ABSTRACT

The classic false belief task (with its numerous versions) is considered the main instrument for assessing the children’s abilities to reason on other person’s beliefs. It is the task most frequently used for the study of social-cognitive abilities development. Despite these, last years researches show that this task is inadequate for the study of this domain. This inadequacy is manifested at many levels. According to one of these levels, excessive processing requests of the task leads to incorrect answers from children and to failure at this task, while they have the necessary abilities at the conceptual level. The aim of the study is to show that negation, as a supplementary cognitive step, is one of the excessive requests at the performance level, which is not relevant for children’s abilities to reason on the content of beliefs. The results sustain this inference.

 

KEYWORDS: children’s knowledge about the mind, false belief task,

                       negation, affirmation

 

 

CHILDREN’S KNOWLEDGE ABOUT THE MIND

 

An important part of the cultural values that the developmental niche imposes to its new members have no longer a survival function. They became more and more specific and now they target the development and the transmission of a sophisticated and artistic communication system, complex social behaviors that are strongly rooted in the emotional and cognitive abilities of the human beings. All this supposes that the human beings are strongly interrelated and that they perceive the others as humans, like them, with similar goals and feelings. To know what another one feels, to identify the significance of his gestures and his behaviors are vital aspects for human beings surviving and development.

Understanding others as having a mind is the main aspect of the knowledge about the mind. There are many aspects of human behavior that reveals the existence of a mind capable of thoughts and emotions. Even from the first hours after birth, the child shows some behavioral signs that were interpreted as proofs of perception and understanding of people as people. Of course, there are many voices that argue against this kind of interpretations saying that they are just the result of adult projections for those situations and not the objective reality. The simplicity principle must guide this kind of interpretations. It is indeed wise to doubt this kind of results, but to doubt endlessly the children’s abilities to be human beings from the beginning of their ontogenetic evolution and to try to undermine any proofs in this direction through complex and twisted reasoning means, probably, an underestimation of the human species (Flavell, 1999).

In 1999, Flavell promoted a new label for the old concept of theory of mind: knowledge about the mind. Indeed this new label is a more appropriate than the old one and is in accordance with many critics that were formulated about the theoretical aspect of this ability development. But even so, this domain remains poorly shaped. In what will follow I will try to shape this domain of study. All children’s cognitive-behavioral manifestations that indicate the perception of others as human beings that poses a mind are of high relevance. These cognitive-behavioral manifestations are presented respecting their succession along the ontogenetic development.

The most important components of children’s knowledge about the mind are:

The preference for human faces – immediately after birth infants have often been found to look more at face-like than non-face-like patterns and it seems also that they are sensitive to the facial orientation, spending more time looking at the faces that had an upright orientation (and were judged by adults as being attractive) (Slater et al., 2000). This is possible by processing internal facial features, which also explains newborn’s ability to recognize and to prefer his/her mother face as compared with another woman’s face (Field et al., Bushnell et al., Pascalis et al., Walton et al., apud Slater et al., 2000).

Agents versus inanimate objects – differentiating agents from inanimate objects is a basic ability that affords us to understand others as human beings with mental states like desires, beliefs, emotions, etc. There are a number of stimulus properties that differentiate agents from inanimate objects: different aspects of biological motion like self-propelled movement; autonomous nonrigid transformation of an object’s surface (Gergely et al., 1995). Newborns manifest manual gestures in the presence of people, but not in the presence of objects, at 2 months they imitate gestures produced by humans, but not those produced by objects that imitate humans. At 4 months infants vocalize more when people leave the room where the child is (Lagerstee et al., 2000). At 5 months children associate an object with a hand that is trying to grab it but not an object with a stick that touches it (Woodward, 1998). At 6 months they start to follow another person’s actions and his attention orientation, and they expect that people’s action to be related functionally with objects (Lagerstee et al., 2000).

The perception of emotions and the manifestation of emotions – even from the first months after birth, infants not only prefer to perceive human faces, but they also have the ability to discriminate between different types of adult behaviors, especially those one that are emotional in nature. At 2 months they start to monitor and to reciprocate with their partners in the context of intimate dyadic exchanges. At 3 months infants discriminate between different types of facial expressions. It seems that they prefer their mother faces when they are smiley as compared with their mother’s still-face (Striano, 2001). This ability to differentiate between different types of emotions is also partly due to the adult’s special manner of interaction with the child, like the so called motherese speech (Trainor et al., 2000).

The manifestation and the perception of intentions – intentionality is a type of psychological relation between people and objects. To see people as having intentions means to understand that their intentional states (goals, desires, beliefs) are the mental causes of their actions (Gergely et al., 1997). The intentional stance is a useful evolutionary strategy because it allows the prediction of people’s future behavior (Dennett apud Gergely et al., 1995). There are two conditions that are essentially and absolutely necessarily for the identification of goals and for the behavior prediction: 1) action’s equifinality and 2) action’s rationality (Gergely et al., 1995, 1997). At 9 months the infant uses systematically communicative gestures with instrumental purposes, s/he points and s/he changes the eye gaze for joint attention purposes and for social reference purposes (Butterwort et al., Bretherton şi Campos et al. apud Gergely et al., 1997). At 12 months children perceive and understand the intention on an equifinality and rationality basis (Gergely et al., 1995).

Desires – refers to what a person wants, what a person desires. Desires are part of complex mental aspects because they are subjective and because they have fewer external behavioral referents. Desires are expressed mostly at the emotional level. At 18 months children understand that their desires are different from other person’s desire and they act according to this information (Gopnik et al., 2000).

Pretending – is an ability involved not only in children’s games, but also in different cognitive capacities like: counterfactual thinking, conditional planning, empathy, moral understanding, visual imagery. The main features of pretence are (Nichols et al., 2000): a) existence of the initial premise, b) inferential elaboration, c) non-inferential elaboration (embellishment), d) production of appropriate pretend behavior, and e) cognitive quarantine. The existent studies show that this ability is present at the age of 2 (Nichols et al., 2000).

Deceiving – it implies to make another person believe something that is true when it’s actually false (Baron-Cohen, 2000). The central point of deceiving is to create a false belief in someone else’s mind for a future purpose (Carlson et al., 1998). It implies that you know that there are beliefs, which can be true or false, and it also implies to know that these beliefs can be manipulated, that people change their beliefs according to the information they directly depict from the environment or according to the information that is offered by other persons. Therefore, deceiving has a major importance for the understanding of other’s mind. At the age of 2 or 3 children are able to deceive (Carlson et al., 1998).

Belief understanding – beliefs are representational mental phenomena that are the result of the interaction between the individual and the world. Humans build a copy of reality on the basis of the information that they possess. Having false belief means that the person has an erroneous image of the world. Because different people have different experiences, their beliefs will be different. People’s goals and behaviors are specifically based on beliefs. Therefore, understanding people as having beliefs, true or false, is a central element of understanding people.

 

 

MORE THAN ONE REASON TO BE SCEPTIC REGARDING   

THE FALSE BELIEF TASK

 

One of the most frequently used tasks for the study of knowledge about the mind development is the false belief task (FBt). Performances on this task are considered representative for the process of acquiring abilities to rationalize on beliefs during early childhood (either when it is about the acquisition of the concept of belief – Perner, 1995; or when it is about the relation between Theory of Mind Mechanism and the Selection Processor – Roth et al., 1998).

As mentioned above, the two main developmental theories that are studying the knowledge about the mind problem are the “Theory Theory” and “Modularity Theory”.

Theory theorists (Gopnik A., Meltzoff A., Wellman H.M., Perner J., Gelman S.A.) argue that our knowledge about the mind comprises an everyday “framework” or “foundational” theory. The body of knowledge must comprise a system of interrelated concepts and beliefs and not just a collection of unrelated contents. They consider that for the children to pass the false belief task it is necessary to acquire the so called representational mind - to understand that people’s behaviors are related to their representations about the world, these being either true or false.

Modularity theorists (Leslie A. M., Roth D.) consider that young children are not acquiring a theory about the mental representations at all. Rather they postulate the acquisition through neurological maturation of a succession of three domain-specific and modular mechanisms for dealing with agents versus nonagent objects. The third mechanism, Theory of Mind Mechanism₂ (ToMM₂), which begins to develop during the second year of life, allows children to represent agents as holding attitudes toward the truth of proposition or so-called propositional attitudes (Flavell, 1999).

The standard version of the FBt presents to the child the character, Sally, who leaves a desired object (e.g. a chocolate) in a basket and then she leaves the scene. In her absence, another character, Anne, changes the object location and puts the chocolate in a box. Subjects are asked to predict the location to which Sally will search for the desired object (or where she thinks the object is) (Wellman et al., 2001). This task was created as an answer to the critics on true belief task. When reasoning on true beliefs it is possible that children’s answers are not a reflection of their abilities to read other person’s belief, but they are merely the result of their interpretation of the real state of affairs (Carlson et al., 1998).

The adepts of the Theory Theory consider, judging from the performances at this task, that around age 4 there is a conceptual change in Theory of Mind, more precisely children understand the representational nature of human mind and they are able to ascribe beliefs to self and others. The failure at this task is a reflection of serious deficits in understanding mental states – a Theory of Mind deficit (Perner, 1995).

Last years literature about the development of knowledge about the mind and its assessment raises many arguments against the exclusive use of this task. These counterfacts are specific for those studies that have the intention to build a new theoretical perspective as an alternative to the already existing theories (“Theory Theory” and “Modular Theory”). I prefer to present these studies this way because they are not yet organized in a coherent form. This new integrative perspective does not necessarily intend to totally discard the already existent theories, but rather to integrate those elements considered valid in a new unitary structure which also integrates new assumptions and corrections to those criticized aspects.

 

False belief and deception

 

Deception abilities and abilities to compute the content of another person’s beliefs are still considered by some researchers as different abilities. This is so because they have apparently different requirements.

The deceiving task has many versions. One of this is the “kitchen” game (Hala et al., 1996). Two experimenters (E1 and E2) are inviting the subject to play together the “kitchen” game. They show the children which are the toys for the game and most important where are the treats that are going to be shared at the end of the game. At a certain point during the procedures, E2 realizes that she left some of the toys in a backpack in another room and decides to go after them. After E2 left the room E1 suggests to hide the treats from E2. The subject is even invited to actively plan the deception and then to set it to work. For correctly solving the task, the subject has to put the treats to another container, different from the initial one.

Other versions of this task ask the children to erroneously leave a trail of footprints which lead to a container that does not contain the desired object or to use an arrow sign that indicates the direction to the wrong location (Carlson et al., 1998; Ritblatt, 2000).

Essentially, this task asks the subject to mislead a character (present or not at that moment) regarding the desired object location (e.g. chocolate, stickers). The subject disposes of many deceiving strategies, according to the version (leaving a trail of footprints on the sand or ink marks, using markers for the locations or signs etc.).

The very point of deception is to create a false belief in the mind of another for some ulterior purpose (Carlson et al., 1998). To create a false belief, the subject has first to know what the true beliefs of that person are, where does he/she know that the desired object normally is. Transforming a true belief in a false one can be done by working on the present state of affairs (which is consonant with the belief’s content): the child searches through the existent alternatives the one that seems the most convenient and then moves the object to the new location (therefore changing the reality).

Unlike FBt where the request to explicitly present the character’s (false) belief is clearly made, the deception task imposes a supplementary intermediary step: reasoning on the content of beliefs (true or false).

One of the advantages of deception tasks is to actively involve children. Setting the personal goal of creating a false belief in another person’s mind is the central element for the children’s best performance at understanding false beliefs (Hala et al., 1996). When children actively plan for deceiving their performance at this kind of tasks increases (Hala et al., 1996; Carlson et al., 1998; Ritblatt, 2000) and this also means that they have a better understanding of other person’s beliefs. It seems that this active participation makes the central elements of the task context to be more salient for the subject and with a higher activation during the entire procedure (Hala et al., 1996).

According to these aspects there is no basis for some researchers’ preference to restrain the conclusions about false belief understanding solely to those studies that use FBts.

 

The implicit nature of the knowledge about the mind

 

Lately, more and more studies raise questions about the nature of the knowledge about the mind. If the innate-acquired debate, in spite of numerous studies, is far away from being solved, the implicit-explicit issue is still “fresh” on the `knowledge about the mind “market”’, expecting that in the future it will be more properly studied. There is already data that suggests, at least partially, that children’s knowledge about the mind might have an implicit nature. Clements et al. (apud Perner, 1995) made an interesting study: the FBt remained mostly the same except that the character’s position was all the time behind the wall from the back of the scene and became visible only when looking to one of the two locations (A or B), each at one of the extreme ends of the display. Clements not only asked children “Where will Maxi look first for his chocolate?” and waited for children’s response, she also videoed their eye gaze in anticipating Maxi’s reappearing. The number of children (by age) showing a sign of understanding false belief, manifested the anticipatory looking, was bigger than the number of children verbally expressing their knowledge. These results have been interpreted as a sign of some implicit understanding, that cannot be verbally accessed (Perner, 1995).

For Deanna Kuhn (2000) understanding knowledge as the product of human knowing is a critical first step in the development of epistemological thinking, which is metacognitive in the sense of constituting an implicit theory of how things are known. Therefore children’s knowledge about the mind (being aware of self and others as knowers, understanding that other people behavior is guided by desires and beliefs that are not necessarily consonant with the reality) are considered early metacognitive achievements. According to this author metacognition develops. It does not appear abruptly from nowhere like an epiphenomenona in relation to first-order cognitions. Instead, metacognitions emerge early in life, in forms that are not more than suggestive of what is to come, and follow an extended developmental course during which they become more explicit, more powerful, and hence more effective, as they come to operate increasingly under the individual’s conscious control (Kuhn, 2000). Therefore metacognitive knowledge develops from implicit to explicit.

Given this situation, classic tasks for the assessment of false belief reasoning prove to be inadequate because they are targeted to explicit knowledge.

 

Insufficient sensitivity to changes that occur during the developmental process of children’s knowledge about the mind

 

Metaanalitical studies show a growing tendency for research studies to use FBt for the assessment of children’s knowledge about the mind in general and then to draw less specific conclusions about the entire domain (Wellman et al., 2001). This is much alarming as a series of studies show that is more appropriate to talk about a gradual development of children’s mentalistic abilities than a development in stages (Haith and Benson apud Flavell, 2000). The gradual aspect is characteristic for the achievement of every aspect of understanding people as humans (therefore we cannot talk about its sudden manifestation in children’s behavior) and also for the global development of this kind of knowledge (different developmental levels can be registered for different concepts at the same time).

There is a series of studies that advocates for this perspective. They show that the improvement of deception task and FBt, mostly on their executive component, ameliorate performances. This kind of improvement appears constantly, even if the performances do not pass the chance level (Wellman et al., 2001). These improvements of the task are:

1)      when the task is presented in terms of explicit deception;

2)      when actively involving the child in strategic planning and making the essential changes, emphasizing the importance of the protagonist’s mental state or the reduction of the contrasting aspects salience of the present state of affairs (Hala et al., 1996; Carlson et al., 1998; Robinson et al., 1995).

The constant explanation for this kind of results was that those elements are masking the earlier abilities for conceptual understanding of beliefs, desires, etc. These earlier masked abilities are not abilities completely developed, at their final state, but they are primary elements which further develop in a continuous manner.

Individual differences can have different forms: shyness can differentiate individuals along life span; other differences manifest only during one of the development sequences and therefore reflect the individual speed for reaching a central point in the developmental process shared by all people. False belief assessment is integrated in the second category and consequently the interpretation based on it must be regarded carefully from the individual differences perspective (Wellman et al., 2001). More advanced studies on the mechanisms underlying the development of this kind of knowledge could be influential for designing more analytical tasks, able to make visible the more basal elements of every concept needed to understand human mind functioning.

Therefore, false belief task proves once again to be inadequate for the study of children’s reasoning on the content of beliefs, because of its lack of sensibility to the developmental trajectory of this ability.

 

Children’s knowledge about the mind means more than passing FBt

 

As it was shown in the first part of this paper, the abilities to understand human beings as being different from non-agent objects (Gergely et al., 1995; Lagerstee et al., 2000), as being able to have emotional feelings (Stern et al., van Wulfften Palthe, apud Striano, 2001), to communicate and to intentionally act (Gergely et al., 1997, 1995) appear early in life, before the first birthday. All the complex abilities like being aware of someone’s desires (Gopnik et al., 2000), understanding pretence (Nichols et al., 2000) and deceiving (Carlson et al., 1998) occur beginning with the age of 2. At the same age children are able to change their behavior according to other people’s knowledge state (e.g. if their parents posses or not a certain belief about the state of the world) and are also able to establish the circumstances that determined them. For example, if the parent was not present when the object’s location has been changed, the child points to that location in order to get that object, usually a toy, but s/he does not point to the new location if the parent was present when the exchange of locations was done (O’Neill, 1996). Therefore even at this age children have complex abilities which allow them to understand others as human beings with a mind strongly related to their behavior.

The excessive use of the FBt to determine the presence of these abilities is inadequate, since understanding that people have a mind responsible for their behavior involves more than passing this task (Bloom et al., 2000). A proper assessment of social-cognitive development cannot be reduced to the FBt. It requires a battery of tests (Wellman et al., 2001), sensitive to the inter- and intra-individual differences.

 

Solving FBt means more than knowledge about the mind

 

On the other hand, to pass a FBt requires many abilities and it’s not enough to only have knowledge about the mind (Bloom et al., 2000).

 

Executive functions and FBt

According to Leslie, 3- year-olds difficulties at FBts are due to executive deficiencies (Leslie, 1994; Roth and Leslie, 1998). When reasoning about false beliefs, the Selection Processor module deactivates the well accepted assumptions that beliefs are necessarily true or that everyday reality aspects serve a certain function and every time they occur they have the same prescribed function.

Recent research on executive functions targeted the relation between inhibitory control and FBts performance. Inhibitory control can be defined as the ability to suppress potentially interfering thought process or actions (McCall apud Carlson et al., 1998). This inhibitory control, and generally the executive functions, register an important developmental stage between the ages of 4 and 7 (Perner et al., 1999), a critical interval for the development of knowledge about the mind also. More exactly a significant improvement of children’s abilities to pass FBts and deception task is noticed.

It appears from the structure of many theory of mind tasks that inhibitory control skills are at least necessary for successful performance: proactive inhibition (e.g. stimulus presented earlier interferes with the processing of a stimulus presented later) and coactive inhibition (e.g. two or more simultaneously presented stimuli interfere with one another) (Dempster, apud Carlson et al., 1998). For example, in many FBts children have to predict the behavior of a character, based on the present state of affairs or based on the character’s false representation of reality. But the present state of affairs is usually so vivid and salient in the children’s minds that they cannot inhibit the use of it, even if they have some understanding of false beliefs (coactive inhibition). In the deception task children are usually asked to point to a misleading location (the desired object is not there) to deceive a character that wants the object. Many of the children wrongly point to the real location of the object, failing to this task. But pointing is a gesture with prepotent social significance (you point to an object, to something that exists to draw your own and other people’s attention to it). Therefore, pointing leads to proactive interference (Carlson et al., 1998; Roese, 1997). Eliminating these sources of proactive and coactive interference leads to an improvement of the performances on FBt and deception task (Robinson et al., 1995; Carlson et al., 1998).

Therefore, children find it easier to make a link between behavior that has occurred (searching in the wrong place) and conditions giving rise to false belief than to predict a person’s behavior on the basis of conditions giving rise to false belief (Robinson et al., 1995). There is the same situation for the deceiving conditions: if the social significance of pointing (the child uses an arrow image) and the social intimidation (the cases when the deceived person is not present at the moment of deceiving or when the deceived person is a stranger for the subject) are eliminated, the performances for this task improve (Carlson et al., 1998).

Another condition that improves children’s performances on FBt is the strategic planning (Hala et al., 1996). This requires to actively plan how to deceive another person: the child choose the hiding place, she establishes the route for the footprints that will mislead the deceived person. Planning starts with considering the targeted person’s true belief. Then, when the subject actually plans the deception, the mental representation of this belief is available and presents immediate relevance. Also, the subject has to keep in mind this representation when she works on strategic details. Thus, initially true beliefs, that through the transformation of present reality are becoming false, are the central point along the entire procedure and they maintain their saliency for the subject leading to improved performances (Hala et al., 1996).

Maintaining the saliency determines an improvement of performances not only for deception tasks, but also for FBt. For this task they strengthened the own and other representations through their association with a concrete object: the children were asked to look for the image that corresponds to the mental representation content (a card) and to post it to “mail box” toy (Mitchell et al. apud Cadinu et al., 2000) or they tasted a “sample” of reality (within the unexpected content task when the children were judging that the Smarties tube contains Smarties they received a Smartie to eat it) (Robinson et al., 1995). The mental representations enhanced this way were kept during the task and used at the right moment, even if the present reality was not corresponding to them.

 

Working memory and FBt

At the age of 3-4, children show poor memory abilities when asked to remember a passed event (Templeton et al., 2000). More specifically the children of 3-, 4-, 5- and 6-years-old have limited working memory abilities (Keenan, 1998).

It was shown that at preschool age memory space is approximately one third from the adult one. The development of memory abilities is based on the growing efficiency of working memory (Gathercole and Baddeley, apud Fry et al., 2000). Before the age of 6 the main memory changes have a qualitative nature. Most of them are due to the increasing processing speed (Kail et al., apud Fry et al., 2000). But it seems that, alongside processing speed, experience and the maturational process in progress have also their role.

For this age interval two types of memory limits are specified. In the case of 3-years-olds there are reduced abilities for multiple representations encoding, storing and retrieval (Templeton et al., 2000). The difficulties that appear for processing the content of belief are mainly due to difficulties in accessing relevant information. Memory searching for relevant information is facilitated by the clues offered within the FBt (Freeman et al., 1995). For children older than 4, 5 and 6, that in many circumstances prove the ability to store and retrieve multiple representations, the limits are rather imposed by their abilities to work with these representation in conditions under which they are asked to compare them to each other and to analyze them (Templeton et al., 2000). These deficiencies could be the result of insufficiently developed control processes (Case apud Keenan, 1998) that lead to a reduced operational efficiency. As these control and attention abilities develop, more information and abilities will be processed more and more automatically, allowing the development of informational processing (Keenan, 1998) and therefore raising the working memory capacity (verbal and spatial) (Kail et al., apud Fry et al., 2000).

The strong correlation between working memory test scores and FBt performances suggests a vigorous and stable contribution of working memory to the children’ performances in reasoning on false beliefs (Keenan, 1998). Their capacity to simultaneously hold in mind more aspects would allow connecting them for the purpose of reasoning on the content of beliefs.

Therefore, low performances on FBt are not exclusively due to a conceptual deficit as Theory Theory advocates, rather to psychological aspects of performance components.

 

We can depict five main conclusions, ideas that are milestones for the new coming theory:

1)      There are other tasks, like deception task, that are targeting belief understanding, so there is no basis for some researcher’ desire to restrain the conclusions about this ability solely to those studies that use FBts.

2)      At least part of the developmental trajectory of children’s knowledge about the mind could be manifested as implicit and this could be also the case for belief understanding. Given this situation, classic tasks for the assessment of false beliefs reasoning prove to be inadequate because they are targeted to explicit knowledge.

3)      Also, false belief task proves to be inadequate for the study of children’s reasoning on the content of beliefs, because of its lack of sensibility to the developmental trajectory of this ability.

4)      Since understanding that people have a mind responsible for their behavior involves more than passing the false belief task, a proper assessment of social-cognitive development cannot be reduced to the FBt. It requires a battery of tests, sensitive to the inter- and intra-individual differences.

5)      Low performances on FBt are not exclusively due to a conceptual deficit as Theory Theory advocates, rather to psychological aspects of performance components. Therefore, we should not draw quick conclusions about children’ failures without thinking at the masking effect.

 

 

THE SPINOZAN PROCEDURE OF AFFIRMATION AND NEGATION

 

One of the reasons for which the FBt was created is that the true belief task was considered irrelevant for the children’s abilities to represent the beliefs of another person. This situation is the result of the correspondence between the content of someone’s belief about the world and the present state of this reality. For example, if Sally thinks that the chocolate is in the basket, which in reality is indeed in the basket, then she has a true belief about the location of that object. In this case, when the child is asked to reason about the Sally’s belief, it is impossible to accurately state if his/her answer is the result of the inferences about the character’s belief or the result of the inferences about the state of the world (since the content of both is the same). As a result of this ambiguity regarding the assessment tools for children’s ability to represent the beliefs of another person, the FBt was created. The context created by this task supposes that the content of the character’s belief (“the chocolate is in the basket”) does not correspond to the present state of reality (“the chocolate is in the box”) and so increases the accuracy of the given response assessment (the given response is the result of the inferences about the false belief and not about the reality).

Indeed, the new created task (“the FBt”) increases the assessment accuracy of the children’s abilities to infer the way another person represents reality, but at the same time imposes new processing requirements. One of these additional processing requirements is to establish the falsity of this representation or to deny its veracity.

 

Main assumptions

 

The tasks designed to asses false beliefs, desires, pretence or deception are in fact tasks which asses children’s beliefs about another person’s beliefs (“X believes that Sally believes that the chocolate is in the basket”). According to the Spinozan procedure we can talk about belief when the symbolic representation of reality is treated as if it was true (Gilbert, 1991). The belief has two components: the representation component and the assessment component.

The representation component of believing refers to the existence of meaningful information within a mental system which allows to code and to symbolize a proposition (Gilbert, 1991). In other words, someone’s mental system considers that the information it stores about the objective reality, correctly characterizes the real state of affairs. For example, to believe that “armadillos have four legs” we need to have the information “armadillos have four legs”, about which we think that characterizes the reality.

 

 

 

 

 

 

 

 

 

 

 

 

 

The assessment component, or the attitude (Leslie, 1994; Roth et al., 1998; Perner, 1995), refers to the relation between the representational component and reality, establishing a connection between the representation information and other information about that reality which already exists within the mental system (Gilbert, 1991). 

According to this procedure to deny (or to unaccept) a proposition supposes information processing in two stages: first is the representation stage and the second is the assessment stage (Gilbert, 1991).

 

The Spinozan Procedure Principles

 

The unity principle states the unity of comprehension and acceptance. According to this principle a proposition cannot be understood without being represented as true and this is so because to comprehend a proposition one must “imagine how the world should be granted its truth (Johnson-Laird apud Gilbert, 1991).

The asymmetry principle states that the acceptance (the affirmation) and rejection (the negation) are not mere alternative outcomes of a single assessment process, but rather, acceptance is psychologically prior (Gilbert, 1991). The modular systems characteristics and premises underlie this asymmetry principle. One of these characteristics is that “when stressed, modular information-passing systems with multiple exit capabilities will often show a bias toward prematurely outputting the products of early modules” (Gilbert, 1991). A stressed or resource depleted system should represent propositions (a product of the first module) but should occasionally be unable to assess those representations (a product of the second module) and thus it gives us a partially result. Because affirmation or acceptance is prior to negation or unacceptance, in the case of resource depletion, the entire process stops at the level of acceptance or affirmation.

Research on adults

 

Psycholinguistics research shows that people are faster when decide if a proposition is true than when they have to decide if a proposition is false (Gough apud Cadinu et al., 2000). Clark and Chase think that: 1) all propositions are initially coded as true; 2) then they are compared to the state of the world; 3) if the two representations match, the truth index is left alone; 4) if they don’t match, the truth index is changed to its opposite, false.

Also, social psychology emphasized some of the consequences of people’s difficulties processing invalid information. One of the most important examples is that of the fundamental attribution error (people’s tendency to underestimate external or situational factors and overestimate internal or personal factors when explaining the causes of people’s behavior) (Ross apud Cadinu et al., 2000). For example, if people listen to a speech of a target person who has been assigned to read a certain autobiographical claim, they will assume that the reader’s claim is true. In other words, people will underestimate the situational constraints, in this case the fact that the reader has been randomly assigned to his role, and will tend to assume that the target person intended to read something about him. The same happens when people receive feedback about their abilities for certain task, and later they are told that the feedback was bogus, they will still tend to incorporate the invalid feedback in their self-attribution (Ross, Lepper and Hubard apud Gilbert, 1991). More than that, people tend to believe the bogus feedback even when they are told in advance that it is false (Wegner et al. apud Cadinu et al., 2000).

Research on depleted resource in adults shows that negation is the first operation to disappear. Festinger and Maccoby presented to their subject information which was either consistent or contrary to their attitude on a certain topic. Subjects presented with counterattitudinal statements were more likely to accept them if they were engaged in irrelevant tasks (apud Cadinu et al., 2000). According to the persuasion model, resource depletion affects tired or distracted subjects so that their ability to logically reject the presented material is diminished (Gilbert, 1991). Consequently, in certain conditions, adults find difficult to undo mental operations (or to pass from the truth index to the false one). Generally speaking, it seams more difficult to process information which does not correspond to the state of the world. To represent something as being false, first this information has to be represented as true and then we have to operate an extra cognitive step.

 

Research on children

 

As it was shown earlier, FBts create a context which imposes new processing requirements in addition to the target requirement (to compute the character’s belief content). Some of them are: a) to resolve the proactive interference – the child has to respond with different gestures, like pointing to, which has powerful social significance (you point to an object, to something that exists, to direct yours and others attention to that object); b) to resolve the coactive interference – during these tasks the present reality is so salient and vivid for the subject so that they cannot inhibit mentioning it, even if they have some understanding of beliefs.

Research on executive functions, more specific on inhibitive control, have shown that these abilities register a marked development between 3- and 6/7- year-old. During this period differences occur from age to age (Perner et al., 2000).

Changes in mnestic abilities also occur during the preschool age. There are significant differences regarding working memory capacity between children 3-, 4- and 5-years-old (Keenan, 1998). The explanations for these differences and for the memory development are still incomplete. What is known for sure is that these mnestic deficiencies are characteristic for children 3 and 4-year-olds and that there is a relation between them and diminished abilities to solve interferences (Carlson et al., 1998).

There is the possibility that these supplementary requirements, added to limited processing abilities, create a processing overload. In this situation a prematurely resource depletion occurs, so that the subject cannot finish processing the belief content. FBts requires establishing the belief’s veracity or the belief’s falsity. According to The Spinozan procedure of affirmation and negation (Gilbert, 1991), to represent a false belief, children have first to represent the beliefs as true and then operate an extra cognitive step, the “negativity” step, and compute their opposite. Thus, in situations in which prematurely resource depletion occurs, processing the false belief content ends earlier than necessary, without operating the extra “negativity” step. So, it is possible that children 3-year-olds are able to understand beliefs as internal states, but have difficulties to determine their falsity, mostly in situations in which their cognitive resources are depleted.

The objective of this study was to demonstrate that within the framework of FBts, one of the excessive performance requirements is to process an extra cognitive step, the “negativity” step, which is not relevant for children’s abilities to compute the content of beliefs.

 

 

STUDY 1

 

Hypothesis

Negation, as compared to affirmation, requires an extra cognitive step for information processing. Within the framework of false belief task, changing the consonance belief-reality will determine differences between the time required for negation processing and the time required for affirmation processing, differences specific for the 3-, 4- and 5-years-olds.

Experimental design

It was used a within-subject design.

Independent variable: the consonance between character’s belief content and the objective reality: the incongruity results in a negation situation and the consonance results in an affirmation situation.

Dependent variable: the reaction time for a YES answer and for a NO answer respectively.

Label variable: the age (of 3, 4 and 5 respectively).

 

Method

Participants. Forty-three preschool age children were randomly selected from several kindergartens as it follows: 14 3-year-olds, range 3.0 to 3.11; 14 4-year-olds, range 4.0 to 4.11 and 15 5-year-olds, range 5.0 to 5.11.

The strong development of belief processing abilities and also of the executive functions, especially of the inhibitive control, that characterizes this age interval was the main reason for choosing subjects of the age class mentioned. It is considered that as they pass from 3 to 5-6 year olds, these abilities manifest for the first time and then they develop, thus inter-age differences will be registered.

This study aims to emphasize affirmation and negation in false belief tasks. To make sure that the participants are able to rationalize on beliefs and that the responses they give are not accidental, all the children were first tested for their false belief abilities.

After the initial testing, 28 participants remained in the study as it follows: 6 3-year-olds, range 3.0 to 3.11; 10 4-year-olds, range 4.0 to 4.11 and 12 5-year-olds, range 5.0 to 5.11.

Materials. A tape recorder was used to record the conversation between the experimenter and the participant, so that the reaction time (RT) record to be accurate. Later RT was determined using a sound processing soft (Cool Edit Pro from Syntrilium), that measured the duration between the final point of the E’s question and the starting point of the S’s response (the RT) in seconds and milliseconds. A Compaq Presario notebook, 1200 series was also used, two chocolate candy boxes (the type and the design of those boxes clearly indicated that they are chocolate candy boxes and not another type of boxes) and colored pencils.

Procedure. The tasks used for this study were:

1.    Standard unexpected content task to assess the ability to identify own and others false beliefs, so that we can be sure that children’s responses for the false belief task intended to assess the reaction time for negation and affirmation are not at random. The used task is a false belief one: the classic unexpected content version (Wimmer and Perner, apud deVilliers): a) For true beliefs: acting surprised, the experimenter (E) takes out from the backpack a candy box and asks the subject (S) to name the object. After the S correctly identifies the object the E opens the box and shows the content (chocolate candies) and the children has to identify the content. After the S correctly responds “candies”, E asks the true belief question; b) For false beliefs: acting the same manner E takes out another candy box (that has a different content from the first one). The S identifies the object and the content and then E asks the false belief question (Appendix 1). The subjects pass to the next phase only if they correctly answer these questions.

2.                                                 The changed location false belief task (Wimmer and Perner, apud deVilliers) aims to emphasize the rapidity differences between the affirmation and negation processes in the false belief task context:

E is announcing that they are going to watch a movie that is about a little boy named Christian and his friend The Clown.

The movie starts with the characters’ presentation. Then the S can see that Christian has a book and a small bag that contains colored pencils. Then the boy and the Clown start to color together the drawings from the book. At a certain moment Christian decides to leave the room. He closes the book and he puts the pencils back into the bag, then he leaves the room closing the door after him. During this period of time when the child is gone, the Clown takes the pencils out of the bag and he is coloring for a couple of seconds. Then, he hides the pencils behind a big doll (and so he changes the location) and he puts some rocks in the bag. The next sequences present Christian returning into the room and the movie ends.

E asks S to tell what happened in the movie. Then, S has to answer some questions that are assessing if the child memorized the main events and those questions that target the YES answer and the NO answer (to establish the RT for affirmation and the RT for negation). (Appendix 2)

The RT is recorded.

3. The “tell me YES or NO” game – This game is used to teach the children in the study the significance of the word YES and the significance of the word NO or, if the children already know the significance of these words, to help them remember. There are 6 trials (3 affirmations and 3 negations randomly distributed). If the S fails more than 3 trials s/he is excluded from the study.

The Study 1 procedure sequences were interposed with those of Study 2 and some of the tasks were common for both studies (as it will be mentioned later). I will present the procedure as it was actually used (simultaneously for both studies) and I will go into details only for those tasks that are specific for the study in course. Because we ran simultaneously both studies and because the entire procedure lasted more than 25-30 minutes working with the same subject twice was needed. The second session was not delayed more than 2 weeks because we wanted to avoid major changes in the social-cognitive development.

The subjects were individually tested in a quiet room after they had been invited to participate in the “game”. No child was included in the study against her/his will.

The experimenter has previously spent a couple of days playing with the children with the purpose of getting familiar with them and eliminating possible cases of shyness. Also they were allowed to see and touch the materials used in the study.

The entire procedure was carried on as it follows:

Table 1.

Procedure unfolding

 

Session 1.	Phase I	1. The unexpected content task (for both Study 1 and Study 2)
	Phase II	1. The “tell me YES or NO” game (for both Study 1 and Study 2) 2. The true belief task – AFFIRMATION vs. NEGATION (Study 2) 3. The non-social task – AFFIRMATION vs. NEGATION (Study 2) a)        Recognition sequence; b)       Affirmation/Negation sequence (the first 6 images) 4. The false belief task – AFFIRMATION vs. NEGATION (Study 1). 5. The non-social task – AFFIRMATION vs. NEGATION (Study 2) –  Affirmation/Negation sequence (the second group of 6 images)
Session 2.	Phase I	_  (it starts directly with phase II)
	Phase II	1.        The “tell me YES or NO” game (for both Study 1 and Study 2) 2.       The non-social task – AFFIRMATION vs. NEGATION (Study 2) – Affirmation / Negation sequence (the third group of 6 images) 3.       The false belief task – AFFIRMATION vs. NEGATION (for both Study 1 and Study 2) 4.       The non-social task – AFFIRMATION vs. NEGATION (Study 2) – Affirmation / Negation sequence (the fourth group of 6 images) 5.         The true belief task – AFFIRMATION vs. NEGATION (Study 2)

 

 

Results

Mean RT for the false belief task (AFFIRMATION vs. NEGATION) for each age group is presented in Table 2.

 

Table 2.

Mean RT for AFFIRMATION and NEGATION at the false belief task

 

 

 

 

 

 

 

 

 

Comparing (Wilcoxon Test for two-related-samples) the affirmation RT with the negation RT, for all age groups, revealed that the RT necessary for a negative answer is significantly longer than the RT necessary for an affirmative answer. A significant difference between these RT-s appears for all age groups: 3-year-olds Z(NO/YES) = -2.201 (p = 0.02*, p < 0.05); 4-year-olds Z (NO/YES) = -2.668 (p = 0.008*, p < 0.05); 5-year-olds Z (NO/YES) = -2.936 (p = 0.003*, p < 0.05).

An age (3- vs. 4- vs. 5-year-olds) × type of response (affirmation vs. negation) ANOVA revealed that age has no significant effect on negation in false belief task, F = 0.27, p > 0.05 (p = 0.76). Also, age has an unsignificant effect on affirmation within the same task F = 0.53, p > 0.05 (p = 0.59).

Multiple post-hoc comparisons show a more accurate image of the differences between age groups for both types of respose: affirmation (Table 3) and negation (Table 4).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type of response I (age) J (age) Mean difference Significance     Negation 3-year-olds 4-year-olds 5-year-olds .3294 .2417 .449 .426 4-year-olds 3-year-olds 5-year-olds -.3294 -8.78  .449 .376 5-year-olds 3-year-olds 4-year-olds -.2417 8.778 .426 .376 *significant (p < 0,05)

 

Table 3.

Multiple comparisons between age groups for negation

 

 

 

 

 

 

 

 

 

 

Table 4.

Multiple comparisons between age groups for affirmation

Type of response I (age) J (age) Mean difference Significance     Affirmation 3-year-olds 4-year-olds 5-year-olds 3.611 -0.117 .981 .803 4-year-olds 3-year-olds 5-year-olds  3.611 -0.1536 .981 .620 5-year-olds 3-year-olds 4-year-olds 0.1175 0.1536 .803 .620 *significant (p < 0,05)

 

 

 

 

 

 

 

 

 

 

It seems that even if children are older, they do not significantly differ in RT for both affirmation and negation in the false belief task context. Children age 3, 4 or 5 require, in this study, almost the same amount of time to negate or to accept a proposition about someone else’s belief.

STUDY 2

 

Hypothesis

Multiple processing requests need more cognitive resources and more time resources. Different cognitive resources requests within false belief task, true belief task and non-social task will cause differences for the negation RT and affirmation RT respectively, differences that are specific for the ages of 3, 4 and 5.

 

Experimental design

A within-subject design was used.

Independent variable: type of task (false belief task, true belief task and non-social task);

Dependent variable: the reaction time for a YES answer and for a NO answer respectively for every type of task.

Label variable: the age (of 3, 4 and 5 respectively).

 

Method

            Participants. Forty-three preschool age children were randomly selected from several kindergartens as it follows: 14 3-year-olds, range 3.0 to 3.11; 14 4-year-olds, range 4.0 to 4.11 and 15 5-year-olds, range 5.0 to 5.11.

The reason for choosing these age groups for this study is the same as for the first study.

Children were first tested for their false belief abilities, so that their affirmative and negative answers for the belief tasks to be not at chance. After the initial false belief test, 28 subjects remained in the study: 6 3-year-olds, range 3.0 to 3.11; 10 4-year-olds, range 4.0 to 4.11 and 12 5-year-olds, range 5.0 to 5.11.

Materials. Materials were the same as for Study 1. Additionally a soft specially created was used to run the non-social task. With this soft 4 types of display were activated, 2 were designed to run the task and the other 2 were designed to organize the database. The displays used to run the task are: 1) The “CHOICE” display (ro. “ALEGERE”) – it is used for the first stage where the subjects identify the objects (pictures representing those objects) and their persona picture base is created; 2) The “QUESTIONAIRE” display (ro. “CHESTIONAR) it was used for the second stage of this task that aims to establish the affirmation RT and the negation RT. This display presents the pictures that the subjects have identified at the first stage. The E asks a question that targets the identity of the object and the subjects have to accept (affirmation) or to deny (negation) that identity (Appendix 3).

Procedure.The procedure for Study 2 was interposed with the procedure for Study 1. This was imposed by some features of the participants’ age: 1) Children at this age become bored very fast and to prevent this, a variation of the task is required, and 2) When making tasks more attractive, children are more profoundly involved in solving them and they concentrate for a longer period of time.

The same preparation and experimental conditions have been adopted.

The aim of this study is to emphasize the rapidity differences between the affirmation process and negation process that appear in contexts with different cognitive processing requirements. I have considered that the social-cognitive type tasks (false belief tasks, deception tasks etc.) are more complex and create a problem context that solicits a greater amount of cognitive processing as against other tasks that require only identification and recognition cognitive processes. Therefore, I used different tasks with an increased degree of cognitive processing requirements:

1.    The non-social task – AFFIRMATION vs. NEGATION:

a) Recognition sequence – E presents to S pictures with different objects. For each object a recognition question was asked (“What is this?”). The S has to name that object. If the object is recognized, the picture is sent to the child picture base. The procedure is repeated until the picture base counts 24 items. These 24 pictures are divided into 4 series of 6 pictures each.

b) Affirmation/Negation sequence (the first 6 images) – the pictures are once again presented and this time the E asks, for each picture, a negation or an affirmation question, according to the correspondence between the presented object and the question (“Is this a ……?”). The questions targeted affirmation or negation at random so that sequences of more that 2 affirmations or 2 negations would be avoided.

2.    The true belief task – AFFIRMATION vs. NEGATION: E is announcing that they are going to watch a movie about a little boy named Christian and his friend The Clown.

The first segment is similar with the first segment of the movie from the false belief task. After Christian leave the room, the Clown takes the colored pencils out from the bag and starts to color. After a while he puts the pencils back into the bag, then the child returns and the movie ends

     E asks the S to tell what happened in the movie. Then, the S has to answer some questions that are assessing if the child memorized the main events and questions that target the YES answer and the NO answer (to establish the RT for affirmation and the RT for negation) (Appendix 4).

     The RT is recorded.

3.    A false belief task: the changed location version (Wimmer and Perner, apud deVilliers) – the same as for the Study 1

Results quantification – only the correct answer counted for both social-cognitive tasks (false belief task and true belief task) and non-social task (the affirmation and negation of the propositions regarding the objects’ identity task). The RT was measured for the YES answer (representing AFFIRMATION), and for the NO answer (representing NEGATION) respectively for every type of task.

The entire procedure was carried on the same way as for the Study 1, respecting the same sessions and phases (see Table 1).

 

Results

Mean RT for AFFIRMATION and NEGATION respectively, for each type of task and for each age group are as follows (Table 5):

 

 

Table 5.

Mean RT for Affirmation and Negation for false belief task, true belief task and non-social task

 

Task type   Age group	False belief task		True belief task				Non-social task
	Affirmation	Negation	Affirmation		Negation		Affirmation	Negation
3-years-old	0.48s (SD=0.44)	1.52s (SD=0.95)	0.76s (SD=0.34)	1.29s (SD=0.94)		0.53s (SD=0.16)		0.76s (SD=0.33)
4-years-old	0.48s (SD=0.24)	1.19s (SD=0.78)	0.78s (SD=0.58)	1.25s (SD=0.77)		0.51s (SD=0.21)		0.77s (SD=0.18)
5-years-old	0.48s (SD=0.37)	1.28s (SD=0.85)	0.78s (SD=0.59)	1.17s (SD=0.48)		0.54s (SD=0.21)		0.69s (SD=0.30)

 

 

Comparing (Wilcoxon Test for two-related-samples) the RT, for negation and affirmation respectively, between tasks, revealed significant differences between false belief task and non-social task for negation, for 3-years-olds and as well for the 5-years-olds (Figure 2). It was found out that 3-years-olds require longer time to say NO in the context of false belief task than in the case of the non-social task (ZNO non-social/NO false beliefs= -1. 992, p = 0.046, p < 0.05). Also, the 5-years-olds need more time to negate in the context of the false belief task than in the context of non-social task (ZNO non-social/NO false beliefs=-2.668, p = .004, p < 0.05). For 4-years-olds there wasn’t a significant difference (ZNO non-social/NO false beliefs= -1.260, p = .208, p > 0.05).

For the Affirmative answer there weren’t significant differences along the tasks.

Fig. 3. Mean Negation RT for all task types (false belief task,

true belief task and non-social task), by age.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

An age (3- vs. 4- vs. 5-year-olds) × type of response (affirmation vs. negation) for all task types ANOVA revealed that age has not a significant effect on negation RT (false belief task: F = 0.278, p = 0.76, p > 0.05; true belief task: F = 0.06, p = 0.93, p > 0.05; non-social task: F = 0.27, p = 0.76, p > 0.05). Also, a non-significant effect of age was registered in the case of affirmation for all types of tasks (false belief task: F = 0.537, p = 0.59 p > 0.05; true belief task: F = 0.09, p = 0.91, p > 0.05; non-social task: F = 0.06, p = 0.93, p > 0.05). Multiple post-hoc comparisons have shown that there are not significant differences between age categories for the RT needed for a negation or for an affirmation, by task type. This means that as they grow up, children do not become faster in giving a negative answer or an affirmative answer.

 

 

DISCUSSION AND CONCLUSIONS

 

The Spinozan Procedure for Affirmation and Negation – The objective of this study was to demonstrate that within the framework of FBts, one of the suplimentary performance requirements is to process an extra cognitive step, the “negation” step, which is not relevant for children’s abilities to compute the content of beliefs. Study 1 and Study 2 targeted the asymmetry principle that states the followings: the acceptance (the affirmation) and rejection (the negation) are not mere alternative outcomes of a single assessment process, but rather, acceptance is psychologically prior (Gilbert, 1991). To negate or to deny a proposition supposes a two step information processing: the representational stage during which the subject mentally understands reality and simultaneously accepts this understanding’s veracity and then the assessment stage that establishes if the representation corresponds to reality (certification) or not (unacceptance or negation) (Gilbert, 1991). The results from Study 1 sustain the Spinozan Procedure for Affirmation and Negation, especially the Asymmetry Principle. The significant difference between the time needed for negation processing and the time needed for affirmation processing indicates that these two types of responses do not imply mental operations sets that are just alternatives to a singular evaluation process, but rather affirmation is prior to negation. The negation RT is significantly longer than the affirmation RT which means that there are necessary additional cognitive operations for negation. The significance of this difference is maintained for all age groups (3-, 4- and 5-years-old). This suggests that during the preschool age (3- to 5-years-old) negation is processed according to the asymmetry principle.

The Asymmetry principle also states that in the case of excessive processing requests, the existing resources are depleted so that the system has a premature output. This means that as the subject faces processing requests that do not count for the negation cognitive steps, his resources are prematurely depleted and this stops or delays the negativity step. Study 2 has only partially verified this aspect, more exactly it intended to verify the condition of the delayed negation. Three types of tasks were presented to subjects. It was considered these tasks impose different amounts of processing requests and that this will lead to differences for the response latency (negations and affirmations). The results sustain this idea. Significant differences were registered between the negation RT for false belief task and the negation RT for the non-social task. The processing requirement differences between these two tasks is manifested at two levels: the level of problem context (social context vs. perceptual context) and the level of the problem requirement, what type of reasoning is involved (reasoning on beliefs vs. reasoning on a known object’s identity). The negation RT for the two social tasks did not differ significantly. It is possible that these results are due to the fact that the question requiring the negative answer from the true belief task transformed the reasoning so that it became characteristic for the false belief task. This way those two social tasks were equivalent.

 For affirmation there were not significant differences between the RT’s characteristic for each type of task, except the 5-years-old group. Children from this group required more time to give an affirmative answer as the tasks grew in complexity. It is possible (as the protocol analysis shows) that these differences are due to age features like better justifying for their answers than younger children.

As children grew older their performances for affirmation and for negation did not improve. Studies on the development of processing speed show that this registers a nonlinear evolution with a gradual improvement during preschool age (Cerrela et al., apud Fry et al., 2000). In both studies there was no significant effect of age on processing speed for negation. These results may be due to the initial testing for false belief abilities. Through this test, children’s performances at false belief task were matched, for all age groups. But, as it was earlier shown, this task requires not only false beliefs abilities, but also memory abilities, abilities to solve interference phenomenon. By taking in the study only the children that are able to successfully pass this task, there is the possibility of selecting those children that, even if having different ages, are at the same developmental level for a large area of abilities (processing speed, working memory, solving interference phenomenon). This was reflected in the similar negation RT for all three age groups. The same effects were registered for affirmation in false belief task.

Processing speed - During preschool age the relation between working memory and intelligence is stronger than the relation between processing speed and intelligence (Miller et al., apud Fry et al., 2000). Therefore it is possible that during this period working memory is more important. If this is the situation, then it is possible that the initial test for the false belief abilities matched the children, even from the perspective of memory abilities, which was reflected by the insignificant age effect on the time required for affirmation processing and for the negation processing.

Cross-situation studies – The fact that the significantly longer latency for negation than for affirmation maintains across many tasks (false belief task, true belief task, non-social task) is of major importance. Studies on deductive reasoning (at adult age) show that the responses that require negation are slower (have a longer RT), and that this latency grows in conditions that require activating information that is supplementary to those offered by the problem context (Leahey, 1980). Psycholinguistic studies emphasize that when children no older than 12 (they are not using yet a constant interpretation schema) are processing negation in complex sentences involving clauses, they have to understand first the meaning of that sentence in its affirmative form and only after that its negative form (Singer, 1986). Therefore there is one more proof that the asymmetry principle is respected across situations.

Protocol analysis

Emphasizing the Asymmetry Principle – qualitative data obtained trough unsystematic observation during these studies sustain the idea of asymmetry. In the case of some subjects with higher tendency to verbalize the process of solving the task, some sequences in their answer of the “…yes…no, no…” type were observed. This clearly shows that negation is not simply an alternative to affirmation, but rather they are successively accomplished by the following formula: UNDERSTANDING/ACCEPTANCE, NEGATION.

One of the consequences of the asymmetry principle, according to modular systems characteristics, is that as the information processing requests become more numerous, the time necessary to accomplish them increases presenting also the possibility that in the case of resource depletion the system will offer a premature output. In our case the premature output would be the affirmative answer when the correct one should be negative (Gilbert, 1991). Study 2 shows only that the RT for negation in the complex task (false belief task) is significantly longer than the RT for negation in the simple task (non-social task) for al age groups. Therefore, as the cognitive processing requests increase, the negativity step processing is delayed.

Age differences for affirmation processing – it was found that for 3- and 4-years-olds there are not significant differences between the affirmation RT from all three types of tasks: they do not need more time to make an affirmation in a false belief task than in a true belief task or they do not need more time to make an affirmation in a false belief task than in the non-social task or in the true belief task as compared to the non-social task. This could be a sign that during this age interval children use some rules to make an affirmation that is not necessarily related to quantity of information. It was observed that when they have given an affirmative answer, 3- and 4-years-olds first give the YES answer and only after that they tried to establish the identity of that object or situation. So, even if they affirm (understand and accept), they are not reaching the level where they are able to identify the object. But, relying on some elements that suggest that identity, they make some kind of recognition which allows them to understand and accept. Because 5-years-olds have significantly different RT for affirmation across the tasks (differences that are due to specific processing requests for every task), we could consider that at this age children have a more complex information processing that sustain the responses they are giving.

Predictions on consistency (psychological traits and physical aspects) – Kalish (2002) proves that people are using different inductive strategies when reasoning on physical and intentional phenomenon. Adults, as well as children, have a tendency to predict results consistency in the context of physical causality. A single exposure to a new physical relation was sufficient to predict that this new relation will maintain for the future. Adults see people as having a consistent behavior due to their voluntary and involuntary traits. They predict that a person will continue to manifest the same traits as in the past. But, it seems that younger children do not use information on traits to infer consistency for the human behavior. 4- and 5-years-olds do not predict that a person will manifest a certain behavior in the future only because s/he manifested it in the past. In other words, even if children 4- and 5-years-olds have the capacity to infer the consistency of physical characteristics in the context of physical causality, they do not infer so easily the consistency of psychological characteristics, implying unpredictability for people’s future behavior. More than that, children from this age group have the tendency to overestimate people’s voluntary control over their own thoughts and behavior.

Some errors in children’s reasoning were observed that are due to these characteristics of consistency predictions:

·       errors due to physical traits overgeneralization: in this study some subjects manifested the tendency to predict consistency for objects’ physical traits and to induce them to the entire category (for example if the child’s own home door has a window that allows to see through, the door from the movie has the same characteristic, even if it does not really exist in the movie). Therefore, the wrong answer can be the consequence of this kind of error and not necessarily the consequence of the lack of false belief abilities. These errors limit the information that the child gathers.

·       erorrs due to people’s voluntary control overestimation: Kalish (2002) shows that younger children have the tendency to overestimate people’s voluntary control over their thoughts and behavior and to consider volition a causal mechanism. During the present studies there were situations that seem to verify this. Asking one child how come the character could see the hiding moment (since he had told me that he went in another room to sleep) she replied that Christian dreamed about the entire moment, so that he knows where the pencils really are.

Children’s own experience, as well as other’s behavior observation, determines them to deduce some behavioral regularity (Clements et al., 2000) and, not only once, they manifest the tendency to extend them across various situations, even when they don’t match. These aspects could be the result of a narrowed working memory space that results in depicting only the relation between elements and ignoring the details (even when they are essential for a correct inference) (Keenan, 1998; Templeton et al., 2000). Also, it is possible that the information/the regularities gained from the child’s own experience, which were stated as social rules, are more salient for him/her than the present information. Added to low inhibitory abilities, they can determine erroneous answers, for the false belief task inclusively. Therefore, these are due more to executive limits and less to conceptual deficits.

By studying false belief training, it was shown that preschoolers are more receptive for the feedbacks that explain the correct answer (as against simply stating wrong answer-correct answer) and this facilitated performance improvement (Clements et al., 2000). It is possible that this improvement it is due to the fact that this way they are able to see the relevant information.

Concluding, children’s answers to classic false belief tasks (that require only the character’s behavior prediction) offer us less information about the nature of the basal reasoning. The answers to the false belief tasks lead us to a dichotomist result, which allows us to see only if the children have or have not the ability to reason on the content of beliefs. Analyzing this kind of results we cannot say where on his/her developmental trajectory the child is. As it was shown in the first part of this paper, there are many other abilities that attest children’s abilities as mind readers, abilities that usually does not manifest all of a sudden, but suffer many improvements of a starting rudimentary form. If this is the case for the ability to compute the content of beliefs, as it was suggested, then the actual false belief task is completely “insensitive”.

Also, it doesn’t allow us to determine if the errors are the result of the reasoning; of the source of information or of their quality; of the limits imposed for the working memory or of the executive limits. The “acid test” for false belief abilities has to be the reasoning behind the answer (Clements et al., 2000) and I could add that the task intended to test them should emphasize them in the most natural way. It will be interesting to build experimental tasks, which adequately solicite the presumed social-cognitive abilities and which to be able to emphasize the children’s basal reasoning. It is possible that including the false belief type reasoning in a narrative context and involving the subject in planning and deciding the scenario to be a good idea in this way.

Generally, the aim of this study was to demonstrate that the false belief task is inadequate as an assessing tool for children’s abilities to reason on the content of beliefs. Specifically, this study tried to demonstrate that negation requires an additional cognitive step, which can be skipped in conditions of excessive processing solicitations. The first part of this specific objective was demonstrated. Regarding the second part, the results only partially verify it. Longer RT for negation in false belief task as against the cognitive-perceptual task could be explained through multiple processing requests imposed by the first type of task. Other studies could be done to demonstrate if complete resource depletion leads to a premature affirmative answer when a negative one would be the correct one, as the Spinozan Procedure states.

References

 

Bloom, P., & German, T.P. (2000). Two reasons to abandon the false belief task as a test of theory of mind. Cognition, 77, B25-B31.

Cadinu, R.M., &Kiesner, J. (2000). Children’s development of a theory of mind. European Journal of Psychology of Education, 15, 2, 93-111.

Carlson, S.M., Moses, L.J., & Hix, H.R. (1998). The role of inhibitory processes in young children’s difficulties with deception and false belief. Child Development, 69, 3, 672-691.

Clements, W.A., Rustin, C.L., & McCallum, S. (2000). Promoting the transition from implicit to explicit understanding: a training study of false belief. Developmental Science, 3:1, 81-92.

de Villiers, J. (2000). Language and theory of mind: what are the developmental relationships? In Simon Baron-Cohen et al. „Understanding other minds – Perspectives from developmental cognitive neuroscience”, Oxford University Press.

Flavell, J.H. (1999). Cognitive development: Children’s knowledge about the mind. Annual Review of Psychology, 50: 21-45.

Flavell, J.H. (2000). Development of children’s knowledge about the mental world. International Journal of Behavioral Development, 24 (1), 15-23.

Freeman, H.N., & Lacohée, H. (1995). Making explicit 3-year-olds’ implicit competence with their own false beliefs. Cognition, 56, 31-60.

Fry, A.F., & Hale, S. (2000). Relationships among processing speed, working memory, and fluid intelligence in children. Biological Psychology, 54, issues 1-3.

Gergely, G., Nadasdy, Z., Csibra, G., & Biro, S. (1995). Taking the intentional stance at 12 months of age. Cognition, 56, 165-193.

Gergely, G., & Csibra, G., (1997). Teleological reasoning in infancy: The infant’s naive theory of rational action. A reply to Premack and Premack. Cognition, 63, 227-233.

Gilbert, D.T. (1991). How the mental systems believe. American Psychologist, 46, 2, 107-119.

Gopnik, A., Capps, L., & Meltzoff, A.N. (2000). Early theories of mind: what the thory theory can tell us about autism. În Simon Baron-Cohen et al. „Understanding other minds – Perspectives from developmental cognitive neuroscience”, Oxford University Press.

Hala, S., & Chandler, M. (1996). The role of strategic planning in accesing false-belief understanding. Child Development, 67, 2948-2966.

Kalish, Ch.W. (2002). Children’s predictions of consistency in people’s actions. Cognition, april.

Keenan, T. (1998). Memory span as a predictor of False Belief Undersatnding. New Zeeland Journal of Psychology, 27, 2.

Kuhn, D. (2000). Metacognitive Development. Current Directions in Psychological Science, 9, 5.

Lagerstee, M., Barna, J., & DiAdamo, C. (2000). Precursors to the development of intention at 6 months: Understanding people and their actions. Developmental Psychology, 36, 5, 627-634.

Leahey, Th.H. (1980). A chronometric analysis of simple deductive reasoning. The Journal of General Psychology, 102, 225-232.

Nichols, S., & Stich, S. (2000). A cognitive theory of pretence. Cognition, 74, 115-147.

O’Neill, D.K. (1996). Two-year-old children’s sensitivity to parent’s knowledge state when making requests. Child Development, 67, 659-677.

Perner, J. (1995). The many faces of belief: reflections on Fodor’s and the child’s theory of mind. Cognition, 57, 241-269.

Perner, J., & Lang, B. (1999). Development of Theory of Mind and Executive Control. Trends in Cognitive Sciences, 3, 9, 337-344.

Ritblatt, S.N. (2000). Children’s level of participation in a false-belief task, age and theory of mind. Journal of Genetic Psychology, March, 161, 1.

Robinson, E.J., & Mitcell, P. (1995). Masking of children’s early understanding of the representational mind: backwards explanation versus prediction. Child Development, 66, 1022-1039.

Roese, N.J. (1997). Counterfactual Thinking. Psychological Bulletin, 121, 1, 133-148.

Roth, D., & Leslie, A.M. (1998). Solving belief problems: toward a task analysis. Cognition, 66, 1-31.

Singer, M.S., (1986). A developmental study of children’s comprehention of negation in complex sentences involving clauses. Journal of Genetic Psychology, 148, 189-195.

Slater, A., Hayes, R., Brown, E., & Quin, P.C. (2000). The role of facial orientation in newborn infants’ preference for attractive faces. Developmental Science, 3, 2, p181.

Striano, T. (2001). From social expectations to social cognition in early infancy. Bulletin of Menninger Clinic, 65, 3, p 361.

Templeton, L.M., & Wilcox, S.A. (2000). A tale of two representations: the misinformation effect and children’s developing theory of mind. Child Development, 71, 2, 402-416.

Trainor, L.J., Austin, C.M., & Desjardins, R.N. (2000). Is infant-directed speech prosody a result of the vocal expression of emotion? Psychological Science, 11, 3.

Wellman, H.M., Cross, D., & Watson, J. (2001). Meta-Analysis of Theory of Mind Development: The Truth about False Belief. Child Development, 72, 3, p655.

Woodward, A. (1998). Infants selectevly encode the goal of an actor reach. Cognition, 69, 1-34.

 

 

 

Appendix 1

 

If right now your mother(or your friend) would come and would see the box, what would she/he believe that is in the box?

 

 

Appendix 2

 

Where did the boy put the pencils before he left?

Did the boy see what did the Clown with the pencils?

Were are the pencils right now?

The negation question: Is the boy thinking that there are rocks in the bag?

The affirmation question: Is the boy thinking that there are pencils in the bag?

 

 

Appendix 3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                    “CHOICE” display                                 “QUESTIONAIRE” display

 

 

Appendix 4

 

Where did the boy put the pencils before he left?

Did the boy see what did the Clown with the pencils?

Were are the pencils right now?

The negation question: Is the boy thinking that there are rocks in the bag?

The affirmation question: Is the boy thinking that there are pencils in the bag?