CC BY
35
Hristo Kyuchukov УДк 81'23
approaches то theory of MIND
The paper presents overview of different approaches to the study of Theory of Mind with children. Publications in support of three major theories of child development: the theory of Piaget, the theory of Vigotsky and the core knowledge theory are analyzed and discussed. At the same time very detailed information about the establishment and development of the research on Theory of Mind is presented.
Keywords: theory of mind, cognitive development, theories
1. Background
The children start to understand the social life from very early age. What they know and how they know what they know are some of the questions, which the researchers try to answer for last several decades. Actually there are three major theories of child development: theory of J. Piaget, theory of L. S. Vygotsky and the core knowledge theory, which are discussed in the literature and provide the ground for the development of Theory of Mind (TOM).
J. Piaget was the first European researcher who proved that children during the preoperational stage develop their egocentrism and cannot understand the other peoples view, thinking and knowledge. Piaget did not think that the language had a power on the cognitive development, but rather the sensorimotor activities developed children's experiences, which they later labeled with words. The children's speech Piaget calls "egocentric speech" because young children have difficulties to take the perspective of others. Their talk is considered to be a talk for "their self'. The social contacts with age mates make the children to see and understand the viewpoint of others.
Lev Vigotsky was another main XX century psychologist that did work on child cognitive development. In his opinion the language plays an important role in the development of children because it helps children to think about mental activities and behavior. According to Vigotsky the children speak to themselves for self-guidance objecting Piaget's view on "egocentric speech". Lately the researchers called the self-directed speech "private speech". Vigotsky did believe that the children's learning takes place within the "zone of proximal development", where the adult play a role of supporter of children's skills. The parents and the older siblings are scaffolding the child's learning processes and performance. when the child does not know how to proceed than the adult directs the child's actions or breaks the task to small manageable units suggesting strategies how to implement/perform the task. Adults' cognitive support when a child performs a task predicts child's mature thinking and help the child to reach the intersubjectivity-the shared understanding between two participants during a communication process, play or other activities.
The third major theory of child development is the "core knowledge" theory introduced by American psychologists (E. Spelke and collaborators). According to the theorists of "core knowledge" the children are born with core domains of thought. The theorists of core knowledge are more supportive of the theory of J. Piaget and in a way
they develop it further. The core knowledge theorists support the idea that the children are born with physical and numerical knowledge. The newborn child being in contact with objects starts to understand how the objects affect each other and they can add and subtract small quantities. In preschool years children develop also their linguistic and psychological knowledge, and this is the period when the children start to understand the mental state of others, i.e. emotions, desires and beliefs. The theorists of core knowledge believe that children develop each core domain independently. The child development is seen rather as a domain-specific process and the children are viewed as naïve theorists who develop their everyday realities on their physical, psychological and biological core knowledge. According to the Theory theory (theory of children as theorists) children can explain an event after observing it or after developing an internal concept about it. One of the most investigated naïve theory is the children's theory of mind-when preschoolers develop social knowledge and understanding of their own and others beliefs.
D. Premack and G. Woodruff (1978:515) are the first researchers (doing studies with chimpanzees) who give a definition of the theory of mind: "In saying that an individual has a theory of mind, we mean that the individual imputes mental states to himself and to others (either to conspecifics or to other species as well). A system of inferences of this kind is properly viewed as a theory, first, because such states are not directly observable, and second, because the system can be used to make predictions, specifically about the behavior of other organisms". The two philosophers in experiments with apes came to conclusions that organisms have an understanding of other organisms' desires or knowledge.
Commenting on the article of Premack and Woodruff, J. Bennett (1978) suggests that the future experiments should be designed in such a way that keep belief about motivation and belief about cognition. D. Dennett (1978), also commenting on the article of Premack and Woodruff suggests few important things. First of all he suggest that the children should be asked "why questions", and nonverbal tests also could be experimented. Dennett suggests the "minimally complex pattern" should have the following format (p.569)
C beliefs that E beliefs that p;
C beliefs that E desires that q;
C infers from his beliefs in (1) and (2) that E will therefore do x and so, anticipating E's doing x.
Based on the Premack and Woodruf's (1978) paper and the comments of Bennett (1978) and Dennett (1978), H. Wimmer and J. Perner (1983) developed the first test for testing the comprehension of child subject's wrong belief. They constructed a test containing a miss displacement task, where one character puts something in one place and a second character changes the place of it in the absence of the first character. Than the question, which the subjects are asked is where the first character will look for that object. And this procedure test the subjects' understanding of others wrong beliefs. Later other false belief task was developed - "the wrong expectation content" unexpected contents task and this is a task when a subject is asked what is a content of a familiar container when it is shown to him and then the subject discover that the content is something different. The subject is asked what another subject would think if the box is shown to him. (J. Perner, S. Leekam and H. Wimmer, 1987).
2. Theories of Theory of Mind
There are several different theories how the Theory of Mind develops in young children. Each of these theories is somehow related to some of the 3 classical theories in the field of child development. The child's understanding of the theory of mind is called not only on one type of account and in this part I am going to analyze the main ideas of the contemporary theories of the theory of mind.
One of the most important Theory is the one developed in early 1990-s by A. Gopnik and H. Wellman (1992). The authors contrast two main views: one is the view that the child understanding of mind is an implicit theory analogous to scientific theory and changes in the understanding may be understood as theory changes. The second is the view that the child "not really understand the mind, in the sense of having some set beliefs about it. She bypasses conceptual understanding by operating a working model of the mind and reading its output" (p.145). The child's model is her own mind.
Gopnik and Wellman believe that the child's understanding of mind is construed as a theory and the changes in understanding may be interpreted as theory changes. Further the authors make characteristics of the theories and two main features of theories are identified: their abstractness and their coherence. Another characteristics of theories, which Gopnik and Wellman describe is the predictions which theories make based on empirical observations. The authors summarize that "all these characteristics of theories ought also to apply to children's understanding of mind if such understandings are theories of mind" (p.148).
Gopnik and Wellman propose that between age of two and half and four and half there is a change from mentalistic psychological theory to another psychological theory. The change involves transition from one view of the mind to another. At the age of 2 years the child is clearly mentalist and not a behaviorist. It seems that the mentalism is the starting state of psychological knowledge. It seems even at two years psychological knowledge seems to be structured largely in terms of two types of internal states: desires and perceptions but this knowledge excludes any understanding of representation.
"Desire and perception can be, and at first are, understood in nonrepresentational terms. Desires at first are conceived simply as drives towards objects. Perceptions are at first understood simply as awareness of objects. In neither case need the child conceive of a complex prepositional or representational relationship between these mental states and the world. Instead, these very young children seem to treat desire and perception as fairly simple causal links between the mind and the world."(p.150)
In early years like the age of two it seems the children use only terms for desire and perception but more cognitive mental terms (think, know, remember, belief) begin to emerge around the third birthday. At the beginning the 3 year olds' understanding of belief is like their earlier understanding of perception. Belief, like desire and perception, involves direct causal links between objects and believers. By age of 4 and 5 the children develop different view of the mind- one which is called "representational model of the mind". Desires, perceptions, beliefs involve the same fundamental structure. These mental states all involve representations of reality.
There are few characteristics of the child's theory as a theory and they are: explanations, predictions, interpretations and transitional phenomena. Children's explanations of actions show a characteristic theory-like pattern. In open-ended
explanation task children are simply presented an action of reaction and asked to explain it. The 3- and 4-year-old children's answers to such open-ended question are organized around beliefs and desires just as adults. Two-year-olds' explanations almost always mention desires, but not beliefs. The three years old invoke beliefs and desires and 4 and 5-year-olds refer to the representational character of these states, explaining failure in terms of falsity.
The children are able from very early age to make predictions about actions and perceptions- their own and others. The children are able to predict that desires may differ and that desires may be fulfilled and may not be fulfilled. The fulfilling of a desire leads to happiness and not fulfilling of a desire leads to a sadness. Children with desire perception theory should be able to predict the perceptions of others, including those in which the perceptions are different from their own. By age of two and half a child can reliably predict when an agent will or will not see an object. They also can predict how seeing an object will lead to later action (Level-lundersatnding). However they cannot predict about representational aspect of perception (Level-2 understanding). They cannot predict that an object which is clearly seen by both parties can look one way to one viewer and another way to another. The most known examples of incorrect prediction is the false belief error in 3 year olds. The ability to perform correctly on a false belief task is taken as evidence that the child has a representational theory of mind.
From the point of view of the theory theory these incorrect belief predictions are mirrored in 3 years olds' performance on a wide range of other tasks. Gopnik and Wellman present an inventory of these type of tasks: 1) appearance-reality task; 2) questions about the sources of belief; 3) understanding of pictorial representational systems. In some of these tasks the desire-perception theory makes incorrect predictions and children consistently give the same wrong answer.
Children seem to first understand both belief and representation as small extensions of the original non-representational desire-perception theory. This stage appears to be an intermediate one between a fully non-representational and a fully representational theory of mental states. Transforming from two and half to five years have all features being a theory change. "While initially the theory protects itself from counter-evidence, the force of such a counter-evidence eventually begins to push the theory in the direction of change. The first signs of the theory shift may emerge when the counter evidence is made particularly silent. Moreover the theory initially deals with such counter-evidence by making the relatively small adjustments to concepts that are already well-entrenched, such as desire and perception. Finally by 4 or 5 the new theory has more completely taken over from the old. The predictions are widely and readily applicable to a range of cases." (p.158)
According to Gopnik and Wellman (1992) in the theory theory to predict someone's behavior there should be recourse to theoretical constructs such as beliefs and desires. Explaining someone's behavior involves more than empirical generalization. It involves appeal to constructs at a very different level of vocabulary- X wants Y and beliefs Z. A distinction between phenomenal description and a theoretical explanation is crucial. Gopnik and Wellman point that "on the simulation theory the child's understanding of mind is more closely linked to the phenomenal than to the theoretical. Understanding the states of mind involves empirically discovering the states or results of a model." (p.159)
In solving a false belief task according to simulation theory the child does not need to have a theoretical construct of belief (or desire) to solve the task. The child has access to his own first-hand mental system and uses that. When asked what the character "thinks", the child does not need to understand the beliefs as something like representational construct. The child simply simulates the experience and reports his own specific resulting state. The failures to solve this task, on this view, reflects a failure of simulation, rather than a failure of knowledge. It is not that the young child do not understand beliefs as states of misrepresentation, it is just the younger child makes egocentric simulation, projecting her own current mental states onto the others.
For both theories (Theory Theory [TT] and Simulation Theory [ST]), Gopnik and Wellman, predict that there will be development: children should be first good in predicting/explaining "easy" states and then the "hard" ones. The notion of easy and hard should differ between these two theories. For Simulation Theory the critical difference should be between states that are difficult or easy to stimulate. The metric for easy and difficulty must be intimately related to the similarity of the states to the child's own states. In this sense the simulation theory is in another tradition- the tradition of "perspective-taking" views in development. Young children's errors are "egocentric" - the child's early error consists of not correctly adjusting their simulation to other person's condition. In contrast for the Theory Theory the critical metric concerns states that are easy or difficult to conceive of. It is important a difference between an early non- representational and a later more representational understanding. Early on children have a relatively adequate understanding of non-representational desire-perception states. Later they develop an understanding of the representational state of belief, specially, and a representational understanding of mind more generally (including, a representational understanding of certain aspects of perception and desires).
Summarizing Gopnik and Wellman (1992) conclude that the children find some sorts of mental state attributions to be difficult and some to be easy. The difference between the easy and hard attributions is not clearly related to the distinction between the self and others, which is expected from ST. The distinction is related to the ability to conceive of and interpret some types of mental states and not others, for self and for other. From a theory point of view this makes sense. A second difficulty concerns whether children are at first generally egocentric about the mind and then overcome this by learning they must adjust their simulations for others. A third empirical problem is that the simulation theory has difficulty explaining the structure of the explanations that the children offer. It is commonplace to say that the child's theory is not of course an explicit theory but rather an implicit one, which may have to be inferred from behavior rather than being openly stated. A fourth difficulty involves the predictive scope of the simulation theory versus the theory theory. The simulation theory provides a good account of one particular type and deficit, perspective taking difficulties, when they occur (although as mentioned earlier ST seems to mischaracterize the nature and the developmental progression of egocentric errors).
The two theories (TT and ST), which are discussed, by Gopnik and Wellman (1992) show evidences in support to J. Piajet's theory of egocentrism. The emphases which the two authors put on the child's understanding of the world through the desire-perception point of view tie their understanding for developing theory of mind with the theory of Piaget about the child development in general.
K. Nelson and her collaborators (2003) present another view to the theory of mind of young children. The authors introduce the term "entering a community of minds" and contrasting it with the "acquiring of a "theory of mind". The authors say "Acquiring implies taking possession of something, and in this case the "thing" is the theory about an abstraction- a mind with universal properties and mechanisms. In contrast, entering involves coming into new place from where one was previously; here, the place is a community, that is, a space occupied by people who are related by common purposes and understandings. The community that one enters includes minds that communicate (thus the community). Minds differ- this is the point of communicating. These contrasts, we believe, tell two essentially and critically different stories about vital developments in the early childhood years" (p.25).
The aim of Nelson and her collaborators (2003) is to explore the implications of the experiential approach to the important social cognitive understandings achieved in the early childhood and later years. The authors do not limit their work only to the theory of mind tasks, which in their view require an explicit socio-cultural-linguistic based theoretical account. Rather the experiential approach which they propose show ways in which the young children increasingly enter into "new relationships of self and other understanding within the family, the peers and with other group members outside the family, that is the community". Nelson and collaborators think that the relationship is established through activities, symbols and language. The language plays an important role because through the language the children understand the cultural world and the interpretation of relationships.
Nelson and collaborators doing standard experiments on theory of mind aim to uncover the sense-making strategies that child bring from their experimental pass to the TOM experimental situations. In series of experiments with different variations of unexpected contents authors design to establish a discourse framework for children. However, the authors were not interested in children's explanations if they were correct or how they related to performance of false belief task, but rather they were interested in what children's responses are to "why" questions, because it might suggest about their thinking about the task situations and its relation to their previous experience.
Nelson and collaborators found that some children focus their explanations on likes and desires, primarily when explaining other children's beliefs. In these terms the authors claim that the children have earlier understanding of false belief in terms of desire, rather than belief or young children may be encouraged in home context to choose terms of desires. Another thing what the study was interested in was if the children understand what it means to "think" something true or false, which is a requirement of the unexpected content task. The result in this study show that young children do not associate seeing with thinking. According to authors the "young child thinking is not equivalent with either saying or seeing. Rather, children may consider these processes as mutually exclusive."(p.35)
Most of the studies in past showed that TOM research is an accomplishment of the preschools years. Nelson and collaborators show that language is an important tool for cognitive and social competence. Conversation helps the child not only to understand and represent the word meaning but also for their development. Discourse with unfamiliar people and situations, discussions with others about knowledge and different experiences
and articulating their own experiences, are helping children to compare their own view, based on their own experience, with another's contrasting view, as present in language.
Another author who also supports the Vigotskyan theory of children's development is P. Harris (1989, 2012), trying to answer the question how the children learn from what the other people tell them.
There are authors like J. Perner (1991) and B. School and A. Leslie (1999) who have more "core knowledge" theory approach to the development of children's Theory of Mind. J. Perner (1991) in his early 1990-s writings tests belief and desires, because they are treated sometimes "as a equivalent components in theories of action or like symmetrical mirror images among the different mental states" (p. 139). Perner looks on how the two mental states- belief and desire represent the child's cognitive situation using the word think and he separates thinking that something is the case (belief) from thinking of something (pure thought). In his work Perner also aims to explain why children understand easily some stories with the verb think, and other children with difficulties.
In order to explain the representation Perner says that there is a need of two different things, which take different states. The first he calls "representational medium" and the second the "world". It is clear that between the two states there should be a correspondence in order to operate together and he calls it causality. For example the content of a desire represents the desire state of the world. But the casual relationship is opposite than what is required - the desire state represents the desire. This is important but you need to make it clearer. And anything that is causally influenced by something else, according to Perner, it could be count as a representation of it.
Further in his paper Perner discusses the interpretive system. Something is interpreted in one or in another way up to the interpreter and his experiences. As the representation occurs only in biological systems "whose organs and processes are replicated or repeated because they serve an important function. The function of representation is to correspond to aspects of the world. The importance of this function lies in the fact that the organism can use the representation to stand in for the real world." (p. 144)
Based on discussions in the literature and his own observations Perner proposes that the representation should contain two elements: "1) that there must be a correspondence between states of the representational medium and states of the represented world, and 2) and the correspondence must be exploited by an interpretive system so that the representation is used as a stand-in for the represented". However the author claims that the stand-in function is the one, which determines the asymmetry between representational medium and represented world, and not the casual relationship and the causality is the one, which plays an important role in establishing reliable correspondence. In Perner's opinion representations include epistemic mental states like perceptions, beliefs and knowledge and the distinction between thoughts and beliefs is brought with the use of the particle "THAT". The author gives examples with the use of the sentences containing think of and think that and in the second example there are two interpretations of the sentence: a) sense of a belief and b) referent of a belief. Further Perner pays attention to the meaning of the verbs want and think that as asymmetry in function between belief and desire, and uses that for explanation why children are more proficient in understanding and reasoning about wants than about beliefs (thinking that).
According to Perner by the age of 2 children have "some kind of "theory of mind", but at the same time they fail to understand some other basic aspects of the mind. In the age around 4 they understand that the people have false beliefs and distinguish between appearance and reality. However Perner thinks that the children before age of 4 interpret the mental states as relations to situations directly. Later the children can relate them as representations of the situations and can understand that the mental attitudes towards situations have different implications. A person who has an attitude "want" towards a non-existing situation then that person will do something to bring about that situation, while if a person just thinks of a situation he/she will not act. Perner says that the "without a conception of mental states as representations the child cannot understand thinking that. The child has a theory of thinking but can only, at best, assimilate thinking that to thinking of. This inability to understand thinking that provides the basis for understanding why children find wants do much easier to understand than beliefs. Understanding most relevant behavior and emotional implications of desire does not need the understanding of that. For belief, this understanding is essential."(p.149).
Further analyzing earlier experiments Perner comes to the conclusion that the tasks involving "thinking" can be easy or difficult. It depends from the interpretation of the verb think- as a mental representation (belief, thinking that which requires a content-referent distinction) or simply as a relationship to a situation (thinking of). Tasks involving "want" do not appear to require a representational interpretation. It seems that children understand the internationality when a person's goal is explicitly specified. The difficulty comes "when the children have to understand the causal relationship between events in the world, desires, and action. Because the causal relationship between desire and action is constitutive of concept of intention."(p.153)
The 3 year old children understand that there are desire situations and that people do everything to get their desires fulfilled. This kind of understanding allows for the causal reasoning, because someone does something because he/she wanted it. To understand the intentional causation is more required. The child has to understand that the action of someone towards the desired situation has to be caused by mental representation. In conclusion Perner is trying to summarize the answer of the question "why do children find problems involving a person's wants easier than problems involving thinking". The analyses of thinking of and thinking that showed that the children's easy or difficult understanding of something depends on different distinction between the mental states involved as a mental representation or understand them as a relationship to a situation.
B. Scholl and A. Leslie (1999) examine an idea that the theory of mind arises from an innate, which they call a cognitive architecture-module. The authors try to answer two questions: What kind of cognitive module would be required in order to explain the TOM? What kind of development would such a module need to undergo in order to account for relevant data? The debates are between the supporters of TOM and the cognitive modules, where the TOM is taken as a developmental and the cognitive modules are taken "anti-development". Scholl and Leslie take a different approach and they argue that the modularity and development can be intimately related. Leslie and colleagues belief that the child has an early competence with concept of belief, but there are performance limitations.
It has been proposed that the TOM is a specific architectural module and the reason for this are that it operates in a specific domain and the bases for it may be innate. Theory of Mind requires the owner to have acquired the concept of belief. Even though the children do not see, hear or feel internal states they become competent at reasoning about them.
According to Scholl and Leslie "Modules are domain -specific: they only operate on certain kinds of inputs-"specialized systems for specialized tasks". Modules process in mandatory way, such that their operation is not entirely under voluntary control.
Modules are typically fast, perhaps due in part to the fact that they are encapsulated (needing to consult only circumscribed data base) and mandatory (not needing to waste time determining whether or not to process incoming input).
Modules offer highly constrained "shallow" outputs, which themselves often undergo further processing down the line.
Modules may often (though need not) be implemented in fixed specialized portions on neutral architectures.
Modules, and the abilities they support, may be selectively impaired by neurobiological damage" (p.133)
Scholl and Leslie claim that the TOM has a specific innate basis in that the TOM is given as a part of the genetic endowment. This is in contrast to "theory-theories" which say that TOM is learned exactly in the same way as scientific theories are learned. In their view the normal acquisition of the TOM is at least in part due to the operation of a TOM specific architectural module.
One of the predictions which the modular theory makes is that the end state of the development should be the same across individuals. The environmental interaction may affect the precise time when the modular capacity shows itself but the end state should be the same. In keeping with this prediction the acquisition of TOM is largely uniform across individuals and cultures. A person develops the Theory of Mind and it does not depend on the character of their environment. It seem that all human beings have the same basic TOM. Citing Harris (1990) Scholl and Leslie agree that the entire TOM emerges universally in the young children with approximately the same timetable. The emphasis has been that the development of the TOM is identical across-cultures and very few authors argue that there are cross-cultural differences in TOM. However the universality of Theory of Mind is also predicted by "theory-theory" as well as by a modularity theory. Given the similarity of human relationships and children's experiences of people the evidence for building a Theory of Mind could also be argued to be universally available.
Scholl and Leslie argue that the facts about development cross-culturally are relevant to the modularity hypothesis, because there is striking uniformity in the theories that develop. Although the environment in which the child is raised may affect when different capacities emerge -for example children with multiple siblings seems to learn TOM faster - the basic concepts are similar across cultures. People in different cultures do not develop different concepts of intention or desire.
The authors stress one further points which have to be underlined. The point is that the TOM discussed in the contexts of the modularity theory refers only the origin of the basic TOM capacities. They acknowledge that higher order TOM activities may not be uniforms across individuals or cultures. In conclusion Scholl and Leslie say that
the modular capacity may be acquired in four different ways (p. 149): 1) the essence of the modular capacity is that inner modular capacity is fixed; 2) the character of the innate modular capacity may be determined by the environment in a number of highly constrained ways. The modular capacity may reach a number of particular end-states; 3) the modular capacity may have an innate basis but there is internal development given the constrained impute; 4) some of the properties and contents of the capacity or skill may not have an innate basis at all, but develop rather as a expertise that comes to have the properties of the module. The authors conclude that modularity and development can be quite intimately connected, when account is taken of the distinction between competence and performance, and the distinction of the mature TOM as a complex aggregate of capacities and the early TOM out of which the mature competence develops.
conclusion
The understanding of the social life from the surrounding world by young children is not happening only in one way. There are different ways how the children gain knowledge and there are different factors which help the children to understand the relations between the people, their desires and wishes, their thinking and beliefs. The psychologist and linguists for last decades already tried to find out what are the main factors and how particularly the language help the children to understand the other people's behavior. Nowadays more new research is initiated in this area and the interest of the scientist is also towards bilingual children and children with Autism and how they understand the social relations in their communities and what determinants help them to get from very early age the Theory of Mind. Still a lot to be done in these directions.
Acknowledgments: I would like to thank Jill de Villiers for inspiring to write this article and for her grate work reading and reviewing an early version of the article.
References
Bennett, J. (1978) Some remarks about concepts. The behavioral and Brain Sciences, 1, 557-560.
Dennett, D. (1978) Beliefs about beliefs. The behavioral and Brain Sciences, 1, 568-570.
Gopnik, A., Wellman, H. (1992) Why the child's Theory of Mind really is a Theory. Mind & Language, 7, 145-171.
Harris, P. (2012) Trusting what you're told. How children learn from others. Cambridge, MA: The Belknap Press of Harvard University Press.
Harris, P. (1989) children and emotion. Oxford: Blackwell.
Wimmer, H., Perner, J.(1983) Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition, 13, 103-128.
Perner, J. (1991) On representing That: The asymmetry Between Belief and Desire in Children's Theory of Mind. In: Douglas Frye and Chris Moore (eds.) Children's Theories of Mind: Mental States and Social Understanding. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers.
Premack, D., Woodruff, G. (1978) Does the chimpanzee have a theory of mind. The behavioral and Brain Sciences, 1, 515-526.
Scholl, B., Leslie, A. (1999) Modularity, Development and "Theory of Mind". Mind and Language, 14, 131-153.
