Intentionality is the mind’s capacity to be directed at something, to be “about” something beyond itself. When you think about your lunch, fear a storm, or hope for a promotion, your mental state points outward at an object or situation. This aboutness is what philosophers, psychologists, and neuroscientists mean by intentionality, and it touches everything from how infants learn to read other people’s actions to how courts assign criminal blame. The concept is richer and stranger than it first appears, and it has become a flashpoint in debates over whether machines can genuinely think.
The Philosophical Root
The modern idea of intentionality traces to the 19th-century philosopher Franz Brentano, who argued that this directedness toward objects is what separates mental life from everything physical. His claim was bold: intentionality “is characteristic exclusively of mental phenomena. No physical phenomenon manifests anything like it.”1Cambridge University Press. Intentionality as the Mark of the Mental In other words, every mental state, whether a belief, a desire, a fear, or a memory, is about something. A rock is not about anything. A thermostat is not about the temperature in the way your annoyance is about the temperature. For Brentano, this was the dividing line between the mental and the non-mental: all and only mental phenomena exhibit intentionality.2Routledge Encyclopedia of Philosophy. Intentionality
Not everyone since Brentano has accepted this clean division. Some philosophers have argued that certain mental states, like raw pain, don’t seem to be “about” anything in the usual sense, while others have suggested that some non-mental systems, like a missile’s guidance computer, do seem to track targets in a way that looks like aboutness. But Brentano’s framing remains the starting point for almost every serious discussion. It set the terms of the debate that still plays out across cognitive science, AI research, and the philosophy of mind.
Dennett’s Intentional Stance
One of the most influential modern takes on intentionality came from the philosopher Daniel Dennett, who shifted the question from “what really has intentionality?” to “when is it useful to treat something as if it does?” His framework, the intentional stance, is a strategy for predicting and explaining behavior by attributing beliefs, desires, and goals to a system and then figuring out what it would be rational for that system to do given those attributed states.3Behavioral and Brain Sciences. Precis of the Intentional Stance You can adopt the intentional stance toward a chess computer, a dog, or your neighbor. If treating the system as a rational agent with beliefs and desires lets you predict what it will do next, the stance is working.
This is a pragmatic move rather than a metaphysical one. Dennett isn’t necessarily saying the chess computer “really” wants to protect its queen in the way you want your morning coffee. He’s saying that the intentional vocabulary (wants, believes, intends) is a powerful tool for prediction, and the question of whether the system “truly” has those states may be less important than whether attributing them works. This sidestep irritates some philosophers and delights others, but it has been enormously productive in cognitive science, where researchers routinely describe animal and infant behavior in intentional terms without committing to a deep claim about inner experience.
How the Brain Processes Intentions
Neuroscience has started to map intentionality onto specific brain regions, though the picture is still far from complete. One consistent finding is that the posterior parietal cortex plays a central role in the conscious experience of intending to move. Research suggests that our awareness of intending an action is not simply feedback from muscles and eyes telling us we moved; instead, it is tied to the initial intention itself, generated before the movement begins.4Trends in Cognitive Sciences. Intentionality, motor awareness, and the posterior parietal cortex You feel like you’re “about to reach for the cup” before your hand starts moving, and that feeling of agency lives in the parietal cortex.
Beyond your own intentions, your brain also works to decode the intentions of other people. A network of parietal and premotor areas contains neurons that fire both when you perform an action and when you watch someone else perform a similar action, a system sometimes called the mirror mechanism. This system has been linked to understanding other people’s actions and, more controversially, their intentions.5PubMed. Mirror neurons and mirror systems in monkeys and humans The mirror mechanism sits in the same circuits that organize your own goal-directed movements, which means the brain may partly understand what someone else is trying to do by simulating the action internally.6PubMed. Cortical mechanisms underlying the organization of goal-directed actions and mirror neuron-based action understanding
However, the enthusiasm around mirror neurons has been tempered. More recent analysis suggests that while mirror neurons are involved in perceiving actions at a basic level, there is no strong evidence they handle the kind of higher-level intention reading that lets you figure out why someone is doing what they’re doing, as opposed to what they’re doing.7PubMed. Understanding intentions from actions: Direct perception, inference, and the roles of mirror and mentalizing systems That deeper, more reflective reading of intentions seems to rely on a different set of brain areas, often called the mentalizing network, which includes parts of the prefrontal cortex and the temporal-parietal junction.
When researchers used brain imaging to see whether they could decode a person’s specific intentions from neural activity, they found that parietal, premotor, prefrontal, and cerebellar regions all carried information about what a person intended to do, with classification accuracy around 54 to 56 percent across these regions, above chance but modest.8PubMed Central. The effect of context and reason on the neural correlates of intentions The brain doesn’t seem to store intentions in one neat location; instead, intention processing is distributed, and the context and reasons behind an intention shape where and how the brain encodes it.
How Infants Learn to Read Intentions
One of the most active research areas in developmental psychology is when and how babies start treating other people as intentional agents rather than just moving objects. The evidence suggests a gradual process. By about eight months, infants recognize that a hand reaching for an object is goal-directed: they expect the hand to keep going for the same object, and they’re surprised when it suddenly grabs something else. By ten to twelve months, this understanding deepens. Infants at that age can recognize not just successful reaching but also failed attempts, understanding that someone who kept trying and failing to grasp a toy was still aiming for it.9PubMed Central. You can’t always get what you want: infants understand failed goal-directed actions
But this early understanding has limits that are easy to overstate. An influential 2012 study claimed that 11-month-olds could flexibly anticipate that a hand would reach for the same goal object even after that object swapped positions with another one. If true, it would mean infants that young already think in terms of abstract goals rather than concrete movement paths. A large multi-lab replication effort, however, involving over 200 infants across multiple experiments, failed to reproduce this finding. Infants consistently anticipated the hand’s movement path, not its goal, regardless of whether the agent was human or animated.10PubMed. Infants’ perception of goal-directed actions: A multi-lab replication reveals that infants anticipate paths and not goals The upshot is that while babies are clearly sensitive to goal-directed action by the end of their first year, the ability to flexibly track abstract goals across changing situations seems to develop later than some researchers initially believed.
Shared Intentionality and Human Culture
Individual intentionality, the ability to have beliefs and goals, is something humans share with other animals to varying degrees. But several researchers argue that what makes human cognition distinct is shared intentionality: the ability to form joint goals, share attention, and coordinate actions with others in ways that go beyond what any other species does. The idea is that uniquely human cultural achievements, from language to institutions, rest on the ability to do things together in a way that requires both partners to understand and care about the shared nature of the activity.11PubMed Central. Shared intentionality, reason-giving and the evolution of human culture
This extends even to something as simple as social bonding. Many mechanisms of bonding are common across primates: grooming, proximity, synchronized activity. But humans have developed bonding mechanisms that seem to depend on the triadic sharing of experience, situations where two people jointly attend to the same thing and know they are doing so. Walking together, having a conversation, watching a sunset side by side: these all involve a kind of mutual awareness that goes beyond being in the same place at the same time. A recent account proposes that these uniquely human bonding mechanisms evolved because of our species’ reliance on cooperative activities that both depend on and create shared experience.12PubMed Central. A Shared Intentionality Account of Uniquely Human Social Bonding
Do Other Animals Have Intentionality?
Great apes provide the most interesting comparison case. Chimpanzees, gorillas, bonobos, and orangutans use gestures deliberately and voluntarily, adjusting them based on who is watching and whether the audience is paying attention. If the recipient isn’t looking, an ape is more likely to use a contact gesture (a touch or a poke) rather than a visual one (a wave or a beckoning motion).13PubMed Central. Great ape gestures: intentional communication with a rich set of innate signals This audience sensitivity is a strong marker of intentional communication: the ape isn’t just emitting signals reflexively but is choosing which signal to send based on a representation of the other’s attentional state.
What apes seem to lack, or at least show in much weaker form, is the shared intentionality discussed above. Chimpanzees can cooperate to get food, but experiments suggest they don’t seem to form joint goals in the same way human children do. A two-year-old human, frustrated that a partner stopped cooperating, will try to re-engage the partner; a chimpanzee tends to just look for a different partner or give up. The gap isn’t in having goals or in reading others’ behavior. It’s in the capacity to build and sustain a genuinely shared project where both participants understand themselves as jointly committed.
When Intentionality Breaks Down
Several psychiatric and developmental conditions involve disruptions in how intentionality is processed. Autism spectrum disorder (ASD) is perhaps the most studied case. The theory of mind model of autism suggests that people with ASD have a deep difficulty understanding other minds, including others’ emotions, beliefs, and thoughts.14PubMed Central. Interventions based on the Theory of Mind cognitive model for autism spectrum disorder Early research demonstrated this vividly: when shown picture sequences requiring explanation in terms of people’s mental states, autistic children performed far worse than matched controls, and their narrations used causal and behavioral language but almost no mental-state language.15British Journal of Developmental Psychology. Mechanical, behavioural and Intentional understanding of picture stories in autistic children They could explain what happened in mechanical or behavioral terms but struggled to explain why someone did something in terms of what that person wanted or believed.
More fine-grained research has shown that the difficulty isn’t necessarily with understanding what someone is doing, but specifically with understanding why they are doing it in a particular context. In one study, children with ASD performed just as well as typically developing children when asked to identify what action was being performed and what tool was being used for, but their error rates jumped sharply on questions about why an action was directed at a specific location, a task that required reading the actor’s underlying intention.16PLOS ONE. Intention Understanding in Autism The deficit seems targeted at the interpretive layer that connects observed behavior to inferred mental states.
Schizophrenia presents a different kind of disruption. Here the issue is less about reading other people’s intentions and more about the sense of agency, the feeling that your own actions and their outcomes are yours. Clinical symptoms like delusions of control (feeling that an external force is moving your body) and thought insertion (feeling that thoughts are being placed in your mind) suggest that something has gone wrong with intention attribution from the inside. Experimental work confirms this but in a puzzling way: some studies find that people with schizophrenia under-attribute agency to themselves, consistent with the feeling that external forces are in control, while other studies find over-attribution, where people take credit for outcomes they didn’t cause.17PubMed. Sense of agency in schizophrenia: A reconciliation of conflicting findings through a theory-driven literature review One recent study found that a particular marker of agency, the tendency to perceive shorter time intervals between your action and its outcome when you feel like you caused it, was present in healthy controls but absent in people with schizophrenia.18Schizophrenia Bulletin. Dissecting Sense of Agency in Schizophrenia: A Predictive Coding Perspective The mechanism that normally glues your intentions to their outcomes appears disrupted.
Intentionality in the Courtroom
Legal systems around the world rely heavily on the concept of mens rea, the “guilty mind,” which asks whether a person intended to commit a crime or cause harm. In principle, the same physical act can be a serious crime or an innocent accident depending on the actor’s mental state. This makes the law deeply dependent on judgments about intentionality, and those judgments turn out to be more fraught than the legal framework assumes.
Research comparing ordinary people’s intuitive concept of intentionality with legal definitions has found a persistent mismatch. The folk concept of intentionality is structured and consistent: people evaluate whether someone acted intentionally by considering their beliefs, desires, awareness, and skill in a relatively orderly way. Legal definitions, by contrast, vary bewilderingly across jurisdictions and statutes, creating a gap between how jurors naturally think and what the law technically requires.19PubMed. Judging mens rea: the tension between folk concepts and legal concepts of intentionality
Even professional judges are not immune to biases in how they assign intentionality. A study surveying actual judges found that the worse the outcome of an action, the more likely judges were to say the actor acted intentionally, even when the description of the actor’s mental state was held constant.20PubMed. Mens rea ascription, expertise and outcome effects: Professional judges surveyed This outcome bias runs directly counter to the legal principle that mens rea should be evaluated based on what the defendant was thinking at the time, not on how badly things turned out. If a surgeon performs a risky operation with the best of intentions and the patient dies, the severity of the death shouldn’t retroactively make the surgeon’s decision seem more intentional, but psychologically, it does, even for trained professionals.
Culture Shapes How We Assign Intentions
Intentionality judgments aren’t universal. Research comparing participants in India and the United States has found that Indians attribute more intentionality than Americans for helpful side effects of actions, while the two groups don’t differ much on harmful or neutral ones.21Social Psychological and Personality Science. Intentional Sin and Accidental Virtue? Cultural Differences in Moral Systems Influence Perceived Intentionality In other words, if a CEO’s decision to boost profits also happens to help the environment, Indian participants are more likely than American ones to say the CEO helped the environment intentionally. This asymmetry seems linked to cultural systems of praise: in contexts where moral and religious frameworks emphasize the importance of recognizing and crediting good deeds, people are more inclined to read positive outcomes as intended.
Broader cross-cultural work has also found that the relationship between perceived intentionality and punishment varies with cultural orientation. In more individualist societies, perceptions of the actor’s agency (did they choose freely? did they know what they were doing?) strongly predict how much responsibility people assign. In more collectivist societies, the severity of the harm matters more than the actor’s internal states.22PubMed. A World of Blame to Go Around: Cross-Cultural Determinants of Responsibility and Punishment Judgments Neither approach is more “correct.” They reflect different moral frameworks for weighing the same ingredients, and both lead to systematic, predictable patterns of judgment rather than random disagreement.
Can Machines Have Intentionality?
The emergence of large language models has turned what was once a niche philosophical debate into an urgent practical question. When a chatbot says “I think you should try a different approach,” is there any sense in which it means what it says? Or is it producing statistically likely sequences of words without any “aboutness” behind them? The AI research community is actively divided on this, with the debate touching on whether these models understand language and the situations language describes in anything resembling a human sense.23PubMed Central. The debate over understanding in AI’s large language models
Some researchers have tried to settle the question by checking whether language models meet the criteria set by philosophical theories of mental content: if a theory says that a system has genuine beliefs when certain conditions are met, do language models meet those conditions? Others argue this is the wrong approach entirely. Instead of asking whether machines have mental states in the philosophical sense, they suggest we should ask whether machines meet the criteria our best theories give for meaningful language use, which is a different and possibly more tractable question.24Synthese. Large language models and linguistic intentionality The distinction matters because it’s possible for a system to use words meaningfully, in the sense that its outputs are systematically connected to the world in the right ways, without having conscious inner experience. Whether that counts as “real” intentionality depends on which theory you’re working with, and the philosophical community hasn’t settled that question.
When We Project Intentions Onto Robots
Whatever the truth about machine intentionality, humans are remarkably willing to attribute it. Studies in human-robot interaction reveal that people readily see robots as having intentions, but the perception is fragile and context-dependent. In one experiment, participants who had short interactions with a robot actually increased their attribution of intentionality to it afterward, while those who spent longer with the robot showed little change or even a decrease.25PubMed Central. Repetitive Robot Behavior Impacts Perception of Intentionality and Gaze-Related Attentional Orienting Repetitive, predictable behavior seemed to erode the illusion. As long as the robot was novel and its behavior could plausibly be interpreted as purposeful, people were happy to see it as having goals. Once the pattern became obvious, the spell weakened.
The consequences of attributing intentionality to machines go beyond philosophical curiosity. When people perceive a robot as an intentional agent, it actually reduces their own felt sense of agency over shared outcomes. In one set of experiments, participants who performed a task alongside a robot that they perceived as capable of acting intentionally reported feeling less ownership over the results than participants who performed the same task alongside a passive mechanical device with identical physical effects. A human co-actor produced the same reduction in felt agency, suggesting that once a machine crosses the threshold into perceived intentionality, it starts to occupy the same psychological slot as another person.26PubMed. Attribution of intentional agency towards robots reduces one’s own sense of agency As robots and AI systems become more common in workplaces and homes, these effects on human self-perception may become increasingly relevant. The question isn’t only whether machines have intentionality. It’s also what happens to us when we start treating them as though they do.