Description

“Simulation theory” and “theory theory.” Two different prominent theories that have been proposed to explain the basic mechanism underlying theory of mind abilities are usually referred to as “simulation theory” and “theory theory” (101). According to simulation theory, theory of mind capacity is based on taking someone else’s perspective and projecting one’s own attitude onto someone else. By contrast, according to theory theory, theory of mind capacity is based on a theory like conceptual framework for understanding others’ actions based on mental states, acquired during the individual’s ontogenetic development.

Neurophysiological evidence relevant to this debate was provided by Gallese and colleagues, who demonstrated a mirror neuron system in macaques (41). Mirror neurons are premotor neurons that fire when the monkey performs object-directed actions such as grasping, tearing, manipulating, and holding, but also when the animal observes somebody else, either a conspecific or a human experimenter, performing the same class of actions. The discovery of mirror neurons provided a possible mechanism for a simulation theory account of theory of mind (42). However, there is substantial debate about the role of mirror neurons in understanding others’ actions and minds (51).

Some theories consider different theory of mind capacities to be subserved by different cognitive mechanisms. For example, two-systems theories argue that a fast and efficient cognitive system, possibly supported by mirror neurons, underlies the kind of quick theory of mind judgments used, for example, in action anticipation or communication. A more cognitively demanding process is thought to be used for reasoning about mental states in verbal tasks or when observing social scenarios (05). Researchers have also posited different processes subserving affective (reasoning about emotions) and cognitive (reasoning about thoughts) theory of mind (30).

Neurophysiological bases. Findings from imaging and lesion studies indicate that theory of mind is mediated by a widely distributed neural system (93).

Functional imaging has been important to isolate brain regions involved in the theory of mind network, with some consensus of a laterality effect to the right hemisphere (06). Different brain regions seem to support different aspects of social cognition and mentalizing. For instance, many studies have shown that the right temporoparietal junction (rTPJ) is selectively active when a subject thinks about another person’s mental states, but not other social information about the person or nonmental representations (86; 87). The right dorsolateral prefrontal cortex has also been associated with theory of mind task performance in lesion studies (29). In contrast, the superior temporal sulcus has often been associated with reasoning about intentional actions (85; 110).

One study supports the importance of the right temporoparietal junction, as well as the role of connectivity between regions in the theory of mind network (58). The authors compared younger adults to older adults doing a theory of mind task while undergoing fMRI. They found that connectivity between the right temporoparietal junction and the right temporal pole mediated the age-related decline in theory of mind performance. Recent research has attempted to more carefully pinpoint the role of the rTPG in mentalizing, finding that the rTPJ does not show differentiation between bottom-up (inferring mental states from motion kinematics) relative to top-down (belief reasoning) theory of mind tasks, whereas other regions do. The authors suggest that the rTPJ may support the representation of mental states, whereas other regions such as the lTPJ and IFG distinguish the level of top-down cognitive function needed for the specific task (44).

The contributions of specific neurotransmitter systems to theory of mind have been discussed. In his neurochemical hypothesis of theory of mind, Abu-Akel proposes that dopaminergic and serotoninergic dysfunction may impair theory of mind (01). Some evidence for this hypothesis comes from a study investigating the role of the 5-HT1A-R-1019 functional polymorphism in the theory of mind of patients with schizophrenia. Patients with the CC genotype of 5-HT1A-R performed better than patients with other genotypes in a picture sequencing theory of mind task (18). However, the study did not directly test the neurochemical mechanisms mediating this difference.

Relation between theory of mind and other cognitive abilities.

Executive function. Whether theory of mind can be characterized by either a domain-specific cognitive process, or inextricably linked to domain-general cognitive abilities, has been debated. Some argue that basic theory of mind abilities result from an encapsulated cognitive module that operates separately from general executive function. In support this, theory of mind and executive function deficits can be dissociated in patients with brain lesions (83), and there is some evidence that young infants with limited executive function abilities can understand low-demand false belief tasks (75).

Nonetheless, even if theory of mind and executive functions abilities are distinct, they are often found to be related to each other. It may be that executive dysfunctions prevent one’s ability to express understanding of others’ mental states. For instance, 3-year-old children often fail false belief tasks that require their inhibiting attention to the true location of an object (thus, using executive function), and verbalize correctly where a cartoon character with a false belief will search (107). Yet if children are not required to respond verbally, or if a simpler task is given, many more children can pass the task (90; 84). Another possibility is that executive function is important for acquiring mental state concepts and not only expressing them (24).

Empathy. Theory of mind and empathy—the ability to infer and share the emotional experiences of another—seem to be distinct but related processes. Some authors have argued for a distinction between “cognitive” theory of mind and “affective” theory of mind, based on behavioral and neuroimaging task dissociations (92; 30). Affective theory of mind plays a role in the cognitive aspect of empathy, inferring emotional states, whereas another component of empathy is in sharing the feelings of others, sometimes termed “emotional contagion.” Neuroimaging studies suggest that theory of mind and empathy engage different neural processes, but that there are also combined processes that engage both systems at once, and suggest a hierarchical model of understanding the relationship between them (89).

Language and communication. Theory of mind is clearly important to understand the pragmatics of communication, and many authors have investigated perspective-taking in this domain (55). Some argue that a special subsystem of theory of mind is needed specifically for making sense of communicative interactions that require rapid online monitoring of others’ mental states (95).

Some researchers believe that theory of mind is fundamentally tied to language use. Substantial evidence suggests that language development and theory of mind abilities are related, and some argue that specific advances in language development support important changes in theory of mind reasoning (70). The idea that language development promotes theory of mind skills is also supported by findings that deaf children of hearing parents—whose environment can be communicatively impoverished—tend to show theory of mind delays compared to both hearing children of hearing parents and deaf children of deaf parents (78). Receiving early hearing provisions can accelerate children’s theory of mind development by way of language acquisition (115).

Clinical applications

Assessment of theory of mind. A range of verbal and nonverbal tasks has been used to assess theory of mind. Five of the most common tests used are the false belief task (and variants thereof), the character intention task, the faux pas recognition test, the strange stories task, and the reading the mind in the eyes test.

False belief test. In a typical task, subjects hear a story about a character who has a false belief and are asked to predict the character’s behavior. For instance, Sally places her ball in the toy box, then leaves the room. Another girl named Anne sneakily moves the ball to a cupboard in Sally’s absence. When Sally returns, where will she look for her ball? If subjects understand that Sally’s belief (the ball is in the toy box) can be different from their own (the ball is really in the cupboard), they should expect Sally to act on the basis of her false belief and search in the toy box (108; 07). There are variants of the false belief task that measure more complex theory of mind abilities, for instance, an ability to appreciate second-order beliefs. In a second-order false belief task, one character has a false belief about another character’s belief. For instance, after Anne moves the ball to the cupboard unbeknownst to Sally, Dan finds the ball and takes it to the park. When both Sally and Anne return to the room, where should Anne expect Sally to look for the ball? In this situation, subjects must recognize that Anne believes that Sally believes the ball is in the toy box.

Character intention task. The task consists of short comic strips, each featuring a character with a specific intention. Each strip consists of three pictures, and subjects are offered a choice of three possible conclusions to each scenario in the form of answer cards. One answer card is a logical conclusion that matches the character’s intention, and two answer cards are distracters with no logical link to the intentional context of the scenario (21).

Faux pas recognition test. After reading a story containing a faux pas, subjects are asked whether anyone from the story said anything that was awkward or that they shouldn’t have said (97; 45).

Strange stories task. This task assesses subjects’ advanced theory of mind by asking them to interpret nonliteral statements, for example, in the context of pretense or irony (46).

Reading the mind in the eyes test (RMET). This test involves judging from the expressions of another person’s eyes what that other person might be thinking or feeling (08). For each set of eyes, subjects have to choose the word that best describes what the person in the picture is thinking or feeling. The RMET is a commonly used test for theory of mind deficits in adults; however, given evidence of validity limitations in the RMET, researchers are starting to use batteries of tasks more often or to consider the appropriate contexts in which the RMET will be most accurate and practically useful (77; 52).

These tasks are simply different ways to assess “theory of mind” (as operationalized by the researcher using the test and may not strictly tap a single underlying ability (Quesque and Rossetti 2020). For instance, false belief tasks and the character intention task might assess a more cognitive type of theory of mind skill, whereas the faux pas recognition test and the reading the mind in the eyes test could assess a more affective theory of mind component, consistent with evidence that cognitive and affective theory of mind (or empathy) are partially dissociable (89). However, some have argued, based on dissociations between emotion and mental state recognition in disorders such as autism spectrum disorder and psychopathy, that emotion recognition as measured in the RMET is entirely distinct from theory of mind (48; 73). It is also important to be aware of recent reviews and metaanalyses suggesting that the reliability and validity of many theory of mind measures are questionable and their psychometric properties often not fully established (114).

In addition, these tests are often used for different purposes with different populations. Standard false belief tasks are appropriate to assess whether young children have theory of mind deficits because most children pass these tasks by the age of 4. In contrast, comprehension of social faux pas and nonliteral language are more complex; therefore, the faux pas and strange stories tasks are often used to assess theory of mind in adults, adolescents, or older children with sufficient verbal skills. The RMET is also typically used with adults or older children. Because many theory of mind tasks involve abilities beyond attributing mental states (eg, inhibitory control, verbal ability, knowledge of social norms), it is important when assessing theory of mind deficits to choose a task that is suitably engaging and developmentally appropriate.

Implication of theory of mind in psychiatric and neurologic disorders. Many authors have investigated theory of mind deficits in various psychiatric and neurologic diseases. Theory of mind dysfunctions are implicated in many clinical disorders and have especially been described in autism spectrum disorders, schizophrenia, mood disorders, frontal lobe damage, stroke, frontotemporal dementia, basal ganglia disorders, and severe traumatic brain injury.

Autism. It has long been thought that autistic individuals experience difficulties in engaging in theory of mind reasoning (07; 40). Many researchers have suggested that a core theory of mind deficit underlies the social and communicative difficulties experienced by individuals with autism (91). Indeed, theory of mind skills in autistic children are reliably related to levels of disorder severity (56) independently from executive function skills (59). Developmental psychology research also finds correlations between early theory of mind skills and social interaction skills later in childhood (27). Autistic adults tend to perform more poorly on theory of mind tasks than neurotypical controls, and brain imaging evidence (showing increased activity in the STG during a ToM task) suggests that they may spend more time and effort on intention processing (28). However, theories have also been proposed that do not put theory of mind deficits at the core of autism, for example, that the reduced social motivation (26; 23) or sensory deficits (09) of individuals with autism spectrum disorder in early life have downstream developmental effects on theory of mind and social cognition more broadly. A large meta-analysis based on 881 effect sizes found that associations between social functioning and theory of mind were significant but quite small in individuals with autism as well as typically developing individuals, suggesting that theory of mind may not be as central to autism as some have supposed (19). Research is beginning to focus more on identifying specific theory of mind challenges in autistic populations in order to design more targeted interventions, rather than assuming all aspects of theory of mind are impaired. For instance, one study found that 8- to 16-year-old autistic children engaged in similar rates of positive conversational theory of mind behaviors (eg, speech suggesting consideration of their communication partner’s mental state) to typically developing controls (04). However, the autistic children engaged in higher rates of negative conversational theory of mind (eg, providing too much or too little information), and this tendency was associated with deficits in other theory of mind measures.

Schizophrenia. The theory of mind deficit in schizophrenia was first hypothesized by Frith and Corcoran, who presented arguments supporting the idea that some paranoid symptoms and behavioral signs could be a result of difficulties in inferring the intentions and beliefs of others (39). Patients with schizophrenia show delays in the speed of intention attribution (82) and less mentalizing in tasks measuring implicit as well as explicit theory of mind (64; 32). Individuals with first-episode psychosis also show poorer performance than control groups on theory of mind tasks such as the faux pas test and joke comprehension (63; 53). In one study, false belief test scores were correlated with delusion in schizophrenic patients (76), providing evidence for Frith and Corcoran’s initial ideal that theory of mind may be important for core symptoms of schizophrenia. Interestingly, theory of mind deficits in schizophrenia not only make it more difficult for patients to access the minds of others, but also to make themselves understood. Achim and colleagues found that theory of mind scores in schizophrenic patients were correlated with their interaction partners’ ratings of how easy it was to understand them (02).

Importantly, there is considerable heterogeneity in theory of mind skills in individuals with schizophrenia, and these skills are related with different variables and outcomes in this population than in healthy controls (10). However, promising interventions targeting theory of mind in schizophrenia have already been developed. A metaanalysis of 14 controlled trials found that observation and imitation of social emotions led to improved theory of mind performance relative to cognitive therapy on its own (111). Metacognitive training, which targets cognitive biases in psychosis, has small but positive effects on theory of mind skills and general social cognition in individuals with psychosis (54). A new virtual reality based therapy in which patients with schizophrenia interact with an avatar, targeting interactions designed to improve theory of mind schools, has also shown positive effects on both theory of mind after intervention as well as symptoms (100). A mindfulness-based intervention showed greater improvements in scores on the RMET in patients diagnosed with psychotic disorder, when added to a typical integrated rehabilitation treatment (66). Importantly, theory of mind scores on some tasks have been found to be related to functional recovery in patients with schizophrenia, especially in terms of social skills and collaboration, adding to the promise of theory of mind based intervention (03).

Mood disorders. Theory of mind deficits are often associated with anxiety disorders characterized by fear of others’ negative evaluations and judgments. In one study, individuals with social anxiety disorders showed impaired cognitive but not emotional perspective-taking compared to obsessive-compulsive and control groups, even though they reasoned about other individuals in a third-party context (22). Other studies have also found theory of mind deficits in social anxiety disorder (50; 104), though it is important to note that these deficits are typically inaccurate mentalizing or excessive mentalizing (ie, in contrast to autism where the tendency to mentalize is often reduced). Some work has even shown that individuals with generalized anxiety disorder are more accurate than controls in some theory of mind tasks when induced to feel worried, consistent with the notion that anxiety can increase and sometimes even improve mentalizing (117).

In other studies, moderate effects of poor theory of mind performance have also been found in patients with major depressive disorder (118; 72). However, the relationship between depression and impaired theory of mind has been inconsistent (12; 38; 37).

Psychopathy. Early studies suggested that individuals with psychopathy typically perform as well as control subjects on tasks that measure explicit theory of mind, including second-order theory of mind tasks and more complex social reasoning tasks like the faux pas test (13; 35). However, more recent work has found that psychopathic adults perform worse than control subjects in a task that measures automatic perspective-taking (36). These individuals may be able to strategically consider others’ mental states but lack the tendency of healthy adults to immediately consider what others think and feel. Consistent with this, a metaanalysis of 42 studies suggests that psychopathic traits are associated with impaired performance in both cognitive and affective theory of mind tasks (94).

Personality disorders. Individuals with borderline personality disorder have been shown to make more errors in theory of mind tasks, in addition to showing increased biases to over- or under-mentalize relative to healthy controls (16; 43).

Brain trauma. Changes in the social behavior of individuals following severe traumatic brain injury have long been noted. Patients with severe traumatic brain injury have been shown to be impaired in a range of theory of mind tasks (71). A meta-analysis found that theory of mind performance was impaired in patients with traumatic brain injury relative to healthy controls, and performance in theory of mind tasks was positively related to functional outcomes in patients (65).

Impaired social cognition has long been recognized to be common after frontal lobe damage. Many patients with frontal lobe damage exhibit impaired performance in theory of mind tasks despite only minor impairment on standard neuropsychological tests, including those held to be sensitive to frontal lobe dysfunction (83). Research suggests that impairments in executive function are at least partially responsible for mentalizing deficits in patients with prefrontal cortex damage (112).

Among the hypothesized causes of pragmatic and communication impairments in people with damage to the right cerebral hemisphere is an underlying impairment in theory of mind (105). Adult patients with right-hemispheric stroke are selectively impaired in tasks requiring mental state attribution compared to controls and patients with left-hemispheric stroke (47; 113). These impairments support the idea that some of the mental processes involved in theory of mind may be localized in the right hemisphere (86).

Children with mild traumatic brain injuries, who often display social impairments post-injury, score more poorly on tests of emotional and cognitive perspective taking relative to typically developing children (11). A study showed that psychosocial adjustment in children with severe (but not moderate) traumatic brain injury was mediated by theory of mind deficits indicated by both poor scores in theory of mind tasks as well as reduced volume in areas of the right hemisphere typically associated with mentalizing (57). Executive function did not mediate this relationship.

Neurodegenerative disorders. Some research indicates theory of mind deficits in frontotemporal dementia and Alzheimer disease (34; 98), with empathy deficits specific and core to frontotemporal dementia (33). Theory of mind deficits as measured by the RMET were indeed more discriminative between frontotemporal dementia patients and controls than executive function tests (88).

Several lines of research suggest that theory of mind deficits may occur in degenerative basal ganglia disorders (14), as well as Parkinson disease (15). These deficits are largely independent from other cognitive impairments, depressive symptoms, and motor impairment. Parkinson disease patients can be impaired in recognizing both mental states (faux pas test) and emotions (reading the mind in the eyes test) (17). A new metaanalysis suggests that theory of mind deficits as measured by the RMET were consistent across patients with neurodegenerative conditions including Alzheimer, Parkinson, and Lewy body dementia, as well as multiple sclerosis (96).

Differential diagnosis. Theory of mind is a complex, multifaceted cognitive ability, and deficits in theory of mind are seen in a wide range of developmental, neurologic, and psychiatric conditions (31). The tools for evaluating theory of mind need more standardization to be used effectively in both diagnostic and research practice (20). The term “theory of mind” needs to be better defined or broken up into meaningful subcategories; and the clinical significance of specific theory of mind deficits in different disorders needs more investigation. As long as “theory of mind” is used to describe any ability to recognize any mental states in others, many disorders will resemble disorders of theory of mind.

Accordingly, it is very difficult to develop a differential diagnosis review. However, some studies have investigated the specificity of theory of mind deficits for a particular disorder, by controlling for other cognitive deficits. For instance, although theory of mind deficits have been found in patients with chronic depression, at least in one study, these deficits were more likely to result from general cognitive impairments (118). In contrast, a study that examined impaired theory of mind in schizophrenia patients with persecutory delusions found that an intention-reading task predicted delusion variance even when controlling for other cognitive factors (69). A third study found that patients with bipolar disorder showed impairments on a theory of mind task that were only partially accounted for by executive functions (109). Examining the associations between theory of mind impairments and other cognitive impairments might help to develop interventions that would be specific for theory of mind deficits.

Another productive avenue for research is to compare different groups of patients on similar theory of mind tasks. One such study found that theory of mind was impaired in three patient groups (schizophrenia, bipolar, major depressive disorder) when compared to matched healthy controls, but schizophrenic patients were more impaired than bipolar disorder patients, and the pattern of theory of mind impairments differed across the patient groups (102). Recent reviews of the literature on theory of mind deficits in patient populations suggest that these deficits are common in depression, bipolar disorder, and schizophrenia, increasing in severity along the affective-psychotic spectrum (99), and that schizophrenic patients have a similar extent of theory of mind impairment to patients with autism spectrum disorder (74). These findings have important implications for developing interventions that might target theory of mind across different clinical conditions.

In conclusion, although most researchers consider theory of mind impairment to partly underlies the behavioral disturbances occurring after several neurologic disorders, further studies are needed to determine which disorders involve core deficits in theory of mind. Discussion of putative clinical interventions is still premature, given the present state of knowledge about the social difficulties that patients with theory of mind deficits might experience in daily life.