DOI 10.1016/0010-0277(83)90004-5; PMID 6681741.
- Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a “theory of mind”? Cognition. DOI 10.1016/0010-0277(85)90022-8; PMID 2934210.
- Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development. Child Development. DOI 10.1111/1467-8624.00304; PMID 11405571.
- Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development. DOI 10.1111/j.1467-8624.2004.00691.x; PMID 15056204.
- Milligan, K., Astington, J. W., & Dack, L. A. (2007). Language and theory of mind: meta-analysis. Child Development. DOI 10.1111/j.1467-8624.2007.01018.x; PMID 17381794.
- Shamay-Tsoory, S. G., & Aharon-Peretz, J. (2007). Dissociable prefrontal networks for cognitive and affective theory of mind. Neuropsychologia. DOI 10.1016/j.neuropsychologia.2007.05.021; PMID 17640690.
- Devine, R. T., & Hughes, C. (2014). False-belief understanding and executive function. Child Development. DOI 10.1111/cdev.12237; PMID 24605760.
- Oakley, B. F. M., et al. (2016). Theory of mind is not theory of emotion. Journal of Abnormal Psychology. DOI 10.1037/abn0000182; PMID 27505409.
- Gernsbacher, M. A., & Yergeau, M. (2019). Empirical failures of the claim that autistic people lack a theory of mind. Archives of Scientific Psychology. DOI 10.1037/arc0000067; PMID 31938672.
- Quesque, F., & Rossetti, Y. (2020). What do theory-of-mind tasks actually measure? Perspectives on Psychological Science. DOI 10.1177/1745691619896607; PMID 32069168.
- Crompton, C. J., et al. (2020). Autistic peer-to-peer information transfer is highly effective. Autism. DOI 10.1177/1362361320919286.
- Crompton, C. J., et al. (2020). Neurotype-matching, but not being autistic, influences self and observer ratings of interpersonal rapport. Frontiers in Psychology. DOI 10.3389/fpsyg.2020.586171.