What Are Examples of Theory of Mind in Everyday Life?

Theory of mind shows up every time you realize someone else doesn’t know what you know. A friend walks into a surprise party looking genuinely shocked, and you understand her reaction because you can separate your own knowledge (you helped plan it) from hers (she had no idea). That mental move, attributing beliefs, desires, and knowledge states to other people that differ from your own, is theory of mind in action. The concept sounds academic, but it’s woven into nearly every social interaction you have, and the examples range from toddlers figuring out deception to ravens outsmarting each other over hidden food.

The Surprise-Party Problem and False Belief

The most famous example used to study theory of mind is the false belief task. In the classic version, a child watches a puppet show: Sally puts a marble in a basket and leaves the room. While she’s gone, Anne moves the marble to a box. The experimenter asks the child, “Where will Sally look for her marble?” The correct answer is the basket, because Sally didn’t see the switch. But to get there, you have to represent Sally’s mind as holding a belief you know is wrong. That’s the false belief task in a nutshell, and it’s been the workhorse of theory of mind research for decades.

Most children pass this test around age four or five. Before that age, kids typically say Sally will look in the box, where the marble actually is. They can’t separate what they know from what Sally knows. Researchers have traditionally treated passing this task as a landmark, the moment a child shifts from an earlier, more intuitive form of social understanding to a more deliberate, explicit version of theory of mind. That transition is generally thought to consolidate between ages four and six.¹

But the picture is more interesting than “young kids fail, older kids pass.” A 2020 study found that three-year-olds performed just as well as older children on false belief trials when they themselves didn’t know the actual location of the object. In other words, when the experimenters removed the child’s own privileged knowledge, the younger kids could reason about someone else’s false belief just fine. The difficulty wasn’t that three-year-olds lacked the ability to think about other minds. It was that their own knowledge got in the way.² This suggests the classic task measures two things at once: theory of mind and the ability to override your own perspective, which is a separate cognitive skill.

Everyday Examples You Might Not Recognize

False belief tasks are useful for research, but theory of mind in daily life is far more varied and constant. Consider these situations, each of which requires you to model what’s happening in someone else’s head:

White lies: You tell a friend their new haircut looks great even though you’re not a fan. You do this because you understand how they’d feel if they heard your honest opinion.
Sarcasm: A coworker says “Well, that meeting was productive” after a disastrous hour of arguing. You laugh because you understand they believe the opposite of what they said, and they know you know that.
Teaching: When you explain something to a child, you automatically adjust your vocabulary because you model their knowledge as less than yours. You’d explain the same thing differently to a colleague.
Persuasion: A salesperson pitches a product differently depending on whether the customer seems budget-conscious or quality-focused. They’re reading the customer’s priorities and tailoring the message.

Sarcasm is worth pausing on because it’s a layered example. To understand sarcasm, you need to track at least three things: the speaker’s actual belief, the literal meaning of their words, and the speaker’s intention that you recognize the mismatch. That’s multiple levels of mental-state reasoning packed into a single sentence. Children who haven’t yet developed robust theory of mind tend to take sarcastic statements literally, which is why “nice job” said sarcastically to a five-year-old often produces confusion rather than laughter.

Two Flavors of Theory of Mind

Not all mental-state reasoning works the same way. Researchers distinguish between cognitive theory of mind, which involves figuring out what someone thinks or believes, and affective theory of mind, which involves figuring out what someone feels. These sound like they’d be the same skill, but they rely on partly different brain networks and can break down independently of each other.

The cognitive variety leans on regions like the precuneus and parts of the temporal lobes, while the affective variety recruits prefrontal structures and areas involved in emotional processing. Brain imaging studies have found that a classifier could distinguish which type of reasoning a person was doing with better than 85% accuracy just by looking at patterns of brain activity.³ Lesion studies tell a similar story: damage to the ventromedial prefrontal cortex tends to impair affective theory of mind specifically, while broader prefrontal damage is needed to disrupt cognitive theory of mind.⁴

Here’s an example that separates the two. Imagine you’re watching a friend open a gift box. Cognitive theory of mind lets you realize she expects a necklace inside (because she saw the jewelry-store wrapping) even though you know it’s actually a prank rubber spider. Affective theory of mind lets you predict that she’ll feel startled and maybe annoyed when she sees the spider. The first is about tracking beliefs; the second is about predicting emotions. People with damage to different brain areas can lose one of these capacities while keeping the other largely intact.

In adolescence, these two types of reasoning are also associated with different cognitive skills. Cognitive theory of mind performance correlates with working memory, language comprehension, and certain kinds of intelligence, whereas affective theory of mind is more closely tied to attention and emotional intelligence.⁵ This matters practically: a teenager who struggles to read emotions in a friend’s face isn’t necessarily bad at understanding beliefs, and vice versa.

The Brain Network Behind It

Theory of mind isn’t housed in a single brain region. Neuroimaging reviews have identified a distributed network with several core nodes, including parts of the prefrontal cortex and the superior temporal sulcus, along with additional regions that pitch in depending on the specific task.⁶ When people read stories about characters’ thoughts, three areas respond selectively: the right and left temporo-parietal junction and the posterior cingulate. The medial prefrontal cortex, by contrast, activates more broadly for social reasoning in general, not just for tracking beliefs.⁷

This network also connects to a separate system involved in understanding physical actions, sometimes called the mirror system. When you watch cooperative interactions (two people building something together), connectivity between action-processing regions ramps up. When you watch emotionally charged interactions, connectivity between those same regions and the ventromedial prefrontal cortex strengthens instead.⁸ The brain, in other words, routes social information through different pathways depending on whether you’re trying to understand what someone is doing versus what they’re feeling.

Language as a Scaffold

There’s a chicken-and-egg quality to theory of mind and language. To talk about someone else’s thoughts, you need sentences like “Sally thinks the marble is in the basket,” where one clause is embedded inside another. And research suggests that mastering this kind of sentence structure, called syntactic complementation, actively helps children reason about false beliefs. After about age four, language development appears to serve as a scaffold for more complex mental-state reasoning.⁹

This has practical implications. Children with language delays often show delayed theory of mind development, and deaf children raised without early access to a signed language can be significantly behind in false belief tasks compared to deaf children who grow up in signing households. The mental-state reasoning depends, at least in part, on having the linguistic tools to represent one belief nested inside another.

When Theory of Mind Breaks Down

The clearest clinical example involves autism spectrum conditions, where difficulty reading other people’s mental states is a core feature. The Reading the Mind in the Eyes Test, which asks participants to identify emotions from photographs of eyes alone, is one of the most widely used assessments for theory of mind differences in autistic individuals.¹⁰ But reduced theory of mind isn’t limited to autism. It shows up across a range of conditions, and sometimes in surprising directions.

In frontotemporal dementia affecting the frontal lobes, patients show broad impairments across theory of mind tasks, from simple false belief tests to more subtle tests involving faux pas detection. Yet their basic comprehension and memory for the stories remain intact, meaning the difficulty is specifically in reasoning about mental states, not in understanding the scenario. Alzheimer’s disease, by comparison, shows a different pattern: patients struggle mainly on more complex theory of mind tasks that demand heavy use of working memory, and their errors on faux pas tests tend to be about forgetting details rather than failing to grasp social nuance.¹¹ A further study found that the two diseases impair different components: Alzheimer’s patients struggled more with inferring someone else’s belief, while frontotemporal dementia patients were selectively worse at inhibiting their own perspective.¹²

Then there’s the opposite problem. Some children don’t under-read other people’s mental states but over-read them. Hyper-theory-of-mind refers to the tendency to attribute thoughts, intentions, or conspiracies to others that aren’t actually there. In children reporting psychotic experiences, hyper-theory-of-mind was significantly more common, and the association was strongest with paranoid thinking.¹³ This is a good reminder that theory of mind isn’t simply “more is better.” The goal isn’t maximum mentalizing; it’s accurate mentalizing. Seeing intentions where none exist can be just as disabling as missing them entirely.

Do Animals Have Theory of Mind?

The question of whether non-human animals can reason about mental states has been debated since the late 1970s. Chimpanzees are the most studied case, and the evidence is mixed in an interesting way. There is solid evidence from multiple experimental setups that chimps understand goals, intentions, perception, and knowledge in others. But despite several careful attempts, there’s currently no evidence that chimps understand false beliefs, the hallmark test of full-blown theory of mind. The best characterization is that chimpanzees operate with a perception-goal psychology: they know what others can see and what others want, but they don’t track beliefs that differ from reality.¹⁴

Some non-primate species show surprisingly sophisticated behavior. Ravens adjust their food-caching strategies based not just on whether a competitor is present right now, but on whether that competitor was present earlier and could have observed the hiding. They also seem to understand that physical barriers block another bird’s line of sight.¹⁵ Whether this reflects genuine mental-state attribution or a simpler behavioral rule is still debated, but the fact that ravens track what others have and haven’t seen, and change their behavior accordingly, is a striking example of at least proto-theory-of-mind in a bird.

The evolutionary picture suggests that theory of mind emerged gradually in response to increasingly complex social environments. Living in large, dynamic social groups creates pressure to predict what others will do, to form alliances and detect cheaters, and these demands may have driven the cognitive arms race that eventually produced human-level mentalizing.¹⁶ Computational modeling across seven primate species has explored this idea further, finding that the complexity of social networks aligns with the sophistication of social reasoning, a pattern consistent with the Machiavellian intelligence hypothesis.¹⁷

Culture Shapes the Sequence

Children everywhere develop theory of mind in a progressive sequence, where earlier insights lay the groundwork for later ones. But the order isn’t identical across cultures. A meta-analysis of theory of mind scale studies found that while the overall developmental pattern is universal, children growing up in more individualistic societies and those in more collectivist societies progress through slightly different sequences. The milestones are the same, but the order in which certain understandings click into place varies based on cultural learning environments.¹⁸

This makes intuitive sense. A child raised in a culture that emphasizes group harmony and social hierarchy gets more practice tracking obligations and social roles early on, while a child raised in a culture emphasizing individual autonomy might develop insight into personal desires and diverse beliefs earlier. The endpoint is similar: by school age, children in both kinds of cultures can pass the core battery of theory of mind tasks. But the path there reflects the social world the child is navigating.

Can AI Pass Theory of Mind Tests?

Large language models like GPT-4 have been tested on theory of mind tasks, and the results are revealing. These systems perform well on first-order false belief tasks, the Sally-Anne type scenario. But they falter on more complex reasoning, like second-order beliefs (“Sally thinks Anne thinks the marble is in the basket”) and recursive inferences that require tracking multiple nested mental states. Humans consistently outperform them on these harder tasks.¹⁹

This is a useful example because it highlights what theory of mind actually requires. An AI can learn statistical patterns in language that let it produce the right answer to a formulaic false belief question. But the richer, context-dependent, layered reasoning that humans do when tracking what multiple people know, believe, and feel about each other’s knowledge remains difficult to replicate. The fact that AI stumbles precisely where the reasoning gets recursive tells us something about how deeply theory of mind is woven into human cognition rather than being a surface-level trick.

In robotics, theory of mind has practical applications for trust. When a robot is equipped with a mature computational model of mental states, it can correctly distinguish reliable from unreliable information sources, accepting suggestions from the former and rejecting the latter. With an immature model, it fails this evaluation.²⁰ On the human side, people’s willingness to trust a robot after it makes a mistake depends partly on how much conscious experience they attribute to it. When people ascribe the robot greater levels of mental life, apologies and denials are more effective at repairing trust.²¹ We can’t help mentalizing, even with machines.

Reading Fiction as a Theory of Mind Workout

One of the more surprising findings in this area involves literary fiction. A series of experiments found that people who had just read literary fiction performed better on both cognitive and affective theory of mind tests compared to people who had read nonfiction, popular genre fiction, or nothing. The effect was temporary, more like a warm-up than permanent training, but it was consistent across five separate experiments.²²

The proposed explanation is that literary fiction, unlike more formulaic genres, presents characters whose motivations are ambiguous and whose inner lives have to be actively inferred. Genre fiction tends to rely on recognizable character types whose desires and intentions are spelled out. Literary fiction forces you to do the work of mentalizing, and that exercise primes the same capacities you use in real social encounters. Whether habitual literary reading produces lasting gains in theory of mind remains an open question, but the short-term boost suggests that the skill is at least somewhat trainable and responsive to practice, not simply a fixed trait you’re born with.

The Philosophical Debate That Won’t Quit

How do we actually do this? Two major theories have competed for decades. Theory-theory holds that we reason about other minds using an implicit set of rules and generalizations, like a folk psychology textbook we’ve internalized. Simulation theory argues instead that we understand others by mentally simulating their situation, running a kind of “what would I think and feel?” routine and projecting the result onto them.²³ Brain imaging and narrative comprehension studies point to the dorsomedial prefrontal cortex and the right temporo-parietal junction as key players when people track a story character’s mental states.²⁴

In practice, the answer is probably “both, depending on the situation.” Quick, automatic social judgments might rely more on simulation: you wince when you see someone stub their toe because your brain briefly models the experience. Slower, more deliberate reasoning about complex social scenarios, like figuring out why your boss sent that ambiguous email, might lean more on theory-like inference. The debate continues, but for a general audience, the takeaway is that your brain has multiple routes to the same destination, and which route it takes depends on the complexity and immediacy of the social situation you’re navigating.