Are there theory of mind regions in the brain

Are there distinct brain regions for ToM? This chapter reviews evidence suggesting that key regions support social cognition, thinking about people medial prefrontal cortex [MPFC] , and ToM, or thinking about thoughts right temporoparietal junction [RTPJ].

Evidence is presented showing that activity in these regions is not modulated by first-person experience or similarity between self and target. The chapter concludes with questions for the next decade of cognitive neuroscientific research regarding ToM.

Keywords: Theory of Mind ToM , social cognition , medial prefrontal cortex , right temporoparietal junction , simulation. Access to the complete content on Oxford Handbooks Online requires a subscription or purchase.

Public users are able to search the site and view the abstracts and keywords for each book and chapter without a subscription.

Please subscribe or login to access full text content. If you have purchased a print title that contains an access token, please see the token for information about how to register your code. For questions on access or troubleshooting, please check our FAQs , and if you can''t find the answer there, please contact us. All Rights Reserved.

During Stage V 12—18 months , the children can retrieve an object when it is hidden several times within his or her view. In summary, when an object was hidden in location A and then hidden in location B, the children would try to find the object in location A during Stage IV and would try to find it in location B during Stage V.

With similar principles, here we enable all of the robots in the experiment acquire the ability of learning object permanence from their own self experiences, and in our model, ACC acts as a central role to realize this cognitive function Gallagher and Frith, Figure 5 shows the visual inputs of the robots in this process. Figure 5. Object permanence learning from self-experience. The robot cannot perceive the object in its visual field, hence cannot find it.

It is similar to the early stages in Piaget's Stages of Object Permanence. As in the experiment of Senju et al. By this way, the participant robot could understand that the actor has the same cognitive ability e. In the final test, the opaque blindfold group and the transparent blindfold group are tested with the same process as shown in Figure 6. Figure 6. Visual inputs of participant robot in the test stage.

The middle screen in the a , and the remaining figures are the visual inputs of the participant robot. The robot learns the visual access of turning around from self-experience. The diversity of belief is caused by different blindfolds in the Opaque-and-Transparent Blindfold Test, and in Turn Around Test, it is caused by the behavior of turn around. The visual inputs of the participant robot are shown in Figure 7.

Figure 7. Turn Around Test. An object ladybird is put on the left black rectangle and hidden it in the yellow box firstly. In the false belief condition, the object was moved to the other box when the actor robot turned around Movie S6. And in true belief task, the actor robot did not turn around when the object was moved to the other box Movie S7. The ability for the ToM comes with individual development process Grosse Wiesmann et al. Grosse Wiesmann et al. But their research focused on the 3- and 4-year-old children in the explicit false-belief tasks, and did not include younger infants who cannot pass the implicit false-belief task.

They did not test whether this finding is also associated with an implicit task, because of the difficulties in performing MRI with toddlers. Although the developmental neural basis for the implicit false-belief task is still not very clear, we hypothesize that the developmental process in implicit false belief understanding is relevant with explicit one, and will also be associated with the maturation of correlate brain areas and their connections.

We aim to test this hypothesis by our computation model, and apply it to the Brain-ToM model that we developed for machine intelligence. The maturation of correlate brain areas could be regarded as calculation capability in our model, and the calculation capability increases with the maturation of brain areas.

The calculation capability in this model is proportional to the number of neurons in the hidden layer. The maturation of the connection between brain areas is critical for information transmission and information integration, especially inhibitory connection and control. The inhibitory control is generally considered as a key mechanism in false-belief task Leslie and Polizzi, ; Scott and Baillargeon, , and we think that the maturation of connections between IFG and TPJ, IFG and vmPFC are the neural basis of self-perspective inhibition and self-belief inhibition, respectively.

In this section, we present the results of our model. Besides, we analyze the temporal and spatial activation of different brain areas during different tasks, the effect of self-experience, maturation of correlate brain areas e.

The opaque blindfold group can be regarded as the false belief condition, as the actor robot's belief is inconsistent with the representations of reality. The transparent blindfold group can be regarded as the true belief condition, as the actor robot's belief is consistent with the representations of reality. When asking the participant robot the upper two questions, for both of them, the participant robot will point to the green box on the right side Movie S5. We repeat this experiment 20 times, the robot could pass the task every time, and we calculate the mean value and standard deviation of time consumption in different brain areas.

The time consumption of false belief reasoning And the time consumption of reasoning about other's belief In the process of reasoning about self-belief, the time consumption of false belief condition In traditional true belief task, the time consumption will be shorter.

The perception in TPJ is identical, hence the IFG will not be activated in this task and the time consumption is reduced. Here we provide the time consumption in different tasks. Based on the belief about the object location, the task can be divided into reasoning about actor robot's false-belief task other-incongruent condition and true belief task other-congruent condition , reasoning about participant robot's own belief task which contains self-incongruent condition self-belief is divergent from other's and self-congruent condition self-belief is corresponding with other's.

Figure 8 shows the temporal and spatial activation of different brain areas during different tasks. This process only contains the perception conflict stage as shown in Figure 6d and motion response stage which have critical differences in different tasks. So we select ms to show the process and the difference in different tasks. Reasoning about other's belief in the transparent group can be regarded as true belief task, but it must be noted that, the information from self-perspective and other-perspective are identical in tradition true belief task, while they conflict with each other in this task which is originally from Senju et al.

Figure 8. The temporal and spatial activation of different brain areas during different tasks. In the process of TPJ activation deciding the output sequence of self and other-relevant stimuli in other-incongruent condition red line and other-congruent condition green line , IFG is activated to inhibit self-perspective.

In the process of mPFC activation deciding belief for motor response in other-incongruent condition, IFG is activated to inhibit self-belief. In the tasks of self-incongruent condition black line and self-congruent condition blue line , the IFG is only activated in the process of encoding action goals. To show the function of IFG easily, we make the arrow mark in the figure.

Our focus is the activation sequence of brain areas and reaction time in the various task, such as the other-incongruent condition spends more time than self-congruent condition rather than the numerical value of time consumption, these result is consistent with the functional neuroimaging studies in Mossad et al.

In the process of inferring visual access which corresponds to Figure 6d , even though the visual inputs in both groups are identical, the output is different when the participant robot infer other's visual access with different self-experience. In other words, when inferring visual access of another person by self-experience, the opaque blindfold group will know the actor cannot see the moving object, and the transparent blindfold group will know the actor can.

When inferring self visual access, the perception of visual inputs and the result of precuneus are identical in both groups. In the process of inferring other's visual access, the IFG will not be activated when self-perspective and other-perspective is identical, as shown in Figures 6b,c,f.

If the self-perspective and other-perspective are in conflict with each other in the process of reasoning about other's belief, the IFG will be activated to inhibit the information of self-perspective, as shown in Figures 6d,e. In the process of motor response in reasoning about other's belief, the IFG will inhibit self-belief if the beliefs are conflictive.

In addition, we also test the effect of blindfold position in this task. Both of the groups can infer the visual access of actor robot correctly, and conclude that the actor robot could see the object move to the right side. And self-belief is corresponding to other's belief in both groups. This additional test could prove that the actor robot does not use low-level features such as whether the blindfold exists when inferring other's mental state, and it also proves the effect of the bodily model in this task.

In the false belief condition, the actor robot's belief is inconsistent with the representations of reality. In the true belief condition, the actor robot's belief is consistent with the representations of reality. When asking the participant robot the upper two questions, for both of them, the participant robot will point to the green box on the right side Movie S7. We repeat this experiment 20 times, the robot could pass the task every time. The mechanism of Turn Around Test is similar to the Opaque-and-Transparent Blindfold Test, the only difference is that the self-perspective and other-perspective are identical in true belief task of Turn Around Test.

So IFG was not activated in this stage. The maturation of correlate brain areas will be regarded as calculation capability in our model, and the calculation capability increases with the maturation of brain areas. As indicated in Myowa-Yamakoshi et al. We think that with the maturation of correlate brain areas as well as their connections, more neurons and synaptic connections will be included in the task processing.

The inhibitory control is generally considered as a key mechanism in false-belief task Leslie and Polizzi, ; Scott and Baillargeon, , and we think that the maturation of connections between IFG and TPJ, IFG and vmPFC are the neural basis of self-perspective inhibition and self-belief inhibition respectively.

As shown in Figure 9 , the inhibitory control uses inhibitory neurons and temporary neurons which store information temporarily for the selection of correct output information of TPJ or mPFC. The information of inhibitory neurons in TPJ is identical with the information of other-relevant stimuli in pSTS, and the information of inhibitory neurons in mPFC is identical with the information of other's belief in dmPFC.

The temporary neurons in TPJ receive self-relevant information from STS, and the temporary neurons' information is identical with the information of self-relevant stimuli in IPL. Then we test the effect of these connections in the false-belief task, and observe that the different maturation of connections leads to different permanence in the task.

These connections will not influence the process of reasoning about self-belief. Figure 9. The effect of IFG connection in the false-belief task. The solid circles indicate that the neurons have been activated. The behavior of the participant robot to predict the actor robot's action is the same as the action caused by self-belief, so it failed in the false-belief task.

The participant robot also failed in this task because the result of inferring other's visual access is wrong which causes by the immature connection between IFG and TPJ. The participant robot could infer other's belief correctly, but it cannot inhibit the effect of self-belief without IFG. In this section, we will discuss the characteristics of the model, the reasons why robot experiments and cognitive experiments are not completely consistent, and the possible mechanisms of why toddlers fail in high inhibition tasks.

Compared to the previous models which we introduce in the related works, our model explores and is fundamentally based on the role of self-experience. In our model, robots learn to understand object permanence and visual access of blindfold or turn around from self-experience, then use it to infer other's belief and predict their actions.

All of the participant robots learn the ability of understanding object permanence from the same experience. In the Opaque-and-Transparent Blindfold Test, they are divided into opaque blindfold group and transparent blindfold group.

Even though the visual inputs of both groups in the test stage are identical, the different experience with an opaque blindfold or transparent blindfold leads to different performances. In Turn Around Test, the different behaviors of the actor robot in the test stage result in different performances. Compared with the recently published work from Patacchiola and Cangelosi : 1 Our model is based on spiking neural networks, and just uses the STDP to successfully reproduce the complex cognitive function of ToM, hence more biological plausible.

The two studies have complementary contributions to the ToM models through bio-inspired mechanisms. Through the integration of biological inspirations and computational modeling, we suggest that the self-experience, maturation of correlated brain areas e. As indicated in Scott and Baillargeon , the false belief tasks contain spontaneous-response and elicited-response tasks that belong to the implicit task and explicit task, respectively.

The difference in spontaneous-response and elicited-response tasks is that the former investigates the capacity of false belief understanding by spontaneous behavior such as anticipatory-looking, preferential-looking, etc with a non-verbal task, and the latter investigates this capacity by answering direct questions that predict agent's behavior who has a false belief with the verbal task. Children can pass the spontaneous-response task before 2 years old, but they can not pass the elicited-response tasks until about 4 years old.

Our tasks on robots are not completely consisted with Senju et al. Both of them used spontaneous-response to test the infants on whether they can pass the task.

And in the test trial, they removed the object from the scene to make infants pass the task easier. In our task, we determine whether the robot can pass the task by detecting the direction of the finger which makes the results more intuitive.

And in the test trial, we move the object to the other box. Setoh et al. Toddlers could pass the elicited-intervention and low inhibition task removing object from the scene which is described by language and picture, but would fail in high inhibition task moving the object to another box. They thought the reason why toddlers failed in high inhibition task is that toddlers cannot inhibit the response of the actual location of the object.

We suppose that the core mechanism of belief reasoning is identical in both tasks, and the only difference should be in the process of motor response. In the motor response process of the elicited-intervention task, it may use the brain areas which control the hand movement.

And in spontaneous-response task, it may use the brain areas which control the eye movement. The main reason why we use high inhibition task to replace low inhibition task is that the behavior of the robots in the true belief task is more intuitive to be understood for the high inhibition task, the robot can point to the position of the object, while for the low inhibition task, the objects are removed outside the scene, and the robot cannot point to their positions.

And we suppose that the reason why toddlers failed in high inhibition task should be related to the lack of motor response ability rather than ToM e. In the low inhibition task, the participant has no idea about the object's location and one of the two candidate motor responses are activated, so the participant robot can succeed in this task.

And in Samson et al. Here we proposed a Brain-ToM model based on existing biological findings of ToM, and this model shows its relevance to ToM of human from the mechanism and behavior perspectives.

In summary, we propose a Brain-ToM model to enable machines to acquire the ability of ToM through learning and inferring based on self-experience. We validate the model by deploying it on humanoid robots. Our model successfully enabled the robot to pass the false-belief task, which is a classical task designed to understand the nature and mechanisms of ToM from Cognitive Psychology.

The model and its application on robots show that current understanding on the mechanisms of the ToM can be computationally unified into a consistent framework and enable the robots to be equipped with the initial cognitive ability of ToM.

YZe conceived the initial idea. YZh and YZe designed the model, carried out, and analyzed the experiments. FZ contributed to the inhibitory control mechanism.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Movie S1. Visual access learning of blindfold from self-experience opaque blindfold group.

This process enables the participant robot understand that other robots wearing a blindfold cannot see. Movie S2. Visual access learning of blindfold from self-experience transparent blindfold group. This process enables the participant robot understand that other robots wearing a blindfold can see.

Movie S3. Familiarization phase. This process enables the participant robot understand the goal of the actor robot. Movie S4. Opaque-and-transparent blindfold test opaque blindfold group. The details in opaque blindfold group of blindfold test. Movie S5. Opaque-and-transparent blindfold test transparent blindfold group.

The details in transparent blindfold group of blindfold test. Movie S6. Turn around test false-belief task. The details in false-belief task of turn around test. Movie S7. Turn around test true belief task. The details in true belief task of turn around test. Abu-Akel, A. Neuroanatomical and neurochemical bases of theory of mind.

Neuropsychologia 49, — Asakura, N. A bayesian framework for false belief reasoning in children: A rational integration of theory-theory and simulation theory. Bardi, L. Repetitive TMS of the temporo-parietal junction disrupts participant's expectations in a spontaneous theory of mind task. Bengio, Y. Towards biologically plausible deep learning.

Google Scholar. Roeyers H, Buysse A, Ponnet K, Pichal B Advancing advanced mind-reading tests: empathic accuracy in adults with a pervasive developmental disorder.

Cognition 89 : 25— Eur J Dev Psychol 7 : — Malle BF The fundamental tools, and possibly universals, of human social cognitiond. Chapter Google Scholar. Cutting AL, Dunn J Theory of mind, emotion understanding, language, and family background: individual differences and interrelations. Child Dev 70 : — Jenkins JM, Astington JW Cognitive factors and family structure associated with theory of mind development in young children.

Dev Psychol 32 : 70— Mind Lang 15 : — Behav Brain Sci 28 : — Acta Psychiatr Scand : — Capps L, Kehres J, Sigman M Conversational abilities among children with autism and children with developmental delays.

Autism 2 : — J Autism Dev Disord 29 : 45— Neurosci Biobehav Rev 34 : — Child Lang Teach Ther 20 : 1— J Neuropsychiatry Clin Neurosci 21 : — Neuropsychologia 48 : — Chakrabarti S, Fombonne E Pervasive developmental disorders in preschool children: confirmation of high prevalence. Am J Psychiatry : — Varga S Pretence, social cognition and self-knowledge in autism.

Psychopathology 44 : 46— Pellicano E The development of core cognitive skills in autism: a 3-year prospective study. Child Dev 81 : — Dahlgren SO, Trillingsgaard A theory of mind in non-retarded children with autism and Asperger's Syndrome: a research note. J Child Psychol Psychiatry 37 : — J Autism Dev Disord 33 : — Ponnet K, Buysse A, Roeyers H, De Corte K Empathic accuracy in adults with a pervasive developmental disorder during an unstructured conversation with a typically developing stranger.

J Autism Dev Disord 35 : — J Speech Lang Hear Res 41 : — J Commun Disord 42 : — Soc Neurosci 1 : — Res Autism Spectr Disord 4 : — Infant Child Dev 11 : — J Clin Ethics 21 : — Child Dev 78 : — J Child Psychol Psychiatry 41 : — Oxford University Press, Oxford pp — J Deaf Stud Deaf Educ 8 : — Schizophr Bull 31 : 21— J Child Lang 21 : — Br J Vis Impairment 26 : — Neuropsychologia 41 : — Dolan M, Fullam R Theory of mind and mentalizing ability in antisocial personality disorders with and without psychopathy.

Psychol Med 34 : — Fonagy P On tolerating mental states: theory of mind in borderline personality. Bull Anna Freud Centre 12 : 91— Australas Psychiatry 17 : — Dodge KA Social-cognitive mechanisms in the development of conduct disorder and depression.

Annu Rev Psychol 44 : — Dev Psychol 45 : — Lin S, Keysar B, Epley N Reflexively mindblind: using theory of mind to interpret behavior requires effortful attention. J Exp Soc Psychol 46 : — Back E, Apperly IA Two sources of evidence on the non-automaticity of true and false belief ascription. Cognition : 54— Pers Soc Psychol Bull 27 : — Hamilton AF Goals, intentions and mental states: challenges for theories of autism.

J Child Psychol Psychiatry 50 : — Decety J, Grezes J The power of simulation: imagining one's own and other's behaviour.

Brain Res : 4— Neuropsychologia 45 : — Br J Dev Psychol 27 : — Burnett S, Blakemore SJ Functional connectivity during a social emotion task in adolescents and adults. Eur J Neurosci 29 : — Brothers L The social brain: a project for integrating primate behavior and neurophysiology in a new domain. Concepts Neurosci 1 : 27— Lieberman MD Social cognitive neuroscience.

Trends Cogn Sci 11 : — A review of the neuroimaging literature. Hum Brain Mapp 30 : — Soc Neurosci 13 : 1— Neuroimage 52 : — Neuroimage 53 : — Curr Opin Neurobiol 18 : — J Clin Neurosci 17 : — Trends Cogn Sci 9 : — Akhutina TV L. Vygotsky and A. Luria: foundations of neuropsychology. J Russ East Eur Psychol 41 : — Vygotsky LS Psychology and localization of functions. Neuropsychologia 3 : — Download references. I express my gratitude to the editor and the anonymous reviewer for their valuable comments and criticism, to Susan Finnel for language editing, and to Charles Njiokiktijien for his encouragement and support.

You can also search for this author in PubMed Google Scholar. Correspondence to Baris Korkmaz. Reprints and Permissions. Korkmaz, B. Theory of Mind and Neurodevelopmental Disorders of Childhood. Pediatr Res 69, — Download citation. Received : 27 October Accepted : 05 January Issue Date : May Anyone you share the following link with will be able to read this content:.

Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative. Journal of Autism and Developmental Disorders Integrative Psychological and Behavioral Science Neurological Sciences Advanced search.

Skip to main content Thank you for visiting nature. Download PDF. Subjects Cognitive neuroscience Neurodevelopmental disorders Paediatrics. Abstract To a large extent, the human infant is socialized through the acquisition of a specific cognitive mechanism known as theory of mind ToM , a term which is currently used to explain a related set of intellectual abilities that enable us to understand that others have beliefs, desires, plans, hopes, information, and intentions that may differ from our own.

Main Human beings continuously make inferences about the psychological states of others. Developmental Precursors of ToM Normally developing children attain ToM at roughly 3—4 y through a progression of stages starting at around 18 mo with the awareness that their own mental states are distinct from those of others 4 , 5 , 7.

Theory of Mind On the basis of precursors and the incorporation of several other neuropsychological functions, children's social interactions with peers and adults triggers and promotes spontaneous development of ToM around the age of 3—4 without any formal instruction or overt effort.

Measuring ToM in Children ToM testing started with the study of autism, and to date, more than 30 experimental tests have been developed for measuring ToM in children Real-Life Consequences of ToM Development and Clinical Conditions With ToM Deficit The development of ToM ability has important consequences for children's social communication, interactions, and behavior, for example, in conversations, negotiations, games, and friendships, which involve interpersonal sensitivity in real social settings including home, school, and the work environment 59 , Autism spectrum disorders.

Developmental language disorders. Attention deficit hyperactivity disorder. Congenital deafness. Congenital blindness. Personality disorders. Neuropsychologia 3 : — Article Google Scholar Download references. Acknowledgements I express my gratitude to the editor and the anonymous reviewer for their valuable comments and criticism, to Susan Finnel for language editing, and to Charles Njiokiktijien for his encouragement and support.