Booksummary with Cognitive development: The learning brain by Goswami - 1st edition


What does cognitive development mean according to Goswami? - Chapter 0 (1)

Cognitive development includes the development of:

  • attention

  • learning

  • memory

  • reasoning

  • language

  • concepts

This development starts as soon as the baby leaves the mother's belly. However, recent research has shown that parts of cognitive development already start in the womb. For example, memories of the mother's voice during pregnancy are already stored in the memory.

Why can cognitive neuroscience be seen as a new era?

Cognitive psychology sees cognition as concepts and ideas in the mind. These are called cognitive representations. It is believed that these representations are both discreet and symbolic.

Due to technical progress, we can now display images of active areas of the brain during a certain action. We can now see what the brain does when it is, for example, solving a problem. Brain research in children can currently be carried out by three techniques:

  • Electroencephalography (EEG): Sensitive electrodes are placed on the skull. These electrodes detect the electrical brain activity. A disadvantage of this technique is that localization is very difficult. However, the technique is very accurate.

  • Functional magnetic resonance imaging (fMRI): an fMRI scan measures the changes in blood flow in the brain. When an increased blood flow to a certain brain area is observed, this means that the water distribution changes. The fMRI gives a blood oxygenation level dependent (BOLD) value. The technique is less accurate than EEG, but the localization is much better.

  • Functional near-infrared spectroscopy (fNIRS): the quantity of hemoglobin in the brain tissue is examined. This indicates changes in oxygen supply. Thus changes in blood supply can also be measured. The localization is better than the EEG-technique and the fNIRS is more accurate than an fMRI scan. Also, with this technique, a child does not have to lie in a noisy magnet, as is the case with an fMRI scan. The disadvantage is that the accuracy is not as good as EEG and the location capacity not as good as fMRI.

Most known neurological examinations have been carried out in adults. So, we know most of the systems that have already been developed with regard to linguistic, perceptual and reasoning tasks. Yet research is increasingly being conducted among children. We know that most brain cells develop before birth (around the seventh month of pregnancy most neurons are present). The environment in the womb can affect later cognitive development. For example, excessive alcohol consumption has an irreversible negative effect on brain development, which affects future arithmetic cognition.

After birth, brain development mainly consists of the growth of connections between neurons. This is called synaptogenesis. This makes the child's brain twice as large in the first year of life. Information is passed between brain cells via low voltage electrical signals (via the synapses).

The primary sensory systems are the first to develop. The higher-level association areas mature later. One of the last areas of the brain going through the maturation process is the prefrontal cortex.

What are two important developmental questions?

  1. What is developing? This question is investigated by observing the cognitive abilities of children during a certain time. Because of this we know that the sensory- and motor cortex develop earlier than language- and spatial areas and that the prefrontal cortex develops last, far into adolescence and early adulthood. The order of brain development corresponds very well with Piaget's stages of development.

  2. Why does development pursue its observed course? To formulate an answer to this question, we need causal reasoning for observed cognitive changes. Experimental research is suitable for this. In the future we will also receive causal explanations from neuroscientific research.

What are the two important explanatory systems?

Traditionally, there are two explanatory systems for explaining cognitive changes in children. The first system focuses on the idea that fundamentals of learning or reasoning are applied to all cognitive domains. This is called the domain-general explanation of cognitive development. The second system states that the development of cognition arises bit by bit, at different times across different domains. According to this view, cognitive development is domain-specific.

The knowledge that we have influences our cognition. The two explanatory systems described above are both structured differently, but they are not mutually exclusive. This book describes that some types of learning match with the first explanatory system and others with the second. The book focuses more on the first developmental question (what is developing?) and less on the second question (why does development pursue its observed course?). The reason for this is that findings on the first question are generally certain, while opinions on the second question may differ.

How do infants learn and what disabilities are associated with learning?

Infants and toddlers can learn in many different ways. Some examples are learning by imitation, learning by analogies and explanation-based learning. The last-mentioned learning method is asking "why?" questions. Because of the focus on causal information, children can explain, predict and ultimately control events.

What do deduction and induction mean?

Deductive reasoning starts at an early age. Deductive reasoning is reasoning based on examples. In n cases, event X leads to Y. In n cases, event A leads to B. You can investigate this by, for example, changing A to X and seeing if B changes to Y.

Inductive reasoning is observed even at a younger age. We reason inductively when we draw conclusions that are not necessarily deductively valid. We generalize on the basis of a well-known example.

Both in domain-general as well as in domain-specific explanation systems causal deduction can be observed. However, the ability of making causal implications appears to be domain-general.

What is the difference between innate and acquired cognition in children?

In addition to the explanatory models as mentioned earlier, there is also the nature-nurture debate. Should development be explained in terms of genes or in terms of an enriched environment? Research has shown that even structures that rely heavily on genetic influences can be altered based on environmental influences. Gene expression is therefore controlled by the environment. This also means environment within us, such as brain tissue. An important question within cognitive developmental psychology is how genes and environment interact with each other, thus creating development.

Piaget's theoretical framework has long been very important in research into children's cognition development. He distinguished three stages:

  1. Sensory-motor stage: cognition is based on actions.

  2. Concrete operations: cognition is based on the symbolic understanding of concrete objects and their mutual relationship.

  3. Formal operations: cognition is completely separate from the concrete world and is described as hypothesis testing and scientific thinking.

What kinds of knowledge are central to cognitive development? - Chapter 1 (1)

Central to the cognitive development of humans are knowledge about the physical world of objects and events, knowledge about the social cognition of the self and others, and knowledge about various things in the world, or conceptual knowledge. These domains can be described as naive physics, naive psychology or naive biology. Children must understand social cognition (to interpret and predict people's behavior on the basis of psychological causation) and they must distinguish between different things that occur in the world. The cognitive development depends on the development of perception, memory, attention, learning and reasoning. After all, all of these processes are needed for most parts of cognition.

It was once thought that cognitive development took place late in life. Piaget, for example, stated that the full object concept (understanding that an object is something that remains, even if it is out of sight) was only present around the age of 18 months. However, this appears not to be the case. Very young children are already developing cognitively by simply looking passively at how things are happening in the world. In addition, they learn from the consequences of direct actions.

Three types of learning are already present at an early stage of development. The first type is associative learning; in the womb, a baby can already make connections between events that are associated. The second type is imitation learning, which is important in the development of social cognition. The third type is explanation based learning that allows infants to make connections between cause and effect. They do this not only by tracing causal relationships, but also by constructing causal explanations for phenomena based on the knowledge they already have.

How can the infant memory be studied?

Bushnell et al. (1984) studied the memory of infants (aged three and seven weeks) by letting parents actively present a stimulus for a period of two weeks (simple shapes in certain colors mounted on a wooden paddle). After two weeks, the researcher offered the child pictures that sometimes differed in color and / or shape from the stimulus that the parents had shown during the two weeks. This research showed that the infants remembered information about the shape, size and color of the objects presented.

Cornell (1979) did a similar study on children from five to six months old and presented photographs of human faces in addition to geometric shapes. Cornell identified recognition by measuring how long a child looked at figures. It was assumed that children would longer look at new stimuli. It turned out that for every set of stimuli children preferred the new stimuli. This means that they recognized the old stimuli. Cornell used a small reminder for the test phase; he quickly repeated the old stimuli. However, this did not appear to influence the results. Because the stimuli were rather abstract, but were still remembered for two days, it can be stated that there is a good development of the recognition memory in young children.

The working memory or short-term memory is the capacity to retain information for a short period of time. The memory span of children aged 5, 7 and 12 months was tested by Rose et al. (2001) by offering stimuli in sets of one, two, three or four items and then pairing them with new items. The number of items that a child recognized was indicated as a memory span. Also, primacy and recency effects were studied. The researchers found that the memory span increases with age and that there is a recency effect for each age. Cornell and Bergstrom (1983) found a primacy effect in children of seven months.

Clifton and colleagues (1990) investigated the memory for events and discovered that six-months-old can remember events well and for a long period of time. This was tested by making the children participate in an experiment at the age of six months. The children had to reach for a Big Bird that made a rattling noise in the light as well as in the dark. Two years later the children were brought to the same laboratory room with the same experimenter to test their memory. They showed little explicit recall of the experiment they had undergone at the age of six months. This was because they showed no preference for the Big Bird compared to other puppets or for the rattling sound compared to other sounds. They did, however, have implicit memories, because they did reach for the Big Bird in the dark without instruction. The control group did not do this. Furthermore, they showed less stress in the dark condition than children from the control group.

Another way to study the memory of events is by using response and reward. With this technique the memory for causal events is examined. This was done by Rovee-Collier et al. (1980) by attaching a piece of string to the ankle of a child. When the child kicked his legs, an attractive mobile was activated above the crib. The children had to learn that kicking caused the movement of the mobile. The memory was tested by measuring how much a child kicked when being put back to the same crib with the mobile after a while. Children of three months old do not forget about it after a period of two to eight days. After fourteen days, however, they completely forgot. As the time between the learning phase and the testing phase lasts longer, these children also forget the specific details about what the mobile looks like and respond to it as if it were a new stimulus. At the same time, as time goes by, the environment (what kind of crib, color, etc.) becomes more important in recognizing the test situation and the contingency of the action and the reward. Details of the learning condition are therefore cues for the reminder of the test situation.

Additionally, a reactivation paradigm was used in this study. Here a reminder is given for a previously learned but apparently forgotten memory that makes this memory available again. In this study, this was done by moving the mobile for three minutes. Infants aged three months show a complete memory with a reminder after 14 and 28 days. Two-month-old infants only after 14 days. Infants of six months hold this memory for at least three weeks. Children can, therefore, develop long-term memories from a very young age and retrieve those memories using the same cues as in adults.

In eleven-month old infants their long-term memory for events causal (causal events) was examined, by using delayed imitation. This involves the degree of imitation after seeing a certain behavior if there has been some delay between seeing this behavior and its imitation. The children remembered it for at least three months. This only happened when it was about causal events and not about non-causal events.

The implicit or procedural memory is the automatic memory that cannot be expressed verbally. For the memory to be explicit or declarative, the past must be remembered and it must be thought about: this requires consciousness. It is assumed that children can only develop an explicit memory if they also have verbal skills (infantile amnesia), but this does not seem to be true.

How can perception and attention be studied in infants?

For a long time, it was thought that young children play a passive role in selecting visual stimuli. However, it is now thought that a baby's visual world is an active environment over which he has no control. To be able to keep track of all this, a baby must create expectations for events that can be predicted so that they can determine their behavior. Attention and perception in infants are therefore measured by looking at their expectations.

Haith et al. (1988) investigated whether infants could create expectations by showing pictures in a logical order and in a random order. Infants showed a faster reaction time and made more eye movements to the previous picture if the pictures were presented in logical order than if they were presented randomly. At the age of 3.5 months, babies are therefore already in control of their own perceptual (attentional) activity.

Gilmore and Johnson (1995) showed that infants by the age of 6 months can also check their visual attention over delays of 3-5 seconds. They demonstrated this by offering a cue to the left or right of the central point that the infant should look at and then see if the infant had a tendency to look at the side where the cue was presented if a picture was presented 3 or 5 seconds later. The infants held a spatial presentation of the cue in their head and used it to plan their later eye movements.

The visual preference technique examines whether an infant can distinguish between two objects or figures. It is assumed that if an infant can do this, he will look at one of the objects for longer because he has a preference for one. If he cannot do it, he will stare at one object as much as at the other. But if he spends the same amount of time on both objects, it could also mean that he finds them both equally interesting. It is not known for sure whether or not an infant can make the distinction. The habituation paradigm can offer a solution for this. If the same stimulus is repeatedly presented to an infant, the time he looks at the object decreases. If a new stimulus is then presented and the infant looks at the stimulus (dishabituation) for a longer period of time, then he is able to distinguish between the two different stimuli.

Soon after birth, children are able to match perceptual information from different sources (cross-modal perception). Meltzoff and Borton (1979) did an experiment with dummies that showed that 1-month-old babies preferred the picture of the dummy with the texture they had just felt in their mouths. This means that infants can make a cross-modal connection between touch and vision at this age. Spelke (1976) and Dodd (1979) discovered that infants can connect auditory and visual stimuli at an early age by offering films with a corresponding sound or another sound that does not correspond with the image. The children preferred to watch corresponding videos and were a bit confused if the image did not correspond to the sound.

Habituation can also be used to see if a baby realizes that different objects belong to the same category. If two new stimuli are offered, one belonging to a known category and the second to an unknown category, the infant will look longer at the stimulus that belongs to the unknown category if he has mastered this skill. Infants aged three and five months old already have this skill. Children form a prototype (prototype formation) of the known shape and to compare the new stimulus with. This is an important cognitive process. It ensures that as much information as possible can be stored with as little cognitive effort as possible.

Rosch (1978) claims that we categorize the world based on the collective appearance of some characteristics (such as wings and feathers are typically birds). Seeing a pattern in the common occurrence of characteristics is the basis for making prototypes and therefore the basis for conceptual representation.

Younger and Cohen (1983) demonstrated with pictures of cartoon animals that ten-month-old babies were already sensitive to the common occurrence of certain characteristics.

Younger (1985) constructed a study showing that when infants were offered pictures of drawn animals in which all possible combinations of the length of the neck and of the legs were presented together, they formed a prototype with average values ​​(ie average length of the neck and legs). If the characteristics were always presented together (i.e. long legs and short neck and vice versa), two prototypes were created.

Younger's findings indicate that infants use statistical learning: they learn about statistical patterns, namely the co-occurrence of certain characteristics. Krikham and colleagues (2002) did an experiment with geometric shapes in which certain pairs of objects always followed each other (for example, a green cross was always followed by a yellow circle). They demonstrated that infants can also learn about the structure of the environment at an abstract level. This type of statistical learning is also available in other domains, such as the auditory domain.

The visual world and its perceptual structure

In addition to forming prototypes of statistical characteristics of organisms and objects, it is important for cognitive development that infants can detect regularities between different events. This is usually described as the relationship between objects (such as: a child pushes the car). This also concerns spatial relationships (above, below, etc.) and quantitative relationships (more than, less than, etc.). For research into this, the violation of expectation paradigm is used, whereby stimuli are offered that conflict with typical regularities in their relationship with objects, creating physically impossible events.

In answering the question of whether an infant is able to see spatial relationships, habituation is used. Hereby an infant first "learns" a spatial relationship. After this, a new spatial relationship is presented. If children look at the new spatial relationship longer, they will be able to distinguish between the perceptual structure of different spatial relationships.

Baillargeon and his colleagues investigated whether five-month-old infants realized that a tall (long) rabbit had to be partially visible if it was walking behind a short wall from left to right. The infants were first introduced in the habituation phase with a short (small) or a tall (long) rabbit walking behind a wall from left to right. As soon as the rabbit passed behind the wall, it disappeared and then reappeared on the right. The middle part of the wall was made shorter in the test phase. The small rabbit could still pass the wall without being seen, but the large rabbit should be visible as it passes the lowered part of the wall. In the experiment this did not happen and the tall rabbit also remained invisible behind the wall. Infants spent a longer time staring at the passing of the tall rabbit in the test phase. This suggests that they understand the spatial relationship between the rabbit and the wall and realize that something is wrong. This is already the case from the age of 3.5 months.

The same researchers also investigated the memory for spatial locations of infants (1988). They presented two locations. A visually attractive object was placed at location A. Two screens were then placed in front of the locations and another visually appealing stimulus was offered to distract the child. After this, the attractive object was obtained from location B. Infants who watched this event for a longer time make the connection that this object cannot come from location B, because it was first located at location A.

Children can therefore remember the location of an object without seeing it: They have understood that an object persists even when it is out of view. Infants from eight months old can keep these spatial memories for up to 70 seconds.

McKenzie seated infants aged six to eight months on their mother's lap behind a semi-circular news desk. At various locations an event was shown that excited the child (an adult person who comes from behind a desk and does 'peekaboo'). The locations where the event was about to take place were first marked with a white ball. Children quickly learned that there was a relationship between the occurrence of an event and the white ball and could therefore predict where the event would take place. The spatial position coding was thus done via the relationship between external signs and the position of the event (allocentric) and not egocentrically, whereby the relationship between one's own position in space and the position of the event is considered.

Object permanence is the idea that an object persists even when it is out of sight. Five-month-old infants were tested for this and stared longer at the impossible condition, in which the object did not form an obstruction for rotating the screen. This means that they understood that the object continued to exist and should therefore actually form a blockade. Further research also varied in the size of the blocking object and the material (sponge vs. wooden block). Infants could use both spatial and physical characteristics to make predictions about whether or not an object would block the screen.

In another study, Baillargeon (1986) demonstrated that infants stared longer at an impossible situation with a car. In the habituation phase, a car was driving off a ramp and a part of its path was hidden by a screen so that the child could not see what was happening behind it. The car then reappeared on the right side of the screen. In the test phase it was shown that there was a box behind the screen that should block the path of the car. The screen was placed in front of the path and the car still appeared on the right side of the screen, while that is actually not possible.

Another study into the relationship between the size of a cylinder and its collision force (a cylinder was rolled into an object which caused the object to move) showed that infants of 6.5 months old and 5.5 month old (girls) are able to use collision-related reasoning about the size and distance relationships in what they see.

Infants aged 6.5 months spend significantly more time looking at an impossible situation where support and stability play a role. A box is placed on a platform and is pushed more and more to the right, so far that at a given moment it is placed 85% over the edge. The box does not fall, while it should fall off. Infants spend significantly more time on this situation than on a reliable situation (30% or no slope). Infants from 5.5 to 6 months cannot make this distinction. Baillargeon et al. believe that these children think that contact with the platform itself provides stability. Experiencing the physical environment plays a role in this. From the age of six months, infants can independently sit in a high chair. They can then see for themselves what happens to objects falling down.

The continuity principle means that objects continue to exist in time and space. This can be investigated by making use of events in which containment plays a role. Baillargeon and colleagues investigated this by showing infants a long and short container in which they put a cylindrical object. The object as a whole could be in the long container, but in the short container, part of the object should normally protrude above the edge.

If an impossible situation is created in the test phase by allowing the cylinder to disappear in its entirety into the short container, children will only show increased looking times at the age of 7.5 months. This is surprising. Infants from the age of 4.5 months are already able to use the relative heights of the object and the container as a cue in a highly similar occlusion condition. Baillargeon and Wang (2004) suggested that children view containment events as different from occlusion events. Infants also see containment events different from events where something is covered because they only look at impossible events that are related to covering an object at the age of twelve months. It is stated that children sort physical events into specific categories and that they learn how to structure each category separately. Physical reasoning is therefore developing in phases.

There is criticism about the use of habituation, visual preference, and violation-of-expectation techniques when studying cognitive processes. These paradigms have been developed for the study of sensory and perceptual processes and not for cognitive processes. According to the critics, it is not possible to produce perceptually identical but conceptually different stimuli for habituation paradigms. Perceptual mechanisms such as novelty, scanning, etc. may also explain longer viewing of certain objects. Haith (1998) claims this and he refers to evidence in neuroscientific research in monkeys. This research indicates that the same neurons are both active when an object is visible and when an object is no longer visible. Haith suggests that this activity may form a neural mechanism for degraded sensory representations. However, recent neurological research suggests that when an individual sees an object, many more brain regions become active (for example, the brain regions that link an object to associating objects and the regions that help reach for the object), which slightly contradicts Haith's arguments. Goswami does not draw a clear conclusion about who is ultimately right.

How are objects processed according to cognitive neuroscience?

EEG is a way to measure electrical activity in the brain by sticking electrodes on the scalp and was used by Kaufman et al. (2003). Kaufman showed infants a train going into a tunnel and then showed them one of the following options:

  1. a hand lifted the tunnel and the train appeared (expected appearance event);

  2. a hand lifted the tunnel and there was no longer a train (unexpected disappearance event);

  3. the train leaving the tunnel, and a hand lifted the tunnel and there was still a train (unexpected appearance event);

  4. the train leaving the tunnel, and a hand lifted the tunnel and there was no longer any train (expected disappearance event).

The six-month-old infants watched an unexpected disappearance event for longer than an expected disappearance event, but there was no difference between the two appearance events. Kaufman measured an increased EEG activity on the right side of the brain with a peak of 500 ms after the tunnel was lifted in an unexpected disappearance event. He suspected that this activity was caused by the brain trying to hold an image, even though that image is not available in the field of view. The higher activity could also be a response to an unexpected event.

If the EEG was measured under the appearance conditions, no increased activity was measured because, according to Kaufman, no effort had to be made to create an image of the train. The train was already in sight. The increased activity was therefore not caused by an unexpected event, because even with the unexpected appearance condition no increased activity was found.

There are two neural pathways along which visual stimuli are processed. The first is the dorsal path, in which spatial and temporal information is processed. This is important when processing information that might require action and is also called the "where" path. The second path is the ventral path (or the "what" path), which is important in processing information that is needed to identify unique objects, such as the color. In addition, the ventral path is important when processing faces. When interpreting infant's looking experiments, the idea that children use these paths should be taken into account. It depends on the objects that are used how the information is processed. Objects to be grabbed at or objects that cause another action are processed via the dorsal path and objects that do not require this are processed via the ventral path. It must also be questioned whether this really influences the research results in certain experiments.

What is the relationship between intelligence at a later age and measurements of learning, memory, perception and attention at an early age?

Bronstein and Sigman (1986) conducted a meta-study that showed that the speed of habituation is related to the degree of intelligence at a later age. Sigman et al. (1986) found in another experiment that when infants aged zero to four months old need to look at a stimulus for a longer time, they score less on intelligence tests later in childhood.

A preference for new stimuli is a significant predictor of intelligence at the age of three. Rose and Feldman (1995) reported that the visual recognition memory was the best predictor of intelligence at the age of eleven.

Central to the cognitive development of humans are knowledge about the physical world of objects and events, knowledge about the social cognition of the self and others, and knowledge about various things in the world, or conceptual knowledge. These domains can be described as naive physics, naive psychology or naive biology. Children must understand social cognition (to interpret and predict people's behavior on the basis of psychological causation) and they must distinguish between different things that occur in the world. The cognitive development depends on the development of perception, memory, attention, learning and reasoning. After all, all of these processes are needed for most parts of cognition.

What is perception? - Chapter 2 (1)

Perception refers to the aspects of the visual system. Examples are perception of color and depth. Perception is almost always directly linked to cognition, and this also appears to be the case in children. Studies of imitation have demonstrated a very early link between perception and cognition.

How does imitation in young children work?

From a very young age, babies are able to imitate certain gestures of adults. In an experiment by Meltzoff and Moore babies were seated in a dark room. For 20 seconds a light illuminated the face of an adult who made a certain gesture. For example, he stuck out his tongue. Then the light was dimmed and the babies were filmed in the dark for the next 20 seconds. This session was then repeated. A researcher who had no idea of ​​the design of the experiment recorded the number of gestures of the baby. As time passed by, a baby - who had seen an adult do this several times - stuck his tongue out more and more. The conclusion was that the baby had imitated the adult.

Meltzoff and Moore argued that successful imitation required representational capacity. A number of cognitive skills are required for successful imitation to take place. After all, children must represent the action of the adult, retain this representation when the adult is no longer visible and they must also discover how to produce the gesture with their own facial muscles. Nowadays, neurological research has shown that children possess certain mirror neurons. These neurons become active when someone carries out an action, he or she sees someone else carrying out an action, or when he or she thinks about a certain action.

How do infants perceive causality?

Some experiments have already provided evidence for the idea that infants can represent causal relationships, for example the violation-of-expectation experiments. However, the causal events in such experiments happened out of sight. Experiments have also been carried out in which the causal events occurred in full view of the infant.

Collision events provide a useful set of events for such experiments. For example, when one billiard ball bumps into the other, this second ball is set in motion. This is a good example of a cause-effect relationship. Leslie and Keeble (1987) were interested in the understanding of children at the age of six months about launching events. These are events in which something is launched, so when an object seems to touch another object and thereby sets it in motion. In the experiment they showed the infants two films. In one film, a red block collided with a green block, after which the green block was set in motion immediately. In the other film, the red block again moved towards the green block and made contact with it, but the green block only started moving after a delay of half a second.

Leslie and Keeble claimed that if the infants in the first movie observed a causal relationship, they showed more dishabituation when the opposite was shown than infants in the delayed launch condition. The researchers claimed that children were interested in the mechanical structure of the launch. Leslie's description of the launching event in mechanical terms gives an example of agency: the understanding of instruments in events.

How are living relationships perceived?

The perceptual system can confuse adults by showing a certain object as 'alive' while it is not alive. In an experiment by Michotte (1963), adults were shown a few simple screens on which certain moving geometric shapes were labeled as animate: "he is trying to get over there." The same results emerged in an experiment by Tremoulet and Feldman (2000). There were three screens in their experiment. On one screen, a dot moved in a certain direction for 375 ms and then changed direction for 375 ms.

On another screen, a rectangle moved in a certain direction for 375 ms and then changed orientation when it changed direction and speed. In the control condition, the rectangle did not change orientation when it changed direction and speed. The result of the study was that adults saw strong perceptions of animacy in the first two conditions, but not in the control condition.

An intentional stance is the attribution of mental causes such as beliefs, desire or goals as the basis for action. This was found in infants in the research of Gergely, Nádasdy, Csibra and Bíró (1995).

What are cross-modal cues for perceptual structure?

Sekuler, Sekuler and Lau (1997) demonstrated that a sound can cause perceptual reorganization of an ambiguous motion display. Adults looking at a computer screen on which two identical disks moved from opposite sides towards each other and then passed each other thought they saw the disks streaming through one another. However, when a noise was presented when the two discs coincided, the adults noticed the discs bouncing off one another. They even perceived this when there was no collision.

Scheier et al. (2003) investigated whether this observation also applied to babies. In their experiment they investigated three groups of children between the ages of four, six and eight months. The infants watched two yellow discs appear in the corner of the screen, move past each other, coincide without stopping and continue until they reached each other's starting point. The children heard a tone at the point of coincidence. It was predicted that the tone would cause a collision perception. This event was repeated until the children were habituated. Scheier et al. claimed that if the children perceived an illusion of collision, they would become dishabituated in the later two tests, but not in the earlier ones. This was exactly what they discovered, but only for the infants between the ages of six and eight months.

The researchers claimed that the behavior of six-month-old infants is becoming more flexible and voluntary.

Are there separate causal frameworks for mechanical agency and human agency?

Evidence has been found that young children distinguish between mechanical sources and living sources of movement. In an experiment by Spelke, Philips, and Woodward (1995), seven-month-old infants were tested on their principle of contact as a cause of movement for inanimate objects versus people. The result was that the infants could indeed make a distinction and that they know that people can move and make contact themselves.

Meltzoff (1995) showed that infants can understand the causal intentions of others. In his experiment, the infants saw several adults trying to perform a certain action but failed. Then the infants got the same items. They succeeded in carrying out certain actions. This suggests that the infants understood what the adults wanted to achieve and they therefore had intentions for a certain purpose. However, when a non-living device performed exactly the same actions as the human hand, the infants were unable to understand their intentions. The infants in this condition were less capable to imitate the actions of the device when the device failed to reach its goal.

Meltzoff claimed in his experiment that two separate causal frameworks exist at the age of eighteen months: a physical causality to explain the behavior of things and a psychological causality to explain the behavior of people. He suggested that infants represent the behavior of people in a psychological framework that contains goals and intentional actions and not in terms of purely physical movements. Psychological causation includes the understanding of unobservable events such as desires, beliefs, and goals which are all part of the theory of mind. Autistic children appear to lack a theory of mind.

Are there perceptual representations of causality or cognitive representations?

Meaning-based knowledge is described as representations that indicate what important information is about an event, so that unimportant details are omitted. Concepts or schemes are sets of ideas in the brain that explain the world and are therefore not perceptual copies of what the world is like. They can represent a category (for example, birds) or an event (for example, going to the doctor). Schemas for events are also called scripts.

Neurological research has shown that the activation of a concept produces neural activity in the sensory parts associated with those concepts, as well as in other networks such as association areas. As a result of this research, it was concluded that perceptual details are not omitted, although they can be activated less strongly.

It was thought that infants had categories that would be purely sensory or perceptual. However, this appears not to be the case, infants can also store other information. For example, they can store information about launching events that goes beyond the reach of perceptual features. After all, the launching events are based on action. Such experiments show that what is presented and remembered is not simply the perceptual characteristics, but also the structure or meaning. It appears that infants do not need language to understand something. Moreover, it appears that meaning-based knowledge is dependent on the number of experiences. When multiple experiences take place, only the essence of these experiences will be strongly remembered. As a result, the perceptual details will automatically be activated less strongly when such an experience is remembered.

Are there specialized modules for certain information?

Domain-general is the capacity that can be applied to different situations or domains. Domain-specific is the capability that is applicable to only one area or domain. An example of this is the mechanism that acquires the syntactic structure of language. Other modules are numbers and music. According to Leslie, there are two core domains that are central to the infant's initial capacities for causal conceptual knowledge. These are object mechanics and theory of mind. In these two domains, the central organizational principle is the observation of cause and effect.

According to Leslie, there is a specialized learning mechanism that creates conceptual knowledge about the physical world. According to Leslie, the modular organization of the brain ensures that the infant can acquire fast and uniform knowledge about object mechanics. However, the idea that the brain consists of modules is becoming increasingly obsolete. Cognitive neuroscience has shown that there is no number module as originally thought, but that important numerical information is stored in the language system.

What is the non-modular view of development?

This assumption relates to different learning mechanisms, namely statistical learning, learning by imitation, explanation-based or causal learning and learning by analogy. This last learning mechanism means that similarities are sought between two events, situations or domains in order to subsequently ensure that knowledge is transported between the two. With the help of these simple learning mechanisms, the brain tries to build complex representations about what the world is like. Here, repeated experience leads to generalization, because the connections between the neurons in question are strengthened.

How does problem solving in the physical world work?

Two other important characteristics of cognitive activity are reasoning and problem solving. A commonly used definition of reasoning is: processes of information retrieval that depend on the structure, and not on the content, of organized memory. However, it is difficult to decide which parts of the organized memory are structural and which are substantive.

According to Anderson (1990), reasoning and problem solving have three ingredients:

  1. The person wants to reach a certain end state, which usually consists of achieving a goal.

  2. There must be a certain order of mental processes when reaching this end state.

  3. The mental processes must be cognitive and not automatic.

How do infants reason about objects and events?

In an experiment by Baillargeon, Graber, De Vos and Black (1990), it was suggested that infants understand that certain events are impossible. In this experiment the infants showed a longer looking time at the impossible event than at an event that is possible.

How can we be sure that infants actually reason about the physical parameters of certain situations? A possible approach is to see if the infants are surprised by the impossible physical events. This is good evidence of problem-solving skills in childhood, as it contains a sequence of mental processes that is not automatic.

How can numeric relationships be studied?

Understanding relationships such as 'more/less than' is an important aspect of the numerical system. Understanding that a quantity remains the same unless something is added or removed from it is important as well. Habituation studies are used to investigate whether babies already have ideas about this.

In his research, Cooper (1984) presented situations with 'greater than' in the learning phase. In the test phase, an opposite relationship was shown ('less than') or an equal situation. Infants of ten months showed dishabituation for the same situation. This shows that they can distinguish equal amounts from unequal amounts. At fourteen months they also showed dishabituation for the reverse condition.

Starkey et al. (1983) conducted research into the cross-modal understanding of numbers in young children. Three or two items were placed in a row and included either three or two drum sounds. Infants showed a preference for a congruent sound and image. The researchers thought the infants must have a representation of the amount of drum sounds. However, this result was difficult to reproduce in other research.

In another study by Jordan and Brannon (2006), infants were shown films of either three women who said 'look' or two women. In addition, either three voices were heard or two. Infants also preferred a similar image and sound. However, it is uncertain whether these skills of the seven-month-old infants are present even if faces and voices are not used, as this is of crucial social significance for the infant (see Chapter 3).

One of the most well-known studies is by Wynn (1992). He examined the ability of five-month-olds to add and subtract small amounts. Infants looked significantly longer at the impossible event. This indicates that they can understand simple things about addition and subtraction.

Simon et al. (1995) replicated this study. They suggested the results could mean that an infant noticed when something is happening that is impossible in the physical world. It would therefore not necessarily be caused by children having knowledge of numerical relationships. They did a study in which, in the impossible condition, they did not change the quantities, but the identity of the dolls used. The result was that infants looked longer if an arithmetic impossibility took place than if an identity impossibility took place. Simon et al. argued that the spatio-temporal information allowed the infants to focus on the number of objects and not on the identity of the objects.

Other researchers state that changing numbers in the above "number" paradigms also causes a change in perceptual variables. The surface and the density of the image change. This could also be an explanation for the increased viewing behavior of infants. It does not necessarily have to do with arithmetic representations.

Recent experiments have controlled these variables. The results show that infants of six months can distinguish the difference between 8 and 16 but not the difference between 8 and 12. They can also distinguish 16 from 32, but not 1 from 2. These results lead to the idea that children process small quantities differently from large quantities. For quantities of 1, 2 and 3, they use perceptual variables in the displays. For larger quantities they use an analog magnitude representation.

A distinction has also been made between the stimuli used in the research. This is because visually appealing stimuli are processed via the ventral path and tangible stimuli via the dorsal path. Stimuli must be processed via the ventral path to notice the difference in identity. However, the experiment made use of graspable objects. The question is whether the results will still stand if this is taken into account.

Xu and Carey (1996) said that infants, at the age of three months, already have object permanence just like adults, but that they are unable to hold representations of numerically different objects that are covered. This is only possible from approximately the first year of life. Infants have a generalized representation of objects until they are approximately ten months old. In their experiment, Xu and Carey showed infants different conditions. In the property-kind condition, the infant was shown a screen and a truck on the right side was taken from behind the screen and returned behind it again. A toy kitten appeared on the left side and then disappeared behind the screen. After three times the screen was lifted and there were either two objects or one object. Infants who had representations of numbers would look longer if only one object was presented.

This was also done in the spatio-temporal condition. In this case, however, the truck and the kitten were shown on two sides of the screen at the same time. In this condition, infants were also expected to look longer if only one object was shown after lifting the screen.

The results were as follows: infants looked longer at the outcome where there were two objects in the property-kind condition, but looked longer when there was only one object at the outcome in the spatio-temporal condition. The researchers concluded that infants have an intrinsic preference to look at multiple objects and that this is overruled by the spatio-temporal condition and not by the property-kind condition. From this they concluded that infants of ten months old are unable to see the perceptual differences of the objects. An alternative explanation is that both conditions activated the dorsal path because the stimuli were graspable.

How does imitation in young children work?

From a very young age, babies are able to imitate certain gestures of adults. In an experiment by Meltzoff and Moore babies were seated in a dark room. For 20 seconds a light illuminated the face of an adult who made a certain gesture. For example, he stuck out his tongue. Then the light was dimmed and the babies were filmed in the dark for the next 20 seconds. This session was then repeated. A researcher who had no idea of ​​the design of the experiment recorded the number of gestures of the baby. As time passed by, a baby - who had seen an adult do this several times - stuck his tongue out more and more. The conclusion was that the baby had imitated the adult.

How do children learn? - Chapter 3 (1)

When we are young, we learn to understand that the actions of others depend on the knowledge and beliefs that they have and that this knowledge and beliefs may be different from our own. This is a complicated cognitive skill. Tomasello (1995) believes that a "revolution" is taking place in the understanding of social relationships and activity around the age of nine months. However, this does not seem likely. Prior developments already prepare an infant for social skills. According to this view, the child, the caretaker and the object are central to early social development. The play (or other actions with objects) of infants and parents play a crucial role in this.

When babies are born, they already have a tendency towards social interaction. For example, they have certain mechanisms (such as crying and laughing) to keep the caretaker close. Moreover, they have a strong preference for, for example, the smell and voice of the caretaker. DeCasper and Fifer (1980) demonstrated this by having babies of twelve hours old suck on a dummy. There were two groups. If the babies from group one started sucking louder, a tape was played with the voice of the mother and if they started to suck less loudly, a tape was played with the voice of another woman. In the second group this was done the other way around. The babies soon learned to suck on the dummy correctly so that they would continue to hear the mother's voice. Thus, they recognized the voice of their mother and that made them feel comfortable.

When the test was repeated one day later, the infants had remembered the pace at which they had to suck to hear their mother's voice. But now the conditions were reversed: the babies who were used to hearing their mother's voice when they sucked loudly now heard the other woman's voice and when they sucked less loudly their mother's voice. 80% of the babies learned to adapt to the suck rate of these reverse conditions. The children therefore had a preference for their mother's voice. These results suggest something about the learning and memory of young children. The rapid reversal learning shows that babies are already more developed in the cognitive field than other types of organisms.

The babies had already learned the memory of the mother's voice in the last three months in the womb. DeCasper and Spence (1986) proved this with a study in which the mother read a story every day in the last six weeks of her pregnancy. After birth, the babies appeared to have a strong preference for this familiar story.

The preference for the face, the smell and the voice of the mother are part of the attachment behavior that ensures that there is a close bond between the child and the primary caretaker. This is very important for the healthy psychological development of the infant. However, this book focuses more on the cognitive aspect of understanding. Contingency learning is the ability to discover and learn the contingencies and relationships between one's own actions and events in the environment. This plays a crucial role in the development of social cognition.

Gergely (2001) believed that before infants develop the mechanism to be able to understand the consequences of their own motor actions, they first need to have a representation of their physical 'self'. This representation develops when the baby is two to three months old. Bahrick and Watson (1985) demonstrated this by positioning infants in a seat and showing a video to one group of infants of their direct movements (so they saw on a screen what they were doing at the time) and the other group a video of themselves, but recorded at an earlier moment (causing them to make different movements on the screen). The infants looked longer at the video which was not contingent with their own movements at that time.

However, children do not recognize themselves in the mirror at this age. This develops around the age of 2 years. With the mark test, a noticeable mark is made on the infant in a location that the baby cannot see (for example, a red dot on the nose). If the infant looks in the mirror and points his own nose, it can be concluded that he realizes that whoever he sees in the mirror is actually himself. Children who do not yet realize this, point to the red dot on the nose of the mirror image. Amsterdam (1972) used this task in her research. She concluded that self-recognition was only possible from the age of eighteen months, but that children already had a sense of it (for example, by saying their name, by reaching for the mirror image, by looking at what was behind the mirror, or by carrying out motor actions and look with interest at what is happening in the mirror). There are four phases that play a role in development:

  1. Social responses to the mirror.

  2. Inspecting the physical aspects of the mirror (for example, by looking behind it).

  3. Repeated mirror-testing behavior.

  4. Mirror behavior directed towards themselves (so when they touch their own nose).

Plotnik et al. (2006) did experiments with elephants, among others, and concluded that the awareness of the difference between yourself and another is present in many types of species. Social interaction plays a role in this.

Infants have a relatively early social understanding. This could be because parents treat their child as a social partner early on, sometimes even before the child tries to communicate. Striano et al. (2005) concluded from their experiment that children show social understanding from the moment they are three months old. In the first phase they let the mothers interact with their baby as they normally did. They recorded this. A while later the infant came back and there were three possibilities: (1) the mother interacted with her child as normal, (2) the mother received headphones with the recording of the last time and had to repeat this as much as possible, or ( 3) the mother had to imitate the behavior and facial expressions of her child. Infants of three months old stared longer at condition 3 and had a normal smiling pattern at condition 1.

Gergely (2001) states that children at the age of three months have learned that they have their own body and that they can apply this knowledge in their social environment. They know that their own actions influence the actions of their caretakers. Very early in development, children show social interactions based on evolutionary concepts, because it keeps the caretaker close. Later these interactions arise from the realization that they can influence the behavior of their caretakers through their own behavior. Meltzoff (2002) says that early interaction with people is crucial, because the infant then recognizes other people as "they are just like me." Imitation plays a major role in this. Gergely therefore emphasizes the awareness 'I am different from everyone else' and Meltzhoff emphasizes 'I am the same as everyone else'.

Why do the actions of others play a central role?

When adults observe someone doing something, they assume that this person has a purpose and that he therefore performs a certain action. Children do this too. Gergely et al. (2002) demonstrated this in a study. Meltzoff (1988) had previously shown that when someone pressed a light switch with his forehead instead of his hand, children imitated this. There were two conditions in Gergely's research. In the first condition it was told that the person who performed the behavior was cold. That is why he had wrapped a blanket around him. Thus, his hands were covered. Children of fourteen months old who had seen this and later had to do it their selves, only used their foreheads in 21% of the cases to imitate this action. 79% used their hand.

In the second condition, the person who carried out the action was able to use his hands, but he still used his forehead. 69% of the children also used their forehead in this condition. The context of the situation was therefore different in the two conditions. The different imitation due to a different context is strong evidence for assigning a goal to the actions of others.

In another experiment by Carpenter et al. (2005) a toy dog ​​was used. In the first condition there was a mat on which a house was present. The researcher grabbed the dog and let him enter the house by eight jumps or by one long 'slide'. In the second condition there was no house present and the researcher simply let the dog jump eight times or make a long 'slide' over the mat. An infant might think that the purpose of condition one is to get the dog into the house and that the purpose of condition two is "just playing." When the researcher asked if the infants wanted to do it themselves, the twelve and 18-month-old infants in condition one immediately put the dog in the house (congruent to the goal) and in condition two they jumped or slid with it. Infants therefore interpret the actions of others as actions with a purpose.

In their experiment, Carpenter et al. (1998a) investigated whether infants also imitate actions that are carried out 'by accident'. It turned out that intentional actions were imitated in 78% of the cases and accidental actions in just 43% of the cases. The infants looked at the ultimate goal of the person they imitated and ignored the accidental actions that came with it.

Behne et al. (2005a) investigated whether the infant can understand the different types of intentional actions. In their experiment they showed three conditions in which the experimenter passed a toy to the child, but in the end, he did not give it. There were three different explanations for this: (1) the experimenter did not want to give the toy, (2) the experimenter wanted to give the toy, but was unable to do so, and (3) the experimenter was distracted and therefore did not give the toy. In all conditions the infant did not receive the toy and the result was ultimately the same. However, the infants in the different conditions reacted differently.

In condition 1 they reached for the toy longer and after a while they looked away. In condition 2, they reached for the toy for a shorter time and looked away less. In condition 3 they did not reach for the toy as long as when the experimenter was unwilling to give the toy. They understood that the experimenter wanted to give the toy in condition 2 and 3, but that there was no possibility. They adapted their behavior to the social situation. This concerned infants of twelve and eighteen months of age. Infants of nine months showed the same pattern, but infants of six months did not react differently in the three conditions. We can therefore assume that a revolution is taking place in social cognition at around the age of nine months. However, we cannot state with certainty that younger infants do not yet have any sense of intentions from others.

Woodward (1998) investigated whether younger infants also looked at the actions of others in terms of achieving goals. He had infants look at a small stage where a teddy and a ball were placed side by side. The first group of infants was shown an arm that reached for the teddy or the ball and grabbed it. The second group of children was shown a poster tube with a sponge at the end. The tube reached for the teddy or for the ball and it rested there. The babies were habituated under these conditions.

In the test phase, the teddy and the ball changed positions and two things were possible. The first possibility was for the hand or tube to reach the same object and therefore to a different location (the so-called new path (same goal) condition). The second possibility was for the hand or tube to reach the same location but to the other object (the new goal (same path) condition). Infants who had seen an arm looked longer at the new goal condition. For infants who had seen the tube, there was no difference between the two conditions. The infants were therefore only interested if the person had a new purpose. Experiments with six-month-old infants provided the same results.

Sommerville et al. (2005) discovered that even babies of three months old are aware of goal-directed actions, given that they have the opportunity to perform these actions themselves.

What is the function of goal-directed actions and the attribution of the mental state?

It was suggested that infants should be aware of the mental state of others in order to be able to interpret their goals. Gergely and Csibra (2003) argued that this realization was not necessary. They proposed that infants around 12 months of age adopt a teleological stance about the representations of actions. The actions are represented by (1) the goal state, (2) the action to reach the goal state, and (3) the relevant aspects of reality as limitations on possible actions.

Shimizu and Johnson (2004) supported this hypothesis with the results of their research by demonstrating that infants can also assign goal-oriented behavior to objects. In the habituation phase, they allowed a green oval object to behave as if it could make contact with the social environment and as if it could work purposefully. The infants looked at the object longer in the test phase if it performed a familiar action with a novel goal. This was also the case with human agents.

Teleological explanations depend on the relevant aspects of a situation and not on understanding the mental state of a person. The person who interprets the situation is mind blind: he cannot assign internal states to others and thinks purely from rationality and not from 'what someone wants'.

False beliefs is the understanding that others may have beliefs that are not based on reality. These beliefs are therefore incorrect. If a 12-month-old infant is mind blind, the infant cannot apply false beliefs. If he does understand the actions based on false beliefs, then he must understand the mental state of the person who performs the behavior to consider the actions as rational.

In their research, Kuhlmeier et al. (2003) showed that children are able to assign mental states to objects. During the habituation phase, they showed two films in which a ball had to climb a slope. In the first video the ball was supported by a triangle that pushed the ball up the slope. In the second video the ball was hindered by a square that pushed the ball back. In the test phase, the triangle was on one side and the square on the other. The ball had to move close to one of the two objects. Infants thought it was strange (just like adults) if the ball lay next to the square, because this was not originally his 'friend'. Further experiments need to be done to see if the teleological or mentalistic principle is the best.

How are gaze following and gaze monitoring used in research?

Research into the social cognition of children can focus on the gaze-monitoring behavior of children. With gaze monitoring you follow the focus (gaze) of an individual to find out what attracts someone's attention. Babies, just like adults, find it uncomfortable when someone doesn't make eye contact. In the still face paradigm, the mother discontinues face-to-face interaction with the infant and looks at the infant with a neutral face. The infant avoids the gaze and gets upset.

When parents look at visual stimuli with their children, learning is promoted. A child must be able to follow the focus of the mother. Scaife and Bruner (1975) investigated this among infants from two to fourteen months of age. They let an experimenter play with an infant. At one point the experimenter turned his head 90°. Then there was another moment of interaction and then the experimenter turned his head again, but to the other side. If an infant (1) looked in the same direction as the experimenter, (2) did not look elsewhere first, (3) looked within seven seconds and (4) seemed to be looking for something or was looking at something, this became considered as gaze following. 30% of the two-month-old babies followed, 65% of the infants from eight to ten months old and 100% of all infants over the age of eleven.

Older infants also showed social reference, which uses another person's feedback to understand a situation and to determine how to respond. They did this in the current experiment by looking back at the experimenter and by looking again at the point the experimenter was looking at.

Some argue that following the focus is not necessarily due to understanding that someone voluntarily chooses to look one way. They claim that following the focus of another can come from conditioning, because an infant has learned that when an adult turns his head, something interesting happens there. An infant should also turn his head when the adult turns his head with eyes closed. Brooks and Meltzoff (2002), however, concluded with their experiment that this is incorrect. They allowed adults with eyes closed and eyes open to change the focus while playing face-to-face with the infants. The infants followed the focus of the adult significantly more often when the eyes were open. In addition, pointing and vocalizing that the focus changes have an influence on following the focus of the adult. Correct interpretation of the focus and vocalizing the focus in infants aged ten and eleven months ensures a better understanding of the language at the age of fourteen and eighteen months.

What is the importance of seeing as an intentional action?

When infants follow the focus of an adult to see what the adult sees, they must interpret the focus of an adult as intentional. Woodward (2003) conducted a study in which there were two objects (a teddy bear and a ball) and in which an adult was visible. If the adult and the infant were in contact, the adult said in the habituation phase: "Look!" and thereby looked at one of the two objects. In the test phase (in which the objects were reversed) the adult looked either to the opposite direction and to the same object, or to the same direction but a novel object. Infants of seven and nine months of age often followed the focus of the adult and infants of twelve months as well, but they looked significantly longer if the adult's focus was on a new object.

Moll and Tomasello (2004) investigated whether infants also follow the focus of an adult if their vision is blocked. They found that eighteen-month-old infants changed positions more often to avoid the blockade and see what the adult was looking at than twelve-month-old infants. In the control group, in which there was no blockage present, the infants of twelve months followed the focus of the adult as often as the infants of eighteen months of age. Moll and Tomasello concluded that the mechanism for following the focus for an object in the visual field develops easier than if the view is blocked.

It has been investigated whether infants also understand that if an adult looks at something, this can be a sign of communication with the child and not just an act of the adult. Behne et al. (2005b) concluded from the results of their research that infants from the age of fourteen months understand that shifting the focus can have a communicative intent.

Why and how do infants use social reference?

There are two possibilities why infants use social reference:

  1. They adjust their behavior because they have a mentalistic interpretation of someone else's reaction (for example: he looks scared, so this may be a dangerous situation).

  2. They adjust their behavior because the emotion of an adult acts as a signal (for example: he looks like that because I have to stop).

Only with option 1 is it necessary for the child to understand that another person has a mental state.

The classical study of social reference made use of the visual cliff, which was also used to measure children's depth perception. The infants then crawl over a transparent surface and under that surface different forms of depth can be indicated. An infant of about nine months old does not crawl to a part where it seems much deeper, because he thinks he could fall. However, such a situation can be manipulated. When the infant reaches the cliff, he looks at his mother. When she looked anxious, the child did cross the cliff. However, when the mother looked happy, fourteen out of nineteen infants did cross the cliff.

The same kind of experiment was done by Hornik et al. (1987) with the use of toys. If the mother showed fear or disgust, the infant stayed away from the toy. Positive expression from the mother did not influence the infant's behavior towards the toy.

Mumme et al. (1996) investigated whether the mother's facial expression or the mother's voice was a better cue for social reference. The results indicate that infants aged twelve months do not yet fully understand facial emotional expressions. However, this is the case when vocal cues are used. This is contrary to the results of the experiment with the visual cliff, but probably because there was a clear danger in that experiment (falling down into the cliff). Vocal cues are very important for social reference in children. The face is an important cue in potentially dangerous situations.

Which actions of infants are relevant to study?

Pointing is an important function in the communication of an infant. There are two different functions of pointing. In proto-imperative pointing, the child uses pointing to get something. With this way of pointing it is not necessary for an infant to understand that someone else has a mental state. In proto-declarative pointing, the goal of the infant is to have someone focus their attention on something of interest of the infant. The purpose of this is joint attention and it requires a higher, more mentalistic level of communication. The infant starts to use proto-declarative pointing around the age of twelve months. The proto-imperative point is present in an infant of about fourteen months.

If someone points to something, the infant can follow it at the age of 11.7 months. He must realize that someone is pointing to bring an object to the attention of someone else (the infant), so that joint attention is given to this object.

Around the age of twelve months, infants understand that pointing to something is an action with an intention. Bates et al. (1975) also thought that proto-declarative pointing of infants are intentional actions for communication. They try to achieve joint attention. Liszkowski et al. (2004) investigated this. If the experimenter responded to the infant when the infant pointed to the doll that appeared behind the curtains and so there was joint attention, the child pointed less often than when there was no joint attention. The infant pointed longer if there was joint attention. If the experimenter only looked at the doll, but did not share the interest in the doll with the infant, the infant looked at the researcher significantly more often. These results show that the social context in which the pointing takes place has a significant effect on the behavior of the infant. Proto-declarative pointing serves to share interest with another and has a communicative intention.

How can false belief be understood?

The philosopher Denett found that good reasoning about false beliefs was the only evidence of the attribution of a mental state (understanding and being able to point out the goals of someone else and the intentions underlying them) to others. If someone acts on the basis of a false belief, this does not correspond to what is expected from a real situation.

With the false location false belief task, it can be demonstrated if an infant does or does not understand the concept of false belief. A person sees that something is being hidden at a certain location. Then this person leaves the room. When the person is gone, the object is hidden in a different location. When the person returns, he will first look for the object at the original location, because he has a false belief that the object is still there. Only when infants understand why the person first goes to the wrong location and after this looks for other locations where the object can now be hidden, does the child understand the concept of false beliefs.

In the Onishi and Baillargeon (2005) experiment, fifteen-month-old infants looked significantly longer if they knew that someone had a false belief and that person immediately went to the right location. They also looked longer if they knew that someone had a true belief, because they had observed that the person had also seen that the object had been moved, and that the person was still searching in the original place. This showed that these infants understood the situation and that they did indeed give people a theory of mind: they understand that someone has a mind through which he or she can have beliefs that are wrong. They then not only attribute goals and intentions to others, but also beliefs.

What does social cognitive neuroscience focus on?

Tzourio-Mazoyer et al. (2002) studied whether infants of two months old exhibited activity in the fusiform face area when they were offered faces of unknown women. This area is not yet fully developed in infants this age, but has been identified in adults as the area that plays a role in the processing of faces. Indeed, there was activity in infants in this area. This area therefore appears to have a relatively specialized activity.

Another study looked at the N170 wave in the brains of four-month-old infants and whether it would differ if infants were shown a face that was staring at them (direct gaze) or looking in a different direction (averted gaze). The results showed that infants showed an N170 (later than adults, namely at 240 ms) and that the amplitude was greater (more negative) when the faces made eye contact than when they looked away. The researchers suggested that this could be important for the interpretation of eye gaze signals in communicative actions.

Studies of EEG activity with joint attention found that baseline EEG in the left frontal and some central areas was related to the tendency to initiate joint attention. The parietal areas were active in responding to a request for joint attention. The researchers thought that the frontal and central areas play a role in inhibiting other visual cues to focus on the joint attention and that the parietal areas play a role in the required spatial actions to follow the attention of someone else (for example, moving the head).

When we are young, we learn to understand that the actions of others depend on the knowledge and beliefs that they have and that this knowledge and beliefs may be different from our own. This is a complicated cognitive skill. Tomasello (1995) believes that a "revolution" is taking place in the understanding of social relationships and activity around the age of nine months. However, this does not seem likely. Prior developments already prepare an infant for social skills. According to this view, the child, the caretaker and the object are central to early social development. The play (or other actions with objects) of infants and parents play a crucial role in this.

What is conceptual knowledge? - Chapter 4 (1)

Conceptual knowledge is knowledge about the kinds of things in the world. This is not only about knowledge about objects, but also about actions, events and mental states. Conceptual development consists of inductive learning and categorization. Induction is a way of reasoning in which one deduces the general from the specific. Generalizing on the basis of a known example is one of the most common forms of inductive reasoning and is the basis of categorization. With induction we fill gaps in our knowledge. The ability to reason on the basis of induction is already present early in development. The focus of research is therefore on the organization of knowledge that determines the ability to categorize.

What are superordinate, basic and subordinate levels?

Categorizing is organizing things into a category. There are different levels within a category: superordinate level (animal), basic level (cat), and subordinate level (siamese).

Neisser (1987) describes categorization as the ability to treat a set of things as equal, to put them in the same pile, to call them the same, or to respond to them in the same way. Categorization is important because we cannot consider every object as unique and because it gives us the opportunity to find out more about an unknown object by just looking at it and classify it into a specific category. Perceptual information, conceptual information and beliefs play a role in categorization. The prototype theory is the theory that the normal level of a category (basic level) gives us the most information about what something is and how it should be classified.

How do basic level categories and prototypes develop?

We categorize a lot on the basic level (normal level; from now on we will refer to the levels as basic, subordinate and superordinate). It is thought that items from the same basic level category have the most characteristics in common and the fewest characteristics in common with items that do not fall within the same basic level category.

In Chapter 1 it was discussed that infants can form prototypes of drawn pictures. Eimas and Quinn (1994) showed that they can also do this with existing animals. They allowed three to 4-month-old infants to habituate to pictures of horses. Then they showed a new picture of a horse in the test phase together with pictures of a zebra, a giraffe or a cat. The infants looked at the other animals longer. This shows that they can see the difference between the prototype of a horse and that of another animal. In the physical world size, sounds and movement patterns can also help to classify objects or organisms into a prototype. Arterberry and Bornstein (2001) investigated this. They found that infants of three months old could distinguish between categories based on rich visual input just as good as based on movement cues.

However, these tests provided information about perceptual categories rather than about conceptual categories. To test conceptual categories, we look at the sorting behavior of infants who cannot yet talk. For example, if they assign toy cars to toy cars and toy horses to toy horses, then they probably have knowledge of conceptual significance.

How can sequential touching be used as a measurement for basic-level categorization?

However, children do not show spontaneous sorting behavior before the age of eighteen months. As a sign of categorization, they use sequential touching. They touch objects that belong to a certain category in sequence, so first all cars and then all horses. Mandler and Bauer (1988) investigated this at a basic level (cat and car) and at a superordinate level (animal and vehicle). Twelve and fifteen-month-old infants distinguished only through sequential touch on a basic level. Only infants aged 20 months made a distinction at superordinate level.

There were many individual differences, for example, that 25% of the twelve-month-old infants were responsive to the superordinate level. In a follow-up experiment, they showed that basic level categories were easily distinguished from each other if they belonged to two different superordinate categories, but more difficult if they belonged to the same superordinate category. Superordinate classification therefore plays a larger role than was thought.

However, these results can also be explained differently. It could also be due to the perceptual context that items from the same superordinate class are more difficult to distinguish from each other. A difference between a car and a dog is easy to see, for example, because a car has no legs and no head. A difference between a horse and a dog is more difficult to see, because they do have these characteristics in common. It is therefore not entirely certain that the results of the study by Mandler and Bauer indicate that it is impossible to distinguish different basic level categories within a superordinate category.

Furthermore, membership of a basic level often means membership of a subordinate level. The theory (about the different levels devised by Rosch) states that perceptual equality correlates with structural equality. The perceptual equality between dogs and horses represents an underlying structural equality, namely that they are both organisms. Basic level and superordinate concepts stem from perceptual knowledge.

Pauen (2002) found that organizing objects into categories is also based on previous knowledge that a child has about certain things. She allowed infants to hold on to some objects of a category and look at them until they were used to it. Then she gave the infant a new object from the same category or a new item from a different category. Pauen, however, created a series of objects that were a mix of animals and furniture. She manipulated the objects in such a way that there were more perceptual similarities between the different groups than within the different groups. The furniture, for example, had eyes. Yet infants spent more time looking at an item from a new category than at an item from a familiar category. So, they used their previous knowledge of categories to determine in which category these objects belonged.

What is the matching-to-sample test?

A child's categorical capacity can also be determined with a matching-to-sample test. A target is shown and then they are asked to select from a number of newly shown stimuli which stimulus forms a pair with the target. Bauer and Mandler (1989b) had children perform a matching-to-sample test based on basic level pairs and super-ordinate pairs. The children (19, 25 and 31 months old) were slightly better in basic-level objects, but they also scored very well on super-ordinate objects. So, from the age of nineteen months there is a sensitivity for both basic level categories and super ordinate categories.

Rosch argued that categorization starts at the basic level and then develops into categorization at the superordinate and subordinate level. However, different studies indicate something else.

What is the core developmental role of the superordinate level?

Mandler and Quinn et al. both suggest that categorization first takes place at a superordinate level and thereafter at a basic level. Mandler says this is due to perceptual (basic level) categorization and conceptual categorization. However, Quinn et al. argue that the perceptual learning processes are sufficient for categorization at superordinate level first and subsequently at basic level.

Mandler (1991) demonstrated his idea by allowing children of 19, 24 and 31 months of age to distinguish between two basic level categories through sequential touching. He paired dogs with horses (low perceptual contrast), dogs with rabbits (medium perceptual contrast) and dogs with fish (high perceptual contrast).

Children of 31 months old could keep the dogs and horses apart. In the medium perceptual contrast, children between the ages of 24 and 31 were able to see the difference well and with regard to dogs and fish, any age group could make a distinction. All groups could also see the difference between dogs and vehicles (different superordinate category). Mandler concluded that the children start at a super ordinate level and are increasingly refining their categories into basic level categories. However, with these results it cannot be concluded with certainty that being unable to distinguish between a dog and a horse means that the basic level categorization for dogs and horses is not present in the child.

Mandler and McDonough (1993) demonstrated that infants of seven, nine and eleven months old do not make a distinction between dogs and fish in an object examination task, but do make a distinction between animals and vehicles. This shows that they can distinguish at a superordinate level, but not at a basic level. In this study, too, toy animals were used and toy animals only show certain characteristics of an animal, but do not present the animal in terms of smell, size and so on. These characteristics could be a key component in distinguishing between different types of animals at the basic level. In addition, even in this study it cannot simply be said that, if a child does not make the distinction, categorization does not actually exist.

Quinn has a theory which is called the global to basic sequence (first global categorization and then basic level). This theory is based on perception. It derives from connectionist modeling. This is a mathematical model of learning via neural networks. Every unit in the network has an output that is a simple numerical function of its input. Cognitive skills, such as language, are represented by patterns of activity at different units. Information therefore enters the input nodes and relevant information about identifying characteristics is filtered and passed on to the output nodes. So, it starts with global categorization, but during learning, more and more nodes are being developed that will distinguish between multiple details. This makes categorization on a basic level possible. The more of these nodes develop, the more categorization takes place at the basic level and ultimately this is preferred.

What are typicality effects in sequential touching tasks?

A prototype is an example from a category that is considered to be very representative of that category. Bauer et al. (1995) investigated whether prototypes play a role in categorization and made use of the sequential touching measurement. They presented children aged 13, 16 and 20 months with sets of objects that consisted of either prototypical examples or non-prototypical examples on subordinate (dog/cat/horse versus crocodile/kangaroo etc.) or on basic level (trout/salmon versus shark/eel). They found that categorization was better with the prototypical sets than with the non-prototypical sets. For thirteen-month-olds, categorization only took place on the basic level and for children aged 16 and 20 months also on the superordinate level. In the non-prototypical set, the categorization of 16-month-old children was only at a basic level. The rest of the sequential touching with non-prototypic sets was not significant. For children of 24 months old a categorization was found for non-prototypic sets, but this only happened at superordinate level and for children of 28 months old this took place at both levels. Furthermore, the degree of prototyping explained a larger part of the variance than age or level of categorization. This emphasizes the importance of prototyping in categorization.

How are child-basic categories represented in cognitive neuroscience?

The brain encodes objects, events, actions and other concepts in terms of sensory units that become active when concepts are experienced directly. For example, when the concept of dog is activated, knowledge about what a dog looks like, but also about how it smells, feels and how a person feels when he sees a dog. There is no modeled system for storing such a concept and the information comes from different parts of the brain.

Since children experience concepts different from adults, different parts of the brain are therefore activated. Mervis et al. (1987) named this child-basic categories. They are opposed to adult-basic categories. The child-basic categories can ensure that children focus on different basic level categories than adults. Those categories can be broader or narrower or can overlap. They studied the categorization of a child with regard to ducks. This categorization changed slowly. First it only contained pictures of ducks, then a duck rattle and then a plush duck head rattle and Donald Duck. Then the child saw ducks in many more instances than his mother, such as in a swan-shaped soap.

How can we look beyond the role of perceptual similarity in categorization?

Perceptual information is correlated with structural information. This is the reason that so much emphasis is placed on perceptual equality between different category members in categorization. However, sometimes the perceptual information does not match the conceptual information. In a study by Gelman and Coley (1990), children were shown pictures of a prototype bird, for example, and were asked if he lived in a nest. Then other pictures were shown. This could be a typical picture from the same category (such as a bluebird) or an atypical picture from the same category (such as a dodo). It could also be a typical picture from another category (such as a stegosaurus from the dinosaurs category) and an atypical picture (such as a pterodactyl, which looks a lot like a bird). The child (2 years old) was asked whether the presented animals also lived in a nest. If children determine their answer to the overall appearance (so perceptually), they would say that the bluebird and the pterodactyl live in a nest and the other two do not. If they base their answer on the structural characteristics that determine membership in a category (i.e. dinosaurs versus birds), they would say that the dodo and the bluebird live in a nest. Children of 2 years old rated dodos (not typical examples from the category) significantly less often as a nest inhabitant than bluebirds. However, if a hint came in the form of language ("this is a dinosaur / bird"), they did succeed.

In another study, Gelman and Markman found that children aged 3 and 4 consistently attributed traits to concepts based on categorical membership rather than on perceptual appearance. These results suggest that children of 3 and 4 years old can use categorical information alone as a basis for making inductive references. They are therefore not necessarily dependent on perceptual information. However, it cannot simply be concluded that categorical membership is more important than perceptual equality. Because the perceptual stimuli in these experiments were merely drawings, many perceptual characteristics disappeared.

What is the role of language in conceptual development?

When children start to learn language, it automatically speeds up categorization. The word 'animal', for example, already represents a super ordinate category. Children often interpret new nouns as superordinate categories and new adjectives as subordinate categories. Furthermore, a study revealed that classification at the basic level almost had a ceiling effect in both conditions (adjectives/nouns). Here, linguistic cues did not play a role in facilitating or inhibiting performance. Waxman argued that children on non-basic levels rely on syntactic cues. She also argued that children have a linguistic bias because they are sensitive to the strong links between conceptual hierarchies and the language that we use to describe it.

What is the distinction between biological and non-biological concepts?

When categories are formed, children learn about the fundamental differences between categories in addition to the conceptual hierarchies within a category. One of the first and most important differences is the difference between biological entities (people/organisms) and non-biological entities (objects). Very young children can already recognize this difference and this basis is further expanded as the child experiences more of the world.

What evidence is found from studies of biological movement?

One way to find out if children make this difference is to see if children see the difference between the movement of a non-living object and the movement of a living object. Bertenhal et al. (1985) used the point-light walker displays to investigate this. These are made by dressing people in black, attaching lights to key points such as their joints and their heads, and then having them move in the dark. Adults easily recognize a person's movement when they see such a movie. Bertenhal et al. made use of the fact that when someone is walking certain lights are covered while taking a step, because another part of the body is temporarily in front of these lights, thus covering the lights. They manipulated videos so that the covering of the lights did not take place. Infants of nine months old only dishabituated for the 'real' videos, which were not manipulated. This suggests that they are sensitive to biological movement. In a follow-up experiment, Bertenhal discovered that this was due to an implicit detection of the person's body, because this effect was not found when the films were shown upside down.

Movement cues are also important for the distinction between biological and non-biological concepts. The lines of movement of a non-biological concept are predictable, in contrast to those of an organism. Lamsfuss (1995) showed that children aged 4 and 5 years old understand this by showing them two tracks (irregular and regular) and asking which one probably belonged to an animal and which one belonged to a machine. Children chose the regular track with a machine and the irregular track with an organism significantly more often.

What evidence is found from knowledge of movements?

Another way to explore the understanding of biological and non-biological concepts is to see if children recognize that biological concepts can move on their own. Massey and R. Gelman (1988) showed children pictures of living and non-living objects and asked them whether or not these objects could move up and down a hill on their own. Unknown pictures were shown of mammals (marmoset), of non-mammals (tarantula), and exemplars of three inanimate categories; statues of animals, of objects with wheels (bicycle; which can only go down a mountain but not up a mountain) and of complex rigid objects (a camera). Children aged 3 years made good decisions about the pictures in 78% of the cases and children aged 4 years in 90% of the cases. They often focused on legs and feet. If an object did have feet (as with the statues), the children ignored the feet because the objects could not move. The 3-year-olds made the most mistakes with non-mammals but realized that they were animals: for example, they said that a tarantula was too small to be able to climb the mountain.

The study by S. Gelman and Gottfried (1993) showed that children of 4 years old already make a distinction between animals and other objects: animals can move themselves and toys and objects can only move with the help of an external force or person.

The presence or absence of a goal also plays a role. In the study by Opfer (2002), children were exposed to moving blob-like shapes that seemed to have a purpose or not. If they had a purpose, they were considered living things. Yet another way to investigate the notion of biological versus non-biological objects is to see if children can distinguish between differences and similarities between the inside and the outside of objects. Animals, for example, all have blood and bones.

What evidence is found from the assumption of shared core properties?

S. Gelman and Wellman (1991) showed children pairs of three and the children then had to judge which two were most similar or which two had the same inside (orange, lemon and orange balloon: orange and lemon have the same inside and orange and orange balloon the same outside). The 3- and 4-year-olds all performed significantly well. The 4-year-olds did better than the younger children. The same number of mistakes were made in both conditions, so mistakes were not made simply because objects looked alike. An equal number of errors were made regarding two objects with the same inside as similar in appearance. The conclusion of Gelman and Wellman was that age was not so much the ability to distinguish between the inside and outside, but that the ability to deal with inside and outside conflicts develops with age. Children therefore know the difference between the inside and outside of different categories, but sometimes find it difficult to make the right choice. This was supported by the research of R. Gelman and Meck.

Simon and Keil (1995) stated that children have a more abstract view of the inside and outside of organisms and non-living objects than a concrete view. They told children that Freddy, an alligator who had never been to Earth before, could see the inside of all objects and organisms. But he was confused and the children were asked to help Freddy by showing what was the exact inside of the objects and organisms. In one test they showed pictures of an object or an organism with a hole in it through which the inside could be viewed. They showed pictures of, for example, a sheep with the inside of a machine and another picture with organs on the inside.

The children then had to choose which picture was right. In the second test, a picture of a sheep was shown and it was asked to choose the inside that belonged to the sheep from three jars. In the first jar there were wires and screws and so on (from a machine), in the second jar the organs of two cats and in the third jar small white stones dipped under gelatin (a mix of non-biological and biological things). The youngest children (he tested them at the age of 3, 4 and 5 years old) knew well how to attribute the machine-insides to machines, but chose 2 and 3 equally often for organisms. Simons and Keil argued that the younger children did not know exactly what the inside looked like, but they did know that some insides were more likely to occur with machines than with animals and vice versa. So, they had abstract representations, but no concrete ones.

Inside and outside are examples of shared core properties. These are characteristics that are important for categorization of non-living and living objects. Another method to measure the understanding of this is to ask whether children can make verbal assessments (not just choose from two pictures). The children will be forced to reason abstractly. Children can make a verbal distinction between the inside of living and non-living objects (Gelman and O'Reilly, 1988).

Keil (1994) suggested that children might look at the shared structure to categorize objects and organisms, but that the shared function is much more important. Pauen (1996a) did an experiment to investigate if this is true. It was found that perceptual inequalities (different parts) are used to create subcategories in living things, but not in non-living things. The latter only happens if the function of the different part remains the same.

What evidence is found from growth studies?

Another way to see if children know the difference between biological and non-biological objects is to see if they understand the concept of 'growing'. Rosengren et al. (1991) investigated this concept of growth in children aged 3 and 5 years. The children were offered a picture of an object or organism and then two pictures were shown and the child had to judge which of the two pictures represented the object or organism after some time had passed.

The two images were either the same size as the first image and larger than the first image (same-size-bigger condition) or the same size as the first image and smaller than the first image (same-size-smaller condition). If children understood that living things are growing and non-living things are not, they would consistently choose the image with the same size as the target object for non-living objects and never the image smaller than the target image for living objects. The 5-year-old children scored 100% correct in the same-size-smaller condition in animals and 97% correct in the same-size-bigger condition. The 3-year-old children scored 78% and 89% correct respectively. For the non-living objects, the 5-year-old children also scored 100% and the 3-year-old children 78% accurate in the same-size-smaller condition and were at chance in the same-size-bigger condition. These results show that the principle of growth is understood better for living things than for non-living things.

How can analogies be used as a mechanism for understanding biological principles?

Children make predictions about the behavior of other objects or organisms based on their knowledge of people. Inagaki and Hatano (1987) asked 5- and 6-year-old children if a person, a rabbit, a tulip and a stone could stay the same size forever, without the object growing. The children responded that this was not possible for a person, a rabbit and a tulip, but it was possible for a stone. In another study, Inagaki and Sugiyama (1988) asked a number of questions about a person, a rabbit, a pigeon, a grasshopper, a tree, a tulip, and a stone. These were questions such as "Can x breathe?" and "Can x think?" The less the target object resembled a person, the less the children attributed psychological, mental and physiological characteristics.

What evidence is found from inheritance studies?

Children appear to know something about the fact that organisms can reproduce and therefore can pass on their properties and that objects cannot. S. Gelman and Wellman (1991) showed that children understand that a kangaroo retains the characteristics of a kangaroo even though it raised with goats.

Furthermore, if you change something about a living object, it will not lose its identity. A zebra with a black-dyed coat, for example, is still a zebra. Objects are changeable and they are able to function as what you have changed it into: you can craft an empty pack of fruit juice into a boat and as a result it can actually sail. Keil (1989) showed that young children assume that there is a change of identity when both organisms and objects are changed. Children aged 7-9 know that this only applies to objects, but not to organisms. Keil explains that this is because older children are operating on the basis of a biological theory. Here organisms are identified on the basis of underlying traits and deep causal relationships. Objects are identified by the usefulness they serve for people. Young children realized that it was impossible to change an organism into an object or vice versa (for example, a porcupine in a cactus).

Inagaki and Hatano (1993) found that Japanese children aged 4 and 5 understand that characteristics such as the color of your eyes cannot be changed, but qualities such as running speed can be changed, by training. They concluded that even young children understand the phenomenon of inheritance. Keil did not conclude this though, but this is perhaps because he used transformations like surgery in his research (for example: stripes are removed from a tiger and manes are added). Inagaki and Hatano simply made use of training. Perhaps the children did not yet fully understand the 'surgery' principle.

What evidence is found from studies of natural cause?

Furthermore, research has been done into the concept of innate features (natural cause). A rabbit cannot jump at birth, but the ability to jump is innate. Gelman and Kremer (1991) investigated this by presenting a number of behaviors of organisms and objects to children and asking whether a person had ensured that this behavior took place or whether it originates from the organism or the object itself. Examples are a rabbit that jumps, a guitar that makes music, a bird that flies and a balloon that goes up into the sky. Children over-generalized that people were the cause of behavior in less familiar organisms. They said that if salt melts in water it is caused by humans (42%). However, nobody said that the coloring of the leaves of the trees during fall was caused by people. The children were accurate in identifying behavior that was caused by a person. Sometimes they attributed an internal cause to an object, but that often occurred with self-sustaining objects: for example, a telephone that rings.

What is the historical perspective about the representation of categorical knowledge?

It was thought that children organized their conceptual knowledge in terms of thematic relationships.

What is the role of thematic relationships in organizing conceptual knowledge?

Thematic relationships are associative, for example: dogs go together with bones and bees with honey. Adults do this in terms of categorical relationships. Smiley and Brown (1979) conducted a study in which, for example, children had to choose what matched best: a bee with honey (thematic relationship) or a bee with a butterfly (categorical relationship). The young children (4 years old) had a preference for the thematic relationship. However, these were rather open questions, and if more instruction was given for sorting, one-year-old children could already sort objects based on categorical relationships. Bauer and Mandler (1989b) demonstrated this in their research. They used the same kind of design but said, "find the other one just like this one (the target)." In 26% of the trials with 16-month-old children and in 15% of the trials with 20-month-old children, there was a preference for thematic relationships. Therefore, the conclusion was that children only prefer thematic relationships under the influence of certain test instructions. As children can use thematic and categorical relationships from an early age, children would be developing both relationships at the same time. Adult research shows that both defining features and characteristic features are used to store conceptual information in the semantic memory.

How are categories represented in terms of characteristic vs. defining features?

A characteristic feature is a feature that is typically associated with the concept (grandmas are old). A definition feature is a feature that applies to all members of a group (a grandmother is always the mother of one of the parents).

It is thought that children first categorize based on the characteristic features and that there is a transition to categorization based on definition features as they learn more about the world. This is the characteristic-to-defining shift. This was demonstrated in a study by Keil and Batterman (1984). Children aged 5 used characteristic features and children aged 9 used definition features. The children did not all have the transition at the same age. The researchers concluded that this is probably due to the increase in knowledge about definition characteristics. However, research in cognitive neuroscience offers evidence for distributed mental representations. This means that concepts and categories in the brain are represented in terms of multiple sensory modalities. These units are well connected networks of neurons. There is increasing evidence for this theory.

How do categories and beliefs about the world develop according to essentialism?

Essentialism states that people make implicit assumptions about the structure of the world and the underlying nature of categories. These assumptions are represented in the categories that they develop. Categories are not detected by observing correlations. In adults there is a causal necessity, which is used to categorize. This could also be present with children.

What is the essentialist bias?

Causal necessity is particularly applicable to natural categories. Gelman (2004) states that psychological essentialism can be an early cognitive bias. She suggests that children have an innate tendency to search for hidden characteristics that allow them to categorize. Young children seem to assume that living things maintain their identity after superficial change and pass on some properties to their offspring. In addition, children value the determination of which properties members of a certain category should share.

Essentialism is not based on dichotomous development, but on two distinctive, interrelated levels: the level of observable reality and the level of explanation and cause. It is stated that the theories and core values ​​of essentialism are innate in order to guide cognitive development. The origin of this innate knowledge is unclear.

Another proposition is that young children take a vitalistic approach to perceive causality as a mechanism or as an intention. Vitalistic causality is that food and water provide vital life forces to make people and other animals active.

It may be possible that learning mechanisms used by our perceptual systems are the source of the core values ​​and causalities. The basic learning mechanisms used by the brain support the early development of the concepts described in the current chapter. As a result, patterns of inductive interference or cognitive bias (essentialism, vitalism and learning difficulties) can be explained.

What do conceptual changes in childhood depend on?

Children experience conceptual changes in their childhood. This is mainly because knowledge is taught to them. However, it is not certain whether this taught knowlegde is the cause of conceptual change. Some researchers say that this can also happen spontaneously. Carey states that the conceptual changes in childhood depend on making mappings between different domains. Objects from one system are then related to objects from another system.

At first, children do not see plants as a living creature, but as they grow older, they do. Gelman and Wellman, however, argue that children do not have a one-sided conceptual framework about concepts, but that they possess several that are different, yet linked to each other. These frameworks are used to describe the foundational domains. For example, persons are biological units, psychological units and also physiological units. Children use two levels within these frameworks: a level that contains the surface phenomena (based on attributions) and a framework that contains the deeper levels (based on relations). Goswami also supports this theory.

Conceptual knowledge is knowledge about the kinds of things in the world. This is not only about knowledge about objects, but also about actions, events and mental states. Conceptual development consists of inductive learning and categorization. Induction is a way of reasoning in which one deduces the general from the specific. Generalizing on the basis of a known example is one of the most common forms of inductive reasoning and is the basis of categorization. With induction we fill gaps in our knowledge. The ability to reason on the basis of induction is already present early in development. The focus of research is therefore on the organization of knowledge that determines the ability to categorize.

How does language develop cognitively? - Chapter 5 (1)

What are different views on language acquisition?

It was thought that language development is a separate development that is independent of the cognitive development of the other domains. Chomsky spoke about the language acquisition device (LAD), with which he indicated that children have an innate ability to learn a spoken language in whatever culture they are born. Every individual is born with knowledge about the rules that exist within a grammar and with knowledge about possible exceptions.

It is now thought that language acquisition depends on the same learning mechanisms as other cognitive developments. Examples of such learning mechanisms are imitation, the detection of statistical dependencies and the making of analogies (for example about grammar). The information that a child receives through hearing is full of cues about the phonotactic patterns of language. These are the sounds that make up a language and the order in which they can be combined. Furthermore, cues to word boundaries (for example the rhythm of speech) and cues to the emotional content of speech (for example emphasis and volume) are also extracted from auditory perceptual information.

The social aspect with, for example, the parents is very important to language learning. The situation, the tone and the facial expressions are all cues for a child about what is being spoken about. Two things promote the language acquisition of a child: (1) "motherese" or infant-directed speech (IDS), and (2) a child's innate tendency to social contact and attachment. IDS is an exaggerated way of speaking that emphasizes word and sentence boundaries. Moreover, IDS makes the segmentation of the word flow easier for a child.

What is phonological development?

Phonological learning is about the sounds of a language. It consists of two aspects:

  1. learning about certain patterns of sound combinations that are allowed in a certain language. This allows the brain to make phonological representations of the sound structure of individual words.

  2. learning to produce these words yourself.

What is categorical perception?

The term phoneme is used for the individual elements that make up words in a language. "Bit" and "bat" differ from each other in the middle phoneme. Languages ​​are based on two types of phonemes: vowel phonemes and consonant phonemes. Children have the ability to learn 200 vowels and 600 consonants, but most languages ​​use only a small number of phonemes (about 40 English for example). A child learns these phonemes in its first year of life and at the same time the child loses the ability to learn phonemes of other languages.

There are many sounds that are not identical, but are heard as a 'B'. Other sounds are heard as a 'P'. There is a point where the sounds that are very similar are no longer heard as "B" but as "P". This is categorical perception. This is already developing when children are one month old (Eimas et al., 1971). The categorical perception of languages ​​that do not have the same phonology as the mother tongue decreases with age. At the age of six to eight months, children could still hear the difference between the non-American sounds 'KI' and 'QI', at an age of eight to ten months old, only 57% could do this, and at the age of 10 to 12 months old, only 10%. Adults can no longer do this (Werker & Tees, 1984). The same researchers conducted a longitudinal study on this topic and it showed that the possibility of distinguishing sounds that do not belong to the mother tongue continues to decrease during the first five years.

Children already know acoustic boundaries that divide a language into phonological categories that can occur in a language from birth. However, this division is not yet very specific and it is therefore necessary that children learn more about this after birth. Children usually come into contact with one language and therefore specialize in the sounds of that specific language.

The ability to distinguish phonemes in another language disappears and children develop prototypes of phoneme categories in their own language. At the age of 1 year this is already quite developed and this indicates a possible sensitive period for learning language.

Forming a prototype in language acquisition happens just like forming prototypes in the physical world. Children detect characteristics that occur together and thus form a prototype. Sounds that we would categorize as "B" or "P", but that are relatively different from the prototype, are called allophones.

The magnet effect means that sounds that resemble each other are classified in the same category, and sounds that do not resemble each other are not, even though there may be no physical difference between these sounds. Prototypical sounds in a language thus act as a magnet for perceptually similar sounds, whereby these sounds are considered to belong to the same category.

Kuhl et al. (1992) proved with their experiment that linguistic experience prevails at the age of six months compared to phonetic perception. Social interaction also plays a role in the development of phonology. Children don't learn just by listening passively. Kuhl et al. (2003) looked at the decrease in sensitivity for the phonetic aspects of languages ​​other than the mother tongue that takes place between six and twelve months. They let nine-month-old children play with a Mandarin Chinese adult for five hours, exposing them to Mandarin Chinese. The adult used child-centered speech, often used the child's name, and made frequent eye contact with the child. In the test phase, they were tested to see if they could recognize the phonological differences in sounds in Mandarin Chinese. Children who had been with a Mandarin Chinese adult could do this significantly better than children who had been with an English adult. Their results were so good that they matched the results of babies in Thailand who had heard Mandarin all their lives.

Kuhl et al. (2003) did the same in a follow-up study, but they used videos. In the videos, it seemed as if the adult directly addressed the child. The children were interested in the video, but afterwards had no better results than the control group who had seen the videos in English. Kuhl et al. concluded that live speakers offer social cues that facilitated language acquisition. Children learn language with communication as purpose and simply being exposed to auditory sensory information is not enough to trigger perceptual learning of phonology.

What does phonotactic learning mean?

Words consist of series of phonemes and children must learn which phonemes belong to one word and in which order. This is based on the phonotactics. These are the rules about the series of phonemes that are used in a language to construct words. This information helps determine where a word begins and ends: in English, words end with –ant, but never with –atn. The combination –atn will therefore consist of two different words (e.g. at night). This is the phonotactic probability and is a good cue to distinguish words.

Children can divide a word stream into words at the age of seven months. Saffran et al. (1996) did an experiment with eight-month-old infants to investigate this. They came up with three words that consisted of three syllables (bidaku, padoti and golabu). These words were repeated for two minutes in one stream without pauses. The phonotactic probabilities were 1.00 for syllable pairs that occurred within the words (bi-da) and 0.33 for syllable pairs that occurred between the different words (ku-pa).

After the learning phase, the three words the children heard were mixed with new words consisting of three syllables (for example, dapiku). Parts of these new words had occurred in the learning phase, but the phonotactic probabilities were 0.00 (da-pi had never occurred, because da was always followed by ku). Children showed dishabituation for the new words. In a follow-up experiment (1996) the new words became combinations of the old words (dakupa: occurs in bidakupadoti). However, the probability of ku-pa was 0.33 and the probability of da-ku was 1.00. The children again showed dishabituation for the new words and had therefore mastered phonotactics through statistical learning.

What does infant-directed speech, rhythm and prosody look like?

The higher tone, emphasis and intonation of child-centered speech probably help the child to pick up words from the word stream. We do not know exactly how this works, but infants prefer to listen to motherese. Fernald and Mazzie (1991) argued that if the function of child-oriented speaking was merely to attract attention, there would be no relationship between the prosodic structure (the melody and the rhythm of the language) and the linguistic structure. In that case, adults should talk to children and adults with the same prosody.

Fernald and Mazzie investigated this with a story in which a new item of clothing was introduced on each page. This story was read to infants and adults by adults. The first time the new word was introduced, in 76% of the children the word was stressed, compared with 42% of occasions when the story was told to another adults. At the second time it was 70% and only 20% respectively. With children, mothers often place the new words at the end of a sentence in order to stress it. New words were most common in the highest part of the sentence (in terms of reading tone) or at the end.

Jusczyk et al. (1999) investigated whether children use prosodic rules to distinguish words from a word stream. In English, words that consist of two syllables often have stress on the first syllable (doctor, patient). The syllable on which stress is placed is usually stronger, louder and has a higher tone than the other syllable. Infants of seven months old distinguish these strong-weak patterns from a continuous flood of words and recognize the words as words that consist of two syllables (they responded to the entire word doctor and not just doc). At this age the children do not recognize the weak-strong patterns of stress. They got the word taris from 'her guitar is too fancy', because it has stress on the first syllable and guitar on the second. At an age of 10.5 months they can distinguish between this.

If words have the same emphasis and stress pattern, but differ in one phoneme, children do not recognize the difference between these words. Swingley (2005) investigated this and made use of a head turn preference procedure, in which spoken stimuli are presented to the child as long as they look at a flashing light. Children look at the light longer when the stimuli are known. If words differed in the first letter (dog and bog), eleven-month-old infants were able to spot the difference, but not if they differed in the last letter (dog and dob).

Infants of four days old are already sensitive to prosodic and rhythmic patterns. Mehler et al. investigated French children of four days old who had already been in contact with the rhythm of their language in the womb by telling them a story in French and in Russian. The children preferred the French story. If the tapes were played backwards, the children made no distinction. The absolute parameters, such as the height of the voice, were preserved, but the relative cues (melody and intonation) were not. Children distinguish the two languages ​​based on prosodic and rhythmic information. Vowels in syllables are important for the prosody of a language and are cues for syntax. Syntax is the knowledge about how words can be combined into sentences. Morphology is the set of rules on the internal structure of words.

What does early phonological production look like?

The vocal development goes through a number of stages. In the first two months, children produce comfort sounds that have a normal phonation and resemble vowels. At two to three months, children produce phonetic sequences that are the precursors to the use of consonants. This is the gooing stage. This is followed by the expansion stage, in which children from four to six months old start to babble. In the canonical stage whole syllables are used for babbling and this babbling is called canonical babbling. These syllables can function as building blocks of words (mamamama). At the time, infants are seven to ten months old.

The same order of chatting occurs in different cultures. It starts with the /d/b/m/n/g/t, because they are easy to produce. The relative frequency in which these easy to say consonants occur corresponds to the native language. The rhythm of chatting also corresponds to the rhythm of the mother tongue.

Oller and Eilers (1988) investigated babbling in deaf infants. The canonical chat developed between 11 and 25 months and the chat of deaf infants was different from that of children who could hear. Infants therefore need their hearing to develop babbling.

The linguistic hypothesis of babbling states that the production of structural rhythmic and temporal patterns of language is a crucial part of language acquisition. The motor hypothesis of babbling states that the rhythm of babbling is determined by the physiological movements of the jaw.

Petitto et al. (2004) investigated these two hypotheses by comparing infants of parents who could hear to a group of infants of parents who were deaf, but who could hear themselves. The last group learned sign language from their parents. The researchers argued that a kind of "babbling" on their hands should arise if the linguistic hypothesis were true. If manual babbling also occurred in these children, this would correspond to the motor hypothesis. Manual babbling are hand movements of deaf children who learn sign language that corresponds to the audible chatter of children who can hear. The results showed that babbling is specifically linguistic. When children are raised with language, they babble in sounds and when they are raised with gestures they babble in gestures.

What are the results of cognitive neuroimaging of phonological development?

Using fMRI, it has been found that there are two processing paths when processing language (just like when processing visual stimuli). The first anterior pathway is analogous to the visual what-pathway and uses information about vocalizations and acoustic-phonetic sounds. The second posterior pathway, analogous to the visual where-pathway, processes the articulatory information. Therefore, this pathway looks at the location of sounds and how they are formed. Speech is processed as sound and as action.

The mismatch negativity (MMN) is measured with EEG. If the brain repeatedly hears a 'P', it habituates to this and the activity is less. When a 'B' is added (a mismatch), the brain reacts with an increased negative potential of 270 ms after the stimulus. Babies exhibit a large MMN when they sleep, because the auditory system is not switched off. The EEG activity of infants corresponds to that of adults. Dehaene-Lambertz and Gliga (2004) stated that this provides evidence for a neural basis for phonetic perception in the posterior temporal lobe. When vowels are used that do not occur in the mother tongue of children, a much lower MMN can be found in 12-month-old infants than when vowels do occur in the mother tongue. At the age of six months this is not yet distinguishable.

MMN was also used in the investigation of stress patterns. Weber et al. (2004) examined German children. In German, the emphasis of words with two syllables is 90% of the time on the first syllable. When infants are four months old, they do not yet distinguish between words with stress on the second syllable from words with stress on the first syllable. When infants are five months old, they start to show a large MMN.

Dehaene-Lambertz and colleagues (2004) demonstrated with fMRI that the cortex of a three-month-old infant is already structured into several regions that play an important role in speech production. Specific activity in the left temporal lobe was observed, especially in Heschl's gyrus. This is an area that is important in speech production in adults. In adults, the left inferior frontal gyrus (Broca's area) is related to the production of speech, repetition in silence, and short-term memory. In children, this was the only area of ​​the brain that was sensitive to repetition of sentences, resulting in a stronger response when a sentence was heard for the second time. There was also greater activity in the angular gyrus and the precuneus. Dehaene-Lambertz and colleagues concluded from their follow-up study (2006) that a memory system for speech and other auditory stimuli is already active at this age.

With the help of ERPs (event-related potentials), Mills et al. (2004) investigated the possibility of distinguishing from familiar words and new words by fourteen and 20-month-old children. Three categories of words were used. Familiar words were shown in the first category.

In the second category, non-existent words were shown that were phonetically similar to the words from the first category. In the third category, non-existent words were shown that were not phonetically similar to category one. Fourteen-month-old children made a distinction between category one and three and between category two and three, but not between category one and two. Children of 20 months old made a distinction between all three categories. These results showed that the phonological representations of familiar words at 20 months of age were well developed.

What is lexical development?

The lexical development is building a vocabulary and learning what words mean. From an early age, children are confronted with many verbal statements from their caregivers ('look…'). Children from higher SES families hear 487 of these statements per hour and children from lower SES families only 178 per hour.

Children learn words around their first year. At the age of sixteen months children know on average 55 words, at the age of 23 months 225 words and at the age of 30 months 573 words. At the age of 6, children have a speech vocabulary of 6,000 words and understand 14,000 words. The first words consist of words of clearly present objects or persons (mama), clearly present categories (cookie), words for actions (on, away), words for social interaction (bye-bye) and words for repetition (more, another time).

At the start of language development, children understand more words than they can produce (comprehension precedes production). Understanding words develops much faster than the production of words. Of the first ten words of which a child understands the meaning, 50% refer to an action.

Fenson et al. (1994) used the Child Language Checklist to investigate the vocabulary development of children. Three key findings are: (1) understanding words develops around the age of eight to ten months, (2) producing words develops around the age of eleven to thirteen months and (3) there is a wide variety in the lexical development between different individuals. They also found that vocabulary development is gradual. This is contrary to the idea of ​​the vocabulary spurt. This is a spurt in which children suddenly learn many new words because they understand that words refer to things. Further research has rejected the idea of ​​this vocabulary spurt.

Another result of the research was that after the skill of word comprehension emerges, the development of communication gestures and routines takes place. These gestures usually take place after verbal input, such as "goodbye!" Then gestures develop about recognizing the function of objects (holding a telephone to the ear). Next comes the verbal naming of things. The use of gestures is a way of production and not a way of understanding. The gestures thus form a cognitive bridge between lexical comprehension and lexical production. Gestures develop between ten and eighteen months and then decrease. At the age of 21 months, language production exceeds the use of gestures. At the age of ten months children already use gestures in combination with proto-words (for example: accompany gestures to close the door with the protowork "shuh"). These combinations become more complicated at the age of eighteen months. For example, if children want their mom to choose the book, they say "book," and at the same time they take her hand and place it on the book.

How do infants learn words?

Infants of 4.5 months old can recognize their name and have a detailed phonological representation. Mandel et al. (1995) investigated this using a head turn preference procedure. Infants looked at the light significantly longer when they heard their own name. If a name with the same stress structure was presented, they no longer looked. This proves that they also master the phonological representation. When words are shown in combination with their own name, children recognize these words better (Bortfeld et al., 2005). This effect was the same when words were shown with 'mama' at the front. Bortfeld et al. concluded that infants use their own name and mother's name as a key point in a stream of words. Infants identify new words through top-down processing. This is a strong language learning mechanism. First, infants can only process bottom-up and the learning mechanisms are only based on perceptual characteristics of input. Infants can also use top-down learning mechanisms at the age of six months.

What are the cognitive expectations of language and categories?

If children understand that nouns refer to objects, they will also develop more complex categories of words, such as verbs and adjectives. The study by Baldwin and Markman (1989) assumed that children showed more attention to an object that was labeled, because then they understood that the word (sound) produced by an adult belongs to an object. A child was allowed to play with unknown objects (for example a snorkel) and if the name of the object was told, infants of ten months old already showed more interest in these objects. A follow-up study looked at the difference between just pointing to the unknown object or pointing and labeling it. The infants showed more interest in the objects that were both labeled and pointed to. The labeling of objects thus ensures a better understanding of the word-object relationship.

Waxman and Markov (1995) investigated the emergence of categories in the word-object relationship. In one condition, they showed infants four objects from the same category (for example, four different cars) and said, "Look, a car." Nothing was labeled in the control condition ("Look at this"). It was predicted that infants would be less interested in the stimuli if they realized that the four things belonged to the same category. This hypothesis was true, but only for more difficult names (for example if "Look, a vehicle" was mentioned). Xu (2002) claims that labeling plays an important role in determining that objects differ from each other. Waxman and Braun (1995) concluded from their research that assigning the same name to a set of objects from the same category contributes to children categorizing objects.

What are linguistic influences on early concepts?

In the case of overextension, a child uses a certain label for naming different objects. For example, he uses the word "dog" not only for dogs, but also for cats, lions, and so on. This usually happens around the age of 2.5 years. The hypothesis to explain this is that the overextension proves that children have fewer differentiated conceptual categories than adults. If more words are learned, the children must organize them into a group: for example, long, thin objects.

In the study by Fremgen and Fay (1980), children aged from 1 year and 2 months to 2 years and 2 months old were tested. In the production test, pictures of objects or animals were shown for which the mother had previosly reported overextension. The child had to name these pictures. In the comprehension test, four pictures were shown: two of them were irrelevant to the label being tested, one of them was a picture used for overextension and the other was the label (in dogs, for example, a picture of a dog, a cat, a car and a vase). Children always performed accurately in this test. Fremgen and Fay concluded that children do know the difference between the objects or animals. In the absence of words to name objects, a label of an object or organism that resembles it is used.

What is fast mapping?

By the age of around 2 years, children learn approximately ten words a day. They use fast mapping for this. This is the ability to form quick and rough hypotheses about the possible meaning of new words. The context in which the word is used and its position in the sentence are used. In the experiment by Heibeck and Markman (1987) it was found that when children are asked, for example, to 'grab a thick turquoise book that lies between the other red books' (and they don't know the word turquoise yet), they grab what they need by clues about the shape, color and structure. Girls performed better than boys. The understanding of a category was tested by asking a child: "this material is not soft, but ...". For example, if the child responded with "rough," it had conceptual understanding, but if it responded with, for example, "blue," the answer was not taken from the same category, so the child did not yet have that understanding. Older children understand this better and at the age of 4, 96% of the children perform well. Performance on shape words was the easiest, then color words and then texture words. When asked to reproduce the newly learned words, the performance was much poorer: children understood more new words than they could produce themselves.

Markson and Bloom (1997) examined whether, in addition to the fast mapping of words, fast mapping of facts could also take place. Children of 3 and 4 years old could do this. However, fast mapping is not a mechanism that works with every retention task. It seems to have the most effect for new information that is given linguistically.

Dogs can also learn that spoken words can be linked to certain objects or organisms. They too can learn about specific word-object relationships through fast mapping and remember these relationships for at least a month.

How can cognitive neuroimaging contribute to studying lexical development?

Studies on lexicality make extensive use of an EEG response: the N400. The amplitude of the N400 plays a role in the integration of a stimulus in the semantic context. If a sentence is shown that is semantically incorrect ('the storm is ironed' instead of 'the shirt is ironed') an increased negative activity occurs after 400 ms. This is the N400. Friedrich and Friederici (2004) investigated the N400 in children. They showed familiar pictures and then labeled the picture with a correct or incorrect label. In children aged 19 months, a non-congruent label resulted in an increased negative activity at around 700 ms. An earlier negativity was also measured (at 150 and 400 ms), but this negativity was greater with congruent words. Friedrich and Friederici thought this was a priming effect: the children expected to hear a word with certain phonemes.

For 12-month-old children, the negativity did not occur with non-congruent words, but the previous activity with congruent words did. This result was explained by arguing that children of this age do have lexical-semantic knowledge about the words used in this study. If pseudowords were used, these children were also able to distinguish real words from pseudowords.

The N400 could be a neural marker for a potential risk of subsequent linguistic impairment. Indeed, a longitudinal study by Friedich and Friederici (2006) found that children who showed an N400 at the age of nineteen months had normal language development at the age of 30 months. Children who did not show this at the age of 19 months had deficits.

How does grammar develop?

Rules about syntax and morphology are implicitly acquired by children, by listening to adults. Children first start with single words, but around the age of 20-24 months they add extra words in order to add extra information (no bath, and wet doggie). These two-word combinations are increasingly used between the ages of 2 and 3 years. It used to be thought that the grammar of a language is universal, but that is not true.

Children between the ages of 2 and 5 make grammatical mistakes. An example is the use of the past tense -ed in English. Children over-regularize this: they apply it to every verb, even if it doesn't belong. The development of this is u-shaped. This means that children did well before they started to regularize (before they were 2.5 years old). After this comes the period of over-regularization in which many mistakes are made, and then the child finds out that this rule does not apply to irregular verbs and they return to the appropriate use. This development used to support the idea that grammatical development is dependent on the acquisition of rules.

However, a study by Marcus et al. (1992) showed that over-regularization is not as common as people think. Especially with frequently used irregular verbs, the perfomance of the child is mostly correct. Marcus et al. suggested that children make overregularizations because they begin to mark tense intentionally.

Berko (1958) investigated the implicit knowledge of grammatical rules by using an analogous task based on nonsense words. He concluded that young children have a good knowledge of morphological rules. When forming plural there was a large variation in performance of the children.

The formation of new non-existent words by children is also a sign of their knowledge about word formation paradigms. Children invent new words: You have good earsight. Verb variants are rarer than variants with nouns.

The first studies into the improvement behavior of parents with regard to these errors showed that the improvement was not very common and if it did, it was not successful. More recent research shows that adults do correct the mistakes, but by reformulating the child's statement and not by correcting it. This happens with every error, whether the errors are grammatical, morphological, phonological or semantic. This occurred in different languages.

Parents also use expansion to improve children. They repeat the child and at the same time form the good form (Child: "Muffy step on that." Mother: "Who stepped on that?" Child: "Muffy stepped on it."). A study into the effect of this showed that children who had been treated with this extension method had made just as much progress in terms of language as children who had not received this treatment. Children whose parents had a conversation with the child (child: "I got apples." Mother: "Do you like them?") made the most progress.

The syntactic development of children can also be measured by the mean length of utterance (MLU). The length of a statement is measured in morphemes (dog = 1, dogs = 2 and so on). This term was first used by Brown and Hanlon (1970) who followed the language development of three children.

Fenson et al. (1994) did research and discovered that sixteen-month-olds use morphemes such as –ed and –s. At the age of 22 months they use –s, -ed, -ing, and possessive 's'. Fenson et al. suggested that the complexity of a sentence would be a better measurement than MLU, especially after passing the age of 30 months. The best predictor of grammatical complexity is the size of the vocabulary.

Tomasello et al. (2007) concluded from their research that children create "pivot" words that can be combined with many words, producing a new sentence (more milk, more food, more candy). It seems that children acquire and remember small parts of language that have a certain communicative function. Only later do children form a more abstract picture of language. Tomasello proposes that the most important learning mechanism is the change to reasoning by analogy. The acquisition of grammar depends on learning. The development is slow but steady. Children learn from input and use the same learning mechanisms as in the development of the other cognitive domains.

What is the development of pragmatics?

Language is about communication and the pragmatics of language development is about learning how to communicate competently. It about the rules of how to use language. For example, when you have a conversation, you need to take turns. Children do not put themselves in the conversation partner's place and can sometimes suddenly change the subject without announcing this. You must be able to use the language socially and properly. Social means being polite when you talk to others. In his experiment, Dale (1980) found that knowledge of pragmatics is strongly age-related. First there was the naming of things, then greetings and then comments on objects and their attributions. Subsequently, requests about the here and now emerged, followed by confirmation and denial. Finally, the reference came to the past and future, and requests for absent objects.

According to Gleason (1980), learning the social aspect of language is part of the development of social cognition. Children learn to say some things as routine, for example out of politeness (such as 'thank you'). It makes no difference to adults whether children mean it, as long as they have said so.

What are different views on language acquisition?

It was thought that language development is a separate development that is independent of the cognitive development of the other domains. Chomsky spoke about the language acquisition device (LAD), with which he indicated that children have an innate ability to learn a spoken language in whatever culture they are born. Every individual is born with knowledge about the rules that exist within a grammar and with knowledge about possible exceptions.

What is causal reasoning? - Chapter 6 (1)

Causal reasoning is a domain-general skill that plays a very important role in cognitive development. The causal structure of events is important in the representation, interpretation and recall of these events. Causal reasoning is based on perceptual information, because the perceptual information gives cues about the intentionality of people. This promotes social cognition. In addition, it also provides information about living and non-living things. Children recognize relationships, detect statistical dependencies and connect cause and effect through perceptual information. Besides the information that children pick up from their environment, they will also form their own frameworks for explaining events, even without being able to see them. This happens once a child is capable of manipulating causes. If children develop such a framework about events, they can also make predictions and a child can control such an event.

Some frameworks about events emerge from consideration of relevant perceptual variables. With physical causality, when key variables are unobservable, we need causal mechanisms to predict the effects of events. Causal mechanisms are the causal characteristics that determine the effects that an object may have on another object. The causal mechanisms that are being used depend on previous experiences.

How does reasoning about cause and effect develop?

Children aged 3 have experienced many physical causes and consequences. The way to measure their understanding of causal reasoning is to ask a child if it knows or understands that a cause can change objects by changing them from one state to another.

How do children reason about causal transformations of known objects?

R. Gelman et al. (1980) investigated cause and effect relationships in children aged 3 and 4 by constantly having them look at rows of pictures from left to right. The left or right picture was always the object in its original form and the other picture was always the object being transformed. The middle image indicated the cause. For example, on the right: a cup; middle: a hammer; left: a broken cup. The children were shown the picture sequences, with one missing picture and they had to choose the right one from three alternative pictures. Children aged 3 chose the right picture in 92% of the cases and children aged 4 in all cases when the middle picture was missing. If the relationship was the other way around (from broken cup to intact cup by means of glue), 75% of the 3-year-olds chose the right picture and 100% of the 4-year-olds. If the last picture was missing, the 3-year-olds performed slightly poorer (83% for sequences from the original form to the altered form and 58% for when the relationship was the other way around) and if the first picture was missing even worse (66% and 58 %). The 4-year-olds always scored 100% correct. Still, the overall performance was significantly above chance.

How do children use reversible reasoning about causal transformation and known objects?

In a second study the reversibility of causal reasoning was investigated. The children were asked to read the sequence from left to right and from right to left, with the middle picture always being the one missing. So they had to assess causes of the same pairs of objects in two different ways. Children aged 3 scored correct in 49% of the cases and children aged 4 in 75% of the cases. Gelman said this was because 3-year-olds tend to use their own preferred causal order in the test. Gelman did claim that the representations of 3-year-olds were abstract enough to understand reversibility.

Das Gupta and Bryant (1989) criticized the study because they argued that the children could also use associative reasoning instead of causal reasoning. If a child saw a broken cup, it was easier to think of a hammer because there is an association between a broken cup and a hammer. Associative reasoning is seen as a less complicated cognitive skill than causal reasoning. They argued that real causal reasoning depends on whether children understand the difference between the initial state and the final state. In their research, the middle picture was always missing and the 3- and 4-year-old children had to choose the cause.

The cause (for example, a hammer) of the shape that was not intact (for example, a broken cup) was sometimes the cause of the images (from an intact cup to a broken cup), but sometimes it was not (for example, from a broken cup to a wet broken cup by means of a bucket of water being thrown over it). Das Gupta and Bryant stated that if a child always chooses the hammer, he has no understanding of causal reasoning. From the results they concluded that 3-year-olds were often distracted by the salient condition of the object (so broken was more salient than wet), so they drew the wrong conclusions about the cause of the last picture.

What is the influence of the salience of non-canonical states in early causal reasoning?

In a follow-up study, Das Gupta and Bryant investigated which sequence (i.e., either from intact cup uto broken cup - the caconic order - or the from broken cup to the intact cup - the non-caconic order) was best to test the ability of children to make causal inferences. The salience of the broken cup could allow the child to choose the hammer in both the caconic and non-caconic order. Children aged 3 had more difficulty with the non-caconic order (47% correct) than with the caconic order (88% correct). They concluded that determining correct relationships between cause and effect develops between 3 and 4 years and is not yet present at the age of 3. In this study, however, images were used and not real situations. Research shows that when real situations are used, children can draw the right conclusions about cause and effect from the age of 2. This will be discussed later.

It is also demonstrated that adults sometimes have difficulty seeing a causal relationship if they have to reason from effect to cause (from fire to spark instead of spark to fire). The causal directedness according to Fenker et al. (2005), is a part of how information about causal links is stored. More research into this is needed in children.

What is reasoning on the basis of causal principles?

Hume (1974) defined the causal principles. The priority principle states that a cause precedes an effect or occurs simultaneously with an effect, but that a cause can never occur until after an effect. The covariation principle states that the cause and the effect must be systematically interrelated. The temporal contiguity principle states that the cause and effect must be contiguous in place and time, and the similarity principle states that the cause and effect must be similar to a certain degree (for example, that a mechanical effect should have a mechanical cause).

What is the priority principle?

Bullock and R. Gelman used a device in which they could put marbles on two sides. If a marble was put in, a devil emerged from a box. It seemed as if this was caused by the marble, but actually the appearance of the devil from the box was controlled by a foot pedal. During the test phase, the experimenter let two dolls slide a marble into one of the tunnels: one marble before the little devil jumped out of the box and one after he had jumped. The children had to determine which marble had caused the jump. Children aged 3, 4 and 5 were successful in 75%, 88% and 100% of the cases, respectively. Then one tunnel was separated from the device, so that it was no longer in contact with the part where the devil came out. Nevertheless, the devil jumped after putting a marble in this part of the tunnel. Children aged 3, 4 and 5 of age attributed the jumping of the devil to this marble in 75%, 94% and 100% of the cases. Temporal cues were therefore more important to the children than spatial cues. The children understood the priority principle. However, they were surprised and asked if it was a trick.

However, Schultz (1982) stated that these results could also be caused by children looking at what is causally relevant to the outcome. He concluded from his own research that children see spatial cues as the cause rather than temporal cues if the spatial cues are more relevant to the outcome of the event. Hume's principles cannot simply be treated separately from the objects and relationships that play a role in a causal event.

What is the principle of covariation?

The principle of covariation states that if there are various causes for an effect, then the cause that covaries regularly with the effect is likely to be the true cause. For example, if two levers can cause a light to switch on, then it is caused by the lever that is always activated with the light switching on.

Shultz and Mendelson (1975) used such examples in their research and showed the children a number of situations in which the levers were raised and lowered and a light being switched on or not. Children aged 3 and 4 were able to choose the right lever as the cause of the light, and the researchers concluded that children from the age of 3 master the covariation principle.

What is the temporal contiguity principle?

The temporal contiguity principle is closely linked to the covariation principle and the priority principle. The temporal contiguity principle, however, states that causes and effects are linked by an intervening series of contiguous events. If there is a rationale for a delay between cause and effect, this principle is still intact. Mendelson and Schultz (1975) concluded with their research that if there was no physical rationale for a temporal delay, children attached more value to the cause to the temporal contiguity principle than to the covariation principle. If they were unable to see a temporal delay, they attributed the cause to the covariation principle.

What is the principle of the similarity of causes and effects?

If there is no information about covariation or temporal contiguity, the principle of similarity of cause and effect is used. If there is a heavy lever and a delicate lever and a loud noise or a soft noise is heard, it is assumed that the loud noise is caused by the heavy lever. Schultz and Ravinsky (1977) investigated this among children and also focused on whether it would change when it conflicted with the covariation principle and the temporal contiguity principle. Children aged 6, 8, 10 and 12 were tested and if there were no covariation or temporal contiguity cues, all ages used the similarity principle. If there were cues of covariation, only the 10- and 12-year-olds abandoned the similarity principle (adults do this too). The younger children were confused about which principle they should use and therefore they did not use the principles consistently. If there were cues of temporal contiguity, the results were the same; only 6-year-olds showed a confused pattern of responses.

This research shows that temporal information is perhaps more important than information about covariation. Temporal information provides information about the causal structure and the covariation about the strength of the causal relationship. Causal structure is more important than the strength of the causal connection, because we first have to look at whether there is a relation at all (does smoking cause lung cancer?) and then how strong this relation is (the more cigarettes, the sooner cancer?).

What are causal Bayes nets?

Causal Bayes nets are causal structure algorithms that underlie covariation data. An example of such a causal Bayes net is that A is the cause of B and that this is the cause of C (A -> B -> C) or that B causes both A and C (A C). The causal Bayes nets can be used to predict the effects of interventions in simple and complex causal structures.

Gopnik et al. (2001) stated in response to their research that if young children reason according to the causal Bayes nets, they only make causal inferences about real relationships. They examined whether children aged 2 to 4 years are able to reject false associations. They left objects (which they called blickets) on a machine. If there was a blicket on the machine, it made music. The children had to determine whether something was a blicket or not. In reality, the music machine was operated by the experimenter. In the one-cause test, block A was first placed on the machine and music was played. Then block B was put on the machine and it made no sound. If block A and block B were put together on the machine, no music was played. Block A was therefore a blicket and block B was not. In the two-cause test, block A was placed three times on the machine and three times the machine played music. Block B was also put on the machine three times; the machine played music twice and one time it did not. A and B are therefore both blickets.

Two-year-olds understood that if blocks A and B were on the machine together and it played music, but block B did not activate the machine by itself, that only A could be a blicket. Gopnik et al. (2001) examined in a follow-up study whether children also understood how to stop the machine. They left block B on the machine, after which no music followed.

Block A was added and the machine started to play. The children were asked to stop the machine. The majority only removed block A. If it had been shown in advance that block B did activate the machine, both blocks were removed from the machine. Gopnik et al. (2004) also demonstrated that children make use of rejecting false associations in biological and psychological causal relationships.

Blaisdell et al. (2006) demonstrated that rats understand the relationship between observations and interventions in causal reasoning on the basis of causal Bayes nets.

How can causal chains be understood?

So far, only studies have been discussed that deal with one set of causal relationships (A -> B), but no studies with more complex sets of causal relationships (A -> B -> C). Understanding sets of causal relationships with three items is more difficult, but crucial for making accurate causal conclusions. In A to B to C, A is the cause of C, but there is no direct causal relationship between A and C. B functions as a causal mediator.

How can mediate transmission be understood?

Schultz and colleagues (1982) investigated whether 3- and 5-year-olds understood that B could function as a mediate causal event. They used a device with tubes. Only a tennis ball could roll in the first tube and only a golf ball in the second tube that was connected to the first tube. However, there was an arch between the two tubes that prevented the golf ball from rolling from one tube to the other. The golf ball could only roll down and allow the other lighter ball to roll at the end of the tube if it was on the far side of an arch. The children had to choose which lane would cause the light ball to roll. Children aged 3 chose correct in 69% of the ten trails and children aged 5 in 86%. Most mistakes were made in the first and the third trial. From the fourth trial, both ages were consistently correct in their answer.

Baillargeon and R. Gelman (1980) also investigated the concept of B as a mediate causal event. They used the 'Fred-the-rabbit' device where (A) a rod had to be pushed, (B) wooden dominoes had to be knocked over, which (C) landed on a lever. This caused Fred the Rabbit to fall into his bed from a platform. The children had to explain how Fred would end up in his bed. The children were first shown what the entire device looked like. Then the dominoes were covered. This resulted in a space of approximately one meter between the rod and the platform on which Fred stood. The children of 4 and 5 years correctly predicted that Fred could be put in his bed by pushing the rod.

Then a small rod was shown that could not reach the first domino and a longer rod that could reach the first domino. The children were shown that both were unable to get Fred in his bed. Fred could not get the longer rod into his bed (due to a trick). The children were asked to explain why this happened. The children gave the correct explanation and said that for the long rod there was something that prevented the long rod from reaching the domino. The children gave the correct explanation for the smaller rod: the smaller rod could not reach the first domino. In a follow-up study (1981), Baillargeon and Gelman investigated whether 3- and 4-year-olds could also predict the correct outcome if something was changed about the first condition (A, the rod) or the middle condition (B, the dominoes). The children were accurate in both predictions. Children understand mediate transmission (A -> B -> C) by 3 years of age.

Concept of logical search

Another way to investigate children's understanding of causal constraints on sequences of events is by using search tasks. For example, if you have lost your keys and you know you still had them when you went to the supermarket, but not when you went to school, you know you lost them somewhere between the supermarket and school. This is an example of a logical search task. This task involves effective searching based on the understanding of the logic of a situation. Wellman et al. (1979) investigated this by taking children around eight different locations in a playground.

The experiment supervisors took a photo at location three. At location seven they wanted to do that again, but at this location they found out that the camera was gone. The children were asked to help him to find it. The children aged 3 and 5 only significantly searched in the critical area (area 3 - 7). However, half of the searches were focused on area three, so it was not clear whether the children understood that they had an equal chance of finding the camera in each location between 3 and 7.

Sommerville and Capuani-Schumaker (1984) conducted a study in which children had to look for a Minnie Mouse doll. There were four possible locations. Two locations always had the same chance of finding Minnie Mouse. In some trials it was more logical to go to the next locations to find Minnie Mouse and in other trials it was more logical to go to the previous locations to find her. The 3- and 4-year-olds were all able to go to one of the possible two locations during their first search. In their second search, however, they did not always go to the location where it was supposed to be. Follow-up research showed that even children aged 2 go to one of the possible two locations during the first search, but that all children younger than 4 years appear to search at only one location at a time. So, they search outside the critical area. It is not clear whether this is a general problem in understanding causal implications or a specific problem with hiding and finding.

What is reasoning based on the scientific method?

Children face difficulties when they have to rule out potential causal variables as the cause of an event. This requires the scientific method. Here, hypotheses are tested in a systematic way and potential confounding variables are controlled. Evidence must also be sought that could disconfirm the hypothesis. Scientific thinking is the way of thinking that distinguishes and coordinates theories and evidence and the evaluation of hypotheses through evaluation and experimental research. Children only possess this way of thinking around the age of 11 or 12. For younger children, the background information they have makes it difficult to accept confounding evidence from purely statistical terms.

Coordination of theories and evidence

In a study by Kuhn et al. (1988), answers based on evidence were given by 11-year-olds in 30% of the cases and by 14-year-olds in 50% of the cases. The level of 14-year-olds corresponds to that of adults. Children can only reason scientifically if they are able to let go of their background knowledge (that apples are healthy) and to base their answers purely on scientific evidence (that apples can cause a cold, as presented in the evidence in Kuhn's study).

Kuhn et al. (1988) also reported that children make inclusion errors. These are mistakes that are made on the basis that variables only covary with the outcome on a single occasion, while it is assumed that it always happens. In the study, for example, diet coke is sometimes present in children with a cold and not in others. Diet coke is believed to cause a cold, while it is also present in the no cold condition. 47% of 11-year-olds made this mistake and 65% of 14-year-olds. Research with adults shows that they also make these mistakes and that there is a conformation bias: they seek out for evidence that supports their own beliefs or hypotheses.

How can hypothesis be tested?

Sodian et al. (1991) examined children aged 6 and 8 and examined whether they were able to distinguish between a good and a bad test in hypothesis testing. They told them that two brothers knew for sure that there was a mouse in the house, but one brother thought it was a big one and the other thought it was a small mouse. To test it, they planned to place a box with cheese with a small opening, so that only the small mouse could enter, or a box with a large opening, so that both the small and the big mouse could enter. The children were asked which box they should use. The majority of children of both ages opted for the box with the small opening.

Children can argue scientifically but have difficulty with this if it contradicts with their background information or if there are many potential causal explanations. Adults also have difficulty with this.

What are multivariable causal inferences?

Causal reasoning is usually multidimensional. We often have to integrate information about different causes in everyday life. An event usually does not have just a single cause. For example, if you consider going to the post office during lunch break, you should not only consider the length of your break and the distance to the post office, but also whether there is a long queue.

Do children integrate knowledge about two dimensions?

The balance scale task is one of the best-known paradigms to investigate whether children are able to integrate information from two causal dimensions. Children must integrate weight and distance information in this task. The child must predict which side of the balance scale will go down if a certain weight is placed on the 'see-saw' at a certain location. Siegler (1978) used this paradigm and asked children to determine which side would go down. He held one variable (weight or distance from the center) constant and varied the other variable. Children aged 5, 9, 13 and 17 were tested. Siegler proposed that the understanding for this test proceeds through four rules. The first three treat the variables separately. The fourth rule involves integration of both variables.

  • Children who used rule one said that the side with the most weight always went down.

  • Children who used rule two included the distance to the center in their reasoning, but only if the weight was the same on both sides.

  • Children who used rule three took both variables into account, but only if they did not contradict each other. If one side had a greater weight and the other side a greater distance, their performance was at chance.

  • Children who used rule four were able to integrate the information about weight and distance to the center.

Wilkening and Anderson (1991) were critical, because younger children might be using a simpler integration rule that could put them in the wrong control group. The children had to balance the scale by either changing the weight or changing the distance to the center. They found that 9- and 12-year-olds and adults used the integration rules to combine weight and distance. The 6-year-olds focused more on either weight or distance. However, Wilkening and Anderson concluded that children are already aware of integration at a younger age than Siegler stated.

Do children integrate information about the causal power effects?

To measure whether children were able to combine two power forces, Pauen (1996b) used a force table. The children had to predict the path the object would follow once it was released by moving the barrier in the exact right place to catch the object. Children aged 6, 7, 8 and 9 were examined and an angle of 45°, 75° or 105° was used between the two weights. Most young children predicted that the object would move in the direction of the stronger force (80 - 85%). 45% of 9-year-olds demonstrated by integrating both forces that the opening was near the heavier weights but not exactly below it. However, they did not show this at every trial and it was suggested that they were in a transition phase. Only 5-10% across all groups showed correct integration, compared to 63% of the adults.

What are false analogies to balance scale problems in reasoning about the causal effects of forces?

Because children in Pauen's study said they saw the power table as a balance scale, Pauen thought that they might make a wrong analogy between a balance scale and the power table that caused them to make mistakes. When they were trained to use a balance scale that had its center of gravity below the axis of rotation, they were taught to apply an integration rule. If they had received this training, they would show a greater tendency to apply integration. The conclusion was that children spontaneously use analogies to reason about physical laws, just as they do with language and biology. Analogies therefore play a very important role in the reasoning of a child.

How do children integrate knowledge about three dimensions?

Children's ability to interrelate information about three different dimensions has also been examined. Wilkening's research (1981, 1982) dealt with the dimensions of time, distance and speed. Speed ​​is calculated by dividing distance by time. Distance is calculated by multiplying time by speed. Wilkening found in his research that children aged 5 and 10 used the multiplication rule 'distance equals time times speed'. The younger children did not always use the right integration rules. However, even adults did not always use the good integration rules. However, this was not a criticism according to Wilkening. Although 5-year-olds sometimes select wrong rules, they do have a good understanding of individual variables.

What are biases in causal reasoning?

Intuitive physics and erroneous assumptions about projectile motion

If one thinks about projectile motion, one starts from the impetus theory, which states that every movement is caused by something. We reason that if a ball rolls from a moving train it falls straight down (because there is nothing that has pushed it in a certain direction) while that is not true. Even when a high-speed ball is thrown through a C-shaped object, the majority of people think that it will move in a curvilinear arc, while the ball actually follows a straight line. These Newtonian rules can only be understood via direct instruction.

What gravity errors do children make?

Children make mistakes with gravity. They start from the straith down principle as described above on the basis of the ball falling off the moving train. Hood (1995) used a tube task to test the children. There are a number of tubes whose openings do not connect straight to the opening at the bottom, but to another opening. So, there is a bend in the tube, so that it ends up somewhere else than right under the opening. The tubes are opaque. If a marble is thrown into one of the tubes, 2-year-olds assume that the marble falls straight down. If the tubes are transparent, the children will look in the right container, but if the tubes were opaque once again, the children will assume that the marble will fall straight down.

So, when objects fall down 'invisibly', children assume that they fall straight down. When 5-year-olds were offered a more difficult test after passing the tube test (Hood et al., 2006), the many tasks they had to perform reduced the inhibition of the straight down principle and they often looked into the opening that was directly below the entrance to the pipe. In this experiment the children had to follow a red and a green ball and predict where the ball would come out. This meant that they had to follow the layout of the tubes, see which entrance belonged to which exit and remember into which tube the red ball and the green ball were dropped down. However, this was too much for the children.

Most of the studies that have now been discussed have measured knowledge through action. Perhaps other results will be found when knowledge is measured through reflection. It is thought that intuitive psychics is best measured through action. Research by Krist and colleagues (1993) suggests that if people have to reason about physical laws, they get better as they become older. If people have to perform through action they understand it from an early age.

What do insights from cognitive neuroscience tell us?

There are no scientific studies of causal reasoning in children yet. Studies in adults do give an idea of ​​what it could be like in children. Similar neuronal networks are activated for various reasoning tasks (networks that deal with attention, inhibition, working memory, etc.). This suggests that scientific reasoning is not a special form of reasoning, but is based on core cognitive processes.

When we learn new scientific theories, this does not interfere with existing concepts such as the impetus theory. The theories simply exist side by side. Both theories are activated in a relevant situation, but the incorrect impetus theory is suppressed. Behavioral studies also support this idea. However, this data is not consistent with the idea that scientific knowledge in the brain is radically restructured during development, because new information comes in and cannot be accommodated with existing information in the memory. Data from neurological research state that naive theories are stored alongside more comprehensive theories. In order to be able to reason correctly, inhibition is necessary. People place great value on plausible theories. This is important when someone wants to explain something based on causal reasoning

Causal reasoning is a domain-general skill that plays a very important role in cognitive development. The causal structure of events is important in the representation, interpretation and recall of these events. Causal reasoning is based on perceptual information, because the perceptual information gives cues about the intentionality of people. This promotes social cognition. In addition, it also provides information about living and non-living things. Children recognize relationships, detect statistical dependencies and connect cause and effect through perceptual information. Besides the information that children pick up from their environment, they will also form their own frameworks for explaining events, even without being able to see them. This happens once a child is capable of manipulating causes. If children develop such a framework about events, they can also make predictions and a child can control such an event.

What is the relationship between language and social cognition? - Chapter 7 (1)

What is meant by theory of mind?

Theory of mind refers to the possibility of imputing mental states to the self and to another person. Having a theory of mind is important for social cognition. Metarepresentational ability is the ability to represent the states of knowledge (the mental representations) of the self and others. This is present from the age of fifteen months. It is necessary for us to understand the mental states of other people and to predict the behavior of others based on this. Metarepresentational ability stems from imitation, the development of pretend play and language.

What is pretend play?

Early pretend play is usually accompanied by playing with peers. A child must understand that another child is also pretending. When playing with other children, there is a concept present that others also have metapresentations. Language plays a major role in helping children to understand mental states. Leslie (1987) is an important name in research into the theory of mind and metarepresentation and he emphasized the importance of pretend play in metarepresentation.

Language and pretend play both play an important role and research shows that they follow a joint timetable. Pretend play starts in the first year with a truthful way of playing (for example: drinking from an empty cup and making a sipping sound). During the second year, children start to show a more abstract way of pretend play (having a teddy bear drink from an empty cup or cleaning something that is not dirty). These actions are then combined (a teddy bear is given a drink, do some cleaning and then be put to bed). Late in the second year the child becomes able to plan pretend games and makes use of objects (a stick that serves as a horse).

The use of planned pretend play occurs at 18-26 months. The use of language is approximately the same as the pretend play pattern. The first words occur approximately at the same time as the first behaviors of pretend play. The combinations of words also occur approximately at the same time as the order of pretend play. In addition, the production of rules-based, syntactically structured, statements emerge when the child also starts with planned pretend play. It was thought that co-occurrence of language and pretend play was an internal cognitive structure that makes it possible to relate symbolic elements to one another.

Fenson and Ramsey (1981) showed that language and pretend play may need to be separated from each other in terms of development, because the development of pretend play can be encouraged if adults modeled pretend behaviors. Normally combinations of pretend play actions occur spontaneously around 24 months, but their research showed that imitation can occur as early as 19 months. With language acquisition, this age equals the occurrence of single words. Fenson and Ramsey stated that certain cognitive components had to be present in both domains before they could occur together.

Bigelow et al. (2004) demonstrated with their research that imitation provides a better understanding of representative activities. They were interested in pretend play during joint attention and investigated children aged 12 months. At this age, children intentionally use objects while playing: they attribute functions to the toys they play with. They first allowed the children play alone with a number of objects for a while and then together with their mother. The joint attention led to an advanced level of playing and if there was no joint attention, the play was more stereotyped.

What is desire-based psychology?

It is often said that social interaction is important for attaining a theory of mind. Meltzoff and Decety (2003) state that children imitate others but also notice if someone else imitates them. This provides an understanding of the existence of the self. When parents imitate their children, this ensures that children do not just perform behavior but also link goals and expectations to it.

Intentionality and desires are part of the imitation process and ensure that psychological attributions are returned to the child. Therefore, the child learns about the self and that of others.

It may be more difficult for children to understand the internal states for beliefs than for wishes. Wellman et al. were the first to suggest that early psychological understanding is based on desires and later on interactions between desires and beliefs: simple desire psychology precedes belief-desire psychology. This is because wishes motivate behavior and behavior based on beliefs is more difficult to describe. Children up to 3 years are therefore 'desire psychologists' and children from 3 years on 'belief-desire psychologists'. To understand wishes, it is only necessary to understand objects and events, but to understand beliefs, it must also be understood that thoughts are entities. Reasoning based on wishes is not enough to explain why someone would do something that does not correspond to what can be predicted from their wishes. It does not require an understanding of beliefs, but an understanding of intention.

To test whether 2-year-olds are desire psychologists only, Wellman and Woolley (1990) told these children in the desire condition a story about Johnny who was looking for his dog and thought it was either in the garage or in the house. If he wasn't in the garage, where would Johnny look? The children were able to predict this correctly. When they were told that the dog was either in the garage or in the house, the child was asked in the belief-desire condition where he thought it was. If the child thought in the garage, the experiment counselor said that Johnny thought he was in the house. Where would Johnny search? With strict criteria, only 45% of the 2-year-olds gave the correct answer in the belief-desire condition. Based on this and other experiments, Wellman and Woolley concluded that children aged 2 are desire psychologists.

Repacholi and Gopnik (1997) investigated when children develop a psychological understanding of the wishes of others. The experimenter ate a piece of broccoli in front of the child and said they really enjoyed it. Then the experimenter ate a piece of cracker and said they thought it was disgusting (or vice versa). The child was then asked to give the experimenter a piece of broccoli or a cracker (he was allowed to choose which one). 54% of the 14-month-olds gave the preferred food to the experimenter and 78% of the 18-month-olds. For this experiment it is necessary to understand that someone else can have a psychological desire that may differ from your own (almost all children prefer crackers). By the age of eighteen, children have this understanding.

What are conclusions from research into mental representations, belief and false belief?

Dennett (1978) stated that understanding false beliefs (that others may have beliefs that do not match reality) was the only evidence that children attribute a mental state to others. It used to be thought that this was only understood by 3- and 4-year-olds and that a fundamental representation change took place. However, this is now criticized by many.

Wimmer and Perner (1983) were the first to publish studies on false beliefs. They let Maxi put a piece of chocolate somewhere (place X) and he then goes out to play. His mother then moves the piece of chocolate to place Y. The children were asked where Maxi would look for the chocolate when he returned. Children who understand false beliefs would say that he is going to place X, while children who have not yet mastered this predict that he will choose place Y. They concluded that the concept of false beliefs is emerging between 4 and 6 years. Callaghan et al. (2005) investigated this in several cultures and stated that this skill is developing universally between 3 and 5 years of age.

These results are often replicated, but there is also criticisms. Siegal and Beattie (1991) argued that this study made use of misleading assignments because the child might think that they should say where Maxi would need to look to find the chocolate. In their research they used the question: "where should Sam look first?" (they used a story about Sam who had lost his pet). In this case, children aged 3 performed better.

Moreover, the young children may have given a wrong answer, because executive functioning at this age is not yet fully grown. This was investigated by having children do a window task. There were two boxes and one of the two contained chocolate. In the learning phase, both boxes were opaque and the child had to choose the box that the experimenter had to open. If it was the box that contained the chocolate, the experimenter would keep the chocolate and otherwise the child. In order to get as much chocolate as possible, the child had to point to the empty box. During the test phase the child could see in which box the chocolate was and the experimenter did not. Children aged three pointed more often to the box with the chocolate and children aged four did not. For 3-year-olds, the physical knowledge about the location of the chocolate was so important that it controlled their behavior.

Hughes (1998) argued, however, that this cannot be the explanation for not performing successful on the false belief test of Wimmer and Perner. Carlos and Moses (2001) proved that inhibitory control tasks shared variance with theory of mind testing, but that they did not measure the exact same skills.

The original interpretation of the results of the research by Wimmer and Perner received support from the appearance-reality distinction. This is the notion that appearance and reality can differ in terms of physical objects (a sponge can look like a rock) and psychological states (appears to be interested but is actually bored). Children aged 3 do not yet understand this and children aged 4 and 5 years do (Flavell et al., 1983).

Zaitchik (1990) investigated with a different method whether children are able to keep two different representations in their minds. She used non-mental representations in the form of a false-photograph task. In this task, children took photos with a polaroid camera for a while so that they understood the principle of the camera and afterwards the children watched a video. In this video the children saw that Bert took a photo of a rubber duck and he put this photo on the bed. Ernie then placed the rubber duck in the bathtub. The child was then asked where the rubber duck would be in the photo. Children of both 3 years old and 4 years old were at chance in this task and therefore worse than in a standard false belief task. Zaitchik thought that this suggests that children do not have so much trouble with mental representations but with the fact that they are representations. Children make more mistakes because of misrepresentations rather than because they have to use mental representations.

Slaugther (1998) doubted the conclusions of Zaitchik. He had three-year-old children perform a false-photograph task, a false-drawings task (uses the same principle as the false-photograph task) and a standard false belief task. His results were that the false-photograph task and the false-drawings task were significantly easier for 3-year-olds than the standard false belief task. He concluded that the concept of image representations versus mental representations were not developmentally related.

This was confirmed by another study that investigated the understanding of mental states in deaf children. These children miss conversations about mental states that contribute to the development of an understanding of mind. Peterson and Siegal (1998) studied deaf children who used sign language (but only learned it later) from 5 to 11 years old. The children participated in a false-photograph task and a standard false belief task. They scored well on the false-photograph task but not on the false belief task.

The conclusion was that these children have difficulty with false mental representations, but not with image representations, because they are not facilitated by conversations with others. However, this effect does not exist if children learn sign language from birth and the parents can also sign fluently, because then there is a communication possibility. This was found by Woolfe et al (2002), who compared children who had learned sign language from birth and have parents who could sign fluently, with deaf children who had learned sign language later in life. This shows that conversation is very important for a normative development of an understanding of mental states.

What is the role of language in the development of metarepresentation?

In the second year of life, children use terms for mental states during their conversations. Some are emotional states, but most are not (tired). These emotional concepts emerge between 20 and 28 months. Bretherton and Beeghly (1982) had mothers keep track of the language expressions of their children between 20 and 28 months, divided into different categories such as perception, physiology and positive and negative effect. They found that 90% of 28-month-old children talk about pain and that 'knowing' was the most commonly labeled cognition. The scores for cognition labels lagged behind the scores for affect and morality and this is, according to the researchers, because they contain more explicit behavioral correlates. Their conclusion was that the exchange of information by language played a central role in the development of social cognition.

Dunn et al. (1991) investigated the role of conversations with family members and the development of the understanding of the emotional states of others. Children were observed at 36 months and later tested at 6 years of age. At 36 months, no differences were found in the number of conversations about feelings between boys and girls, and most conversations about feelings were most common when children were disputed. Children who had talked more about feelings and mental states of themselves and others at 36 months, were better at identifying emotions of others at the age of six. This did not depend on how well they could talk at the age of six or how much communication took place between the mother and child at 36 months.

Dunn et al. (1991b) examined whether conversation with parents also influenced the beliefs of others in children. They found that when children talked more about feelings with their parents at the age of 33 months, they had a better emotional understanding and false belief at 40 months.

Lohmann and Tomasello (2003) investigated whether the conversation with the parents can truly be seen as the cause of a better understanding of the mental states and beliefs of others. They did this by using a training to see if that specific training actually improved the understanding of children. They concluded that conversations with parents are indeed a cause of a better understanding of mental states.

What is mind-mindedness?

Meins (1997) used the term mind-mindedness to express that parents and caregivers regard young children as individuals with a mind. Not all parents do this, which explains the difference in the allocation of a mental state to other children. She linked mind-mindedness to the degree of security of attachment. A key idea in the theory of attachment is that the quality of the attachment with the parents develops an 'internal working model' of the self. Children who experience secure attachments have a positive internal working model of themselves. If a child is treated from an early age as an individual with a mind of his own, this may have a positive effect on the child's understanding of the mental states of others. Meins and Fernyhough (1999) investigated whether the mother's mind-mindedness at 20 months predicted both the mother's mind-mindedness at the age of 3 years and the understanding of the theory of mind of children aged 5 years.

The degree of mind-mindedness of mothers was measured when their children were 20 months old and at the age of 3 the mother was asked to describe her children. If she did this on the basis of physical characteristics, this was considered less mind-mindedness than if she did this on the basis of psychological characteristics. The results were as follows: mothers of securely attached infants were more mind-minded. Also, children of mothers who were more mind-minded performed better on the false belief task when they were 5 years old. It is therefore important that children are treated as individuals with a mind.

What is the importance of siblings in pretend play?

Perner et al. (1994) found that children aged 3 and 4 with two siblings are almost twice as likely to pass false belief tasks than children without siblings. This is because their brothers/sisters also treat them as individuals with a mental state and because they get more involved in pretend play. However, this study did not measure the children's cognitive skills, which could be an alternative explanation for the results. Jenkins and Astington (1996) did take this into account in their study. They found that children with better language skills were better at false belief tasks and that children from larger families had better language skills. When age, language proficiency and birth order were controlled, the size of the family explained a significant part of the variance in individual differences in performance on the false-belief task. It does not matter whether the brothers or sisters are younger or older. Children with language deficits also benefit most from brothers or sisters.

Brothers and sisters are usually part of pretend play and therefore this pretend play differs from pretend play that is performed with parents. It is more social and emotional. In their research, Youngblade and Dunn (1995) found that pretend play with an older brother or sister (on average three years older) more often results in conversations about feelings than in children playing with their mother. The conclusion of Youngblade and Dunn was that certain aspects of children's interactions with their siblings are strongly related to their understanding that other people have mental states.

What is the importance of friendship with regard to theory of mind?

At the age of 33 months, children talk predominantly about mental states with their families, but at 47 months more often with their friends. Hughes and Dunn (1998) studied ten boy-boy dyads, ten girl-girl dyads and five boy-girl dyads, aged 3 years and 11 months on average. They recorded them three times for 20 minutes in one year. To allow them to participate in pretend play, they gave the children dress-up clothes and role-playing toys. The amount of talk about mental states was measured and the children were tested for the understanding of theory of mind and emotionality. The extent to which children talked about mental states was significantly correlated with their performance on the tasks. If children talked more about mental states in session one, they also scored higher on tasks in session three. Furthermore, girl-girl dyads talked more about mental states than boy-boy dyads. So, interaction with friends is also important for understanding the mental states of others.

Friendships can vary in quality: sometimes close with a lot of positive affect, sometimes with a lot of arguing and then making up. Dunn and Cutting (1999) examined 128 friendships of 4-year-olds. Children who showed the most pretend play scored better on the theory of mind task, talked more with their friends, showed fewer errors in communication and had fewer arguments (and vice versa: higher score on theory of mind meant more pretend play). Conversation about things other than mental states does not contribute to the development of a theory of mind. The study showed that parents' education and the background of the family also contribute to the theory of mind development.

How are bullying and theory of mind related?

Sutton et al. (1999) investigated the extent to which children who bully have a theory of mind. She divided the group into six categories: (1) the bully (13%) (2) the victim (18%), (3) the assistant who helps the bully (6%), (4) the reinforcer who encourages the bully via watching and laughing (8%), (5) the defender (44%) and (6) the outsider who wants nothing to do with it (11%). The bullies scored higher on total social cognition than any other group. Having a better theory of mind skill than the victims and followers ensures that bullies are in favor. However, this is a correlational study and therefore it cannot be said with certainty what exactly causes what.

Children aged 4 with anti-social behaviors have a poorer emotional understanding, even when language skills, family background and cognitive skills are controlled, according to Hughes et al. (1998). Children with anti-social behaviors were also more often able to snatch toys, to call their friends names and to break rules. This is perhaps a reflection of the behavior that also takes place within their families. They were also significantly more inclined to use violence. When they were tested in a follow-up at the age of six, these children showed deficits in their moral development and social understanding. Children who participated in more violent pretend play at the age of four had a poor moral understanding at the age of six.

What is the importance of language skills?

Developing language skills is beneficial for developing a theory of mind, because pretend play focuses more on mental states. Russel (2005) provides an additional reason why language skills are important for representational understanding. He states that even the most basic aspects of language acquisition entails understanding of beliefs. This is because someone who labels (what you do when you use language) searches for the correct word for an object and because someone who labels realizes that everyone is supposed to agree to the label being correct before it can be used (if everyone has a different name for a fork, 'fork' is not a generally accepted label).

Children aged sixteen months correct someone who uses the wrong label and look longer at a person who uses the wrong label than at a person who uses the right label. If a correct label is used, they will look longer at the object being labeled and if an incorrect label is used, they will look longer at the person who says it wrong and / or at their own parents. According to Koenig and Echols (2003), labeling gives a presentation of someone's conviction. If someone says, "That's a cat," it's someone's "belief report." He or she is then convinced that it is a cat. Children expect labels to reflect intentional states of a person and therefore expect veridical labeling. Russell states that improvement behavior and looking longer at someone who labels incorrectly is a sign that a child knows that someone has a false belief.

Pretend play also contributes to metarepresentational development through the necessity to understand the intent of a playmate who is pretending and through actively determining what is real and what is pretend play. Harris et al. (1991) investigated the occasions on which children confuse reality and pretend play. Play involving monsters can be very disturbing for a child. There were two boxes in the room and the experiment supervisors and the children pretended that there was a puppy in one box that would lick your finger when you put your finger in the box and in the other a monster that would bite off your finger. When the children were asked in which box they wanted to put their finger, the majority of the children chose the box with the puppy. In such situations, children can no longer properly disassemble the difference between reality and pretend play. However, even adults show this behavior for certain pretend entities.

What is meant by theory of mind?

Theory of mind refers to the possibility of imputing mental states to the self and to another person. Having a theory of mind is important for social cognition. Metarepresentational ability is the ability to represent the states of knowledge (the mental representations) of the self and others. This is present from the age of fifteen months. It is necessary for us to understand the mental states of other people and to predict the behavior of others based on this. Metarepresentational ability stems from imitation, the development of pretend play and language.

How does memory develop? - Chapter 8 (1)

The episodic memory is the memory for past events. The semantic memory is the memory for factual knowledge. Together, episodic and semantic memory form the explicit or declarative memory. This memory can be used consciously and deliberately. As a contrast, the implicit or procedural memory requires no consciousness, such as when driving.

Supporters of cognitive psychology state that memory consists of a modular system, where all memory systems are thought to function separately. Results from research from the field of cognitive neuroscience support to this idea. For example, episodic memory is centered in the hippocampus and the medial temporal lobe, but learning skills in the motor cortex.

Research shows that people create memories on the basis of personal interpretation, on the basis of knowledge they already possess and on how much they understand what is happening at that moment. Children learn to store memories in more detail at the time they are able to express things verbally. Memory development also depends on other cognitive processes and on a child's metaknowledge. For example, a child can expand his memory by the use of mnemonic strategies.

What happens during the early development of memory?

Recall is the active retrieval of information from previous events. Bauer et al. (1987) concluded from their research that children aged 17-23 months remember events in a certain order. In their experiment they asked children to give a teddy bear a bath in a certain order, and even after six weeks the order was repeated correctly, without prior modeling. Since taking a bath is a familiar ritual in childhood, Bauer et al. investigated whether they got the same results if they let the child perform an unfamiliar task, such as building a rattle. They found the same results and stated that these young children were sensitive to causal relationships and that this means that young children have representations of purposeful ordering, just like adults.

Bauer and Mandler (1989a) investigated whether children are better at storing tasks associated with causal relationships than without causal relationships. In their research they tested children from 16 to 20 months of age. They had them perform two new tasks (building a rattle and make the frog jump) and two well-known tasks (giving a bath to a teddy bear and cleaning the table) where there was a causal relationship between the sequence of the actions. This means that if the actions are performed in a random order, the end result will not be achieved. They also let the child do a task (the train ride) in which the order of the actions did not matter to achieve the same end result.

There was a significant recall for both the random condition and the new fixed order condition. However, this recall was much lower for the random condition than for the fixed order condition. If an irrelevant action was added to the fixed order task (for example, attaching a sticker), it was often displaced to another position in the sequence. In the random order, it was treated no differently than all other actions within the task. Bauer and Mandler concluded that causal relationships are important for constructing and retrieving memories.

A number of studies into the memory of even younger children revealed the following:

  • 80% of thirteen-month-old children could retain memories with a sequence (such as
    building a rattle) for one month.

  • 80% of twenty-month-old children could retain memories for at least six months.

  • 70% of twenty-month-old children could retain memories for at least one year (Bauer et al., 2000).

  • 50% of nine-month-old children could retain memories for a month, but not for three months.

  • At ten months they could retain memories it for three months (Carver & Bauer, 1999, 2001).

Autobiographical memories of events before the third year do not usually occur. This is called infantile amnesia. However, other cognitive skills are learned during this period. Bauer argues that infantile amnesia does not exist. She states that infants create autobiographical memories, but that they cannot retrieve them. Freuds (1938) explanation for infantile amnesia was that memories were present, but that they were suppressed so that they do not end up in the conscious. However, if scary / bad memories are suppressed to protect the individual, why aren't nice memories remembered?

Another explanation is that early memories are coded in terms of physical action or pure sensation. You cannot retrieve them because they are stored in a different format to later memories that are based on linguistic storage. Goswami supports this view, because girls, who are more linguistically developed than boys, also have earlier memories.
Simcock and Hayne (2002) investigated this and found that children also had non-verbal access to memories, but that the inability to translate these memories into language experiences meant that the memories were not stored in the autobiographical memory. Language therefore plays a key role in infantile amnesia.

Another explanation is that children do not yet have a sense of the 'self' and therefore no memories do exist from the first 2 years of life (which is when this realization arises). The emergence of the self would result in a new organization of information: things happen "to me". However, according to Goswami, this is also questionable because research shows that the concept of the self emerges before the age of 2.

Another explanation for the absence of autobiographical memories is that the systems in the brain that play a role in storage are not yet sufficiently developed. According to Gowami, however, this is not probable either, since the medial temporal lobe develops in the first year already and the frontal lobes continue to develop into the early twenties, but these structures both play a role in infantile amnesia.

Finally, the development of knowledge structures could be important. Children do not yet have separate memory cues and a framework for storing events still needs to be learned. This is likely according to Goswami, because this theory states that infantile amnesia is caused by the absence of abstract knowledge structures for describing temporal and causal sequences of events and not because the basic structures in the memory system change during development.

When storing memories, we do not just use language, but also other features such as pictures, hand gestures and other symbols. Children do this too. An explanation for a better memory for older children compared to that of younger children is therefore that these children are further developed in understanding symbols. DeLoache (1987, 1989, 1991) investigated this. He showed children a scale model of a room with various pieces of furniture and a real room, with the same furniture. The first room is from 'Little Snoopy' and the second one from 'Big Snoopy'. Children get to see where Little Snoopy is hiding in his small-scale model room and they are told that Big Snoopy is hiding in the same place in his room. Children aged 3 immediately know where to look in the real room, but children aged 2.5 randomly search for Big Snoopy. They do not understand that the scale model room and the real room are the same. This conclusion was replicated in another experiment. In this experiment the child was told that there was a "shrinking machine" and if the child thought that the scale model was the same as the real model because it had shrunk, he immediately started looking at the right place. Even when children see the hiding place of Little Snoopy on a picture, they are able to find Big Snoopy.

What is the development of recognition memory?

Recognition memory is the ability to recognize that something is familiar and has been experienced before. It is part of the implicit memory. The recognition memory is found in other animals as well and is present early in life. Because of this, there is a discussion as to whether it is an actual cognitive ability, or whether it is just a measure of processing.

Recognition memory is measured by showing children a series of pictures and then measuring how many they consider as familiar after a certain period of time. This is the traditional measurement method. Brown and Scott used this setup in their experiment with 3- to 5-year-olds. They found that 98% of the children showed an accurate recognition memory. In addition, there was a longer time between the learning and testing phase (1, 2, 7 and 28 days). The children saw twelve images that they had been seen once, 24 that they had been seen twice and 36 new pictures. Up to seven days, 94% of the children still knew if they had been seen the pictures twice. For images that had been seen only once, this varied from 84% (after 1 day) to 70% (after 7 days). After 28 days it was 78% for the pictures that had been shown twice and 56% for the pictures that had been shown once. The superior memory for pictures that had been shown twice is caused by the fact that the picture is shown once again and that an assessment has to be made (has the picture been before, or not?). Children therefore have a very good recognition memory at a young age.

The implicit memory is also called unintentional memory or perceptual learning. In their research, Caroll et al. (1985) used a condition in which children were shown a picture and were asked to count the number of crosses. The child thus scans the picture but does not perceive it consciously. Later the child is asked if he has seen the picture before. In their experiment they also did research on explicit memory and concluded that perceptual learning does not develop with age.

The fragment completion task is another way to measure the implicit memory based on a word test or a picture test. Fragments of a picture or word are shown and the participant is asked to complete it or retrieve it (recall). Natio (1990) used such a test. Two thirds of the words used in this test had already been mentioned in two previous tests. In the first test, in which half of the target words were used, the participants had to classify the words into a category ("is this a garment or a piece of fruit?"). This was to induce deep processing.

For example, in the second preceding test, they had to say whether a certain letter was in the word. This was to induce shallow processing. Familiar items were answered correctly more often than unknown items in the following fragment completion task. Furthermore, the implicit memory for these words did not depend on the depth of processing. In a follow-up experiment, Natio asked the children to retrieve the words, explicitly. Age did influence performance and this suggests that the implicit memory does not develop, but the explicit memory does and that the two types of memory are therefore two different systems.

Russo et al. (1995) carried out the same experiment, by using pictures and came to the same conclusion. The same results were also found by Perez et al. (1998) and Bullock Drummey and Newcombe (1995). However, Cycowicz et al. (2000) claim that there are developments in implicit memory with age. They found that children aged 5, 9 and 14 and students all had an implicit memory for previously seen objects in a fragment completion task, but that it took the children longer to identify the object. With a fragment completion task a few lines of the picture are shown at the beginning and then more and more are added until the participant is able to identify the object. Younger children needed more lines. Cycowicz et al. therefore claimed that there is indeed a development in implicit memory, but that explicit and implicit memory still develop at different rates.

Ellis et al. (1993) investigated whether children are also sensitive to priming effects when processing faces. They indeed found a shorter response time when assessing whether a face was familiar or not, if the face had already been shown in another task. This was the same for 5- and 8-year-olds and slightly faster for 11-year-olds.

Newcombe and Fox (1994) tested the implicit and explicit memory for children's faces by using a galvanic skin response. They showed 10-year-olds photos of 3- and 4-year-old children who had been their classmates. Photos were also shown of 3- and 4-year-old children who had attended the same school, but five years later. Later the photos were shown again and asked if the faces were familiar and how nice the participants thought the children were. The latter was asked to see if the participants showed a preference for their classmates, even though they could not remember them. They found that the children recognized their classmates both implicitly and explicitly. If they divided the group into high explicit recognition and low explicit recognition, the performance was equivalent in both groups. This suggests once again that explicit memory develops separately from the implicit memory.

How does the episodic memory develop?

In adults, episodic memory is made up of schemas or scripts. For example, they have a script for doing the laundry or eating in a restaurant. To study the development of episodic memory, we therefore investigate the development of scripts and schemas.

Nelson et al. (1986) asked children to tell exactly what happened during events like grocery shopping and they found that children's episodic memory is structured around the same important representations as in adults. This corresponds to findings of neural data. Over time, the important representations are activated more strongly than the details, because of a higher focus. The development of schemas plays a major role in the development of memory and it also ensures that events become predictable.

Fivush and Hammond (1990) demonstrated that children can also remember unusual events until at least 18 months later. They agree with Nelson that children focus on routines, but claim that if the child understands the routine, this also helps him to understand novel events, even though these are inconsistent with what usually happens. Events with a high emotional significance are also remembered very well.

However, the results from an experiment by Farrar and Goodman (1990) do not seem to support Fivush and Hammond's claim that children use scripts to develop memories for new events. They had 4- and 7-year-old children come to the laboratory five times within two weeks to play games with animals, always in a fixed order. During one visit, however, a new game was played: this was the novel event. A week later, the children were asked about their experiences. The 4-year-old children made one script for both the routine events and the novel event, and saw this as a routine. The 7-year-old children did form two separate scripts; a script for routine and a script for exceptions. Farrar and Goodman concluded that the ability to create different memories for unusual events is not yet fully developed in children aged 4.

Goswami states that both Fivush and Hammond and Farrar and Goodman could be right, because the difference is in the degree of importance of the novel event to the child. The events in the experiment of Fivush and Hammond were very important and in the experiment of Farrar and Goodman the evens were less important.

Reese et al. (1993) investigated whether the communication of parents with their child plays a role in the development of memory. They had mothers of 40, 46, 58 and 70-month-old children ask questions about past events. These should not be routine events, as this could activate the routine script. There were two maternal narrative styles: the first style was elaborating on the information that their child recalled and then evaluate it. In the second style mothers often switched topics and did not evaluate the information. Mothers using the first style always asked different questions ('where were we?', 'What did it look like?'), while mothers using the second style kept asking the same question ('which animals did you see? Andd what else? And what else?'). Children with mothers who used the first style remember more at 58 and 70 months. This is called maternal elaborativeness and it plays a key role in the development of the ability to store memories. It provides more detailed and organized memories and a better sense of time. Moreover, if the child and the mother disagree, the child comes to understand that everyone experiences an event from their own perspective.

Fivush and Schwarzmüller (1998) discovered that important events are stored in great detail and are also described in more detail when asked after a while than when asked about immediately. They also found that memories are stored better and in more detail if they are discussed verbally, just after it has taken place. This is because language enables a child to form good constructions of what has happened. If adults guide the child verbally, it supports the organization of the child. The events that are discussed create the first memories in the autobiographical memory. Abbema and Bauer (2005) found similar results. They found less recall of the events they investigated, but the events were not as important to the child as in the experiment by Fivush and Schwarzmüller. Abbema and Bauer concluded from their research that a memory can be retained after a while, as long as the initial representation is strong enough.

They also found a growing tendency to describe events from their own perspective as the children grew older. Nelson and Fivush (2004) state that autobiographical memory plays a major role in social and cultural functions.

How accurate are memories of eyewitnesses?

Eye-witness memory is the memory for events that may not seem to be significant at the time that they were experienced. It is a special kind of episodic memory. Ochsner and Zaragoza (1988) tested the eye-witness memory of children aged 6. The children had to make a puzzle in a room and then a man came in and messes around and said he was looking for the headmaster. He then stole a handbag and left the room. The children were more accurate and less sensitive to suggestions than children in the control group (same situation, only the man didn't steal a bag). This indicates that the eyewitness testimony of a young child is no less accurate than that of an adult.

Other studies, however, found that children are much more sensitive to suggestive questions than adults. Cassel et al. (1996) found that unbiased leading questions (such as: did the bicycle belong to (a) the boy, (b) the mother, or (c) the girl) produced as many false memories as biased questions (such as: the bike belonged to the mother, right?). This is interesting because it suggests that the mechanisms that create false memories are likely to be general and depend on how the developing memory system functions, and are not specific ones related to false memories of negative events.

Rudy and Goodman (1991) conducted an experiment in which two children played the game 'Simon says' in a trailer with a strange man. Hereby games were carried out which involved actions such as tickling and taking a photo, because these acts also often occur during sexual abuse. Later the children were asked what had happened and were subjected to leading questions. 83% of the 4-year-olds answered the questions about sexual abuse correct (so that no abuse had taken place) and 93% of the 7-year-olds.

A follow-up experiment revealed that children aged 3 were more sensitive to leading questions. In particular, "did he kiss you?" was often answered with a yes nod (3-year-olds: 20% and 5-year-olds: 2 out of 120). The more worrying misleading questions (he undressed you, right?) were answered correctly by this group.

Eisen et al. (2002) studied children who were being treated because they were severely abused. They found that age, intelligence, and overall psychopathology were a better predictor of eye-witness memories than the extent to which a child was abused. The higher the age and intelligence, the fewer false memories. In addition, other aspects also play a role, such as the emotional tone of the interviewer and the extent to which the child wants to please the interviewer.

In general, the developmental patterns of eye-witness memories and episodic memories are the same. One hypothesis about the link between children's knowledge of routines, information from scripts and their eye-witness recall is that children who have more episodic knowledge about certain events should be less sensitive to leading questions. The correlations between knowledge and memory were significant during each delay interval. It was concluded that the variability of knowledge about a certain domain is accompanied by a corresponding variability in recalling a memory. For example, if you know a lot about a doctor's visit and how this works, you also remember more about your own doctor's visit or vice versa. Eisen et al. (2002) also found a gender difference: girls show more detailed memories than boys.

How does the working memory develop?

The memory system for short-term recall is the working memory. This system has a limited capacity of 'workspace' that only retains information for a short time. The information can be incorporated into cognitive tasks. Both new information and information from the long-term memory can be retrieved. The working memory consists of the central executive, the visuospatial sketchpad and the phonological loop. The central executive is the system that controls attention. The visuospatial sketchpad processes and holds visual, spatial and kinesthetic information and is also responsible for verbal information that is stored as an image. The phonological loop ensures the processing and retention of verbal and acoustic information. The phonological loop can only hold information for a short period of time, so a telephone number, for example, must always be repeated.

The visuospatial sketchpad and the phonological loop are also seen as the two "slave systems" of the central executive. It is claimed that children primarily use the visuospatial sketchpad, but switch to the phonological loop around the age of 5, which is similar to the age at which Piaget proposed there was a fundamental change in children's logical reasoning.

The reason we think that children under the age of 5 only rely on the visuospatial sketchpad is that children are not susceptible to effects related to the coding of speech sounds in the working memory. This is necessary for processing via the phonological loop and because it is not present, it is assumed that children use the visuospatial sketchpad.

Conrad (1971) investigated this and used some kind of memory game. Children aged 3 to 11 years were shown a series of pictures. Then a complete set of pictures was shown and the child had to indicate which picture had just been shown. Conrad used two sets of pictures: one with pictures of words that are phonologically similar (rat, cat, etc.) and one with pictures that are not similar. Adults have difficulty with this task when using the series of pictures with similar words, because they store the words verbally and the words sound alike. Conrad found that only the 3- to 5-year-olds showed no difference in the processing of both sets. This shows that they are not yet processing the images via the phonological loop. Children aged 6 years and above showed a longer memory span for the similar series than for the "sounds similar" series.

In order to investigate whether younger children actually made use of the visuospatial sketchpad, Hitch et al. (1988) tested whether these children had difficulty with remembering pictures that are visually similar (pen, fork, key, etc.; all the same shape). He showed pictures and asked them to repeat them in the same order. The 5-year-olds had the most difficulty with the set of visually similar pictures and the 10-year-olds had the most difficulty with the pictures that depicted a long / difficult word (kangaroo, umbrella). So 5-year-old children use the visuospatial sketchpad and 10-year-old children use the phonological loop.

However, it may also be possible that young children do not necessarily trust their visuospatial sketchpad, but that they are not yet able to use a certain mnemonic strategy. These are strategies such as organization and repeating to promote storage and recalling memories. It is common that young children do not yet have these strategies. Deaf children appear to use their visuospatial sketchpad, even when hearing children move to the phonological loop.

The phonological confusability effect or phonological similarity effect means that words that sound similar are much more difficult to remember than words that do not sound similar. In addition, the word length effect is the effect that long words are more difficult to remember than short words. This can provide a method for determining IQ by using a measurement of the memory span in people who speak a different language. The memory span is larger for easier words than for more difficult words and numbers are more difficult to construct in one language than in another. So, if someone has to remember a series of numbers, people from the 'simple language' have an advantage and it may seem as if they have a higher IQ, but that does not necessarily have to be the case. In addition, speech rate is another important component: children who articulate faster also have a larger memory span than children who articulate more slowly. After all, children who articulate more slowly need more time to repeat and practice the words. Research by Hitch et al. (1989) shows that speech rate does indeed influence the number of items that can be held in the working memory, but only if the items are presented in speech form.

Henry and Millar (1993) argue that this is because words are named. Children who go to school learn to rely increasingly on language because they have to name things and this explains why the memory span is growing rapidly around the age of 7 years. Familiarity with the (phonological representation of the) words also influences the development of memory span. More familiarity means easier storage, processing and retrieval. Henry and Millar therefore suggest that the development of working memory is strongly related to the development of long-term or semantic memory. Speech rate is related to the phonological representation of words in the semantic memory.

When talking about the development of memory, it is unlikely that the long-term memory and the short-term memory operate completely independent from one another. The capacity of the short-term memory also depends on the quality of the long-term memory.

How do remembrance strategies develop?

Children are not convinced that they need mnemonic strategies, but in fact they do need them. In a study by Wellman et al. (1975), children aged 3 used strategies to remember the cup under which the toy dog was hidden. For example, they touched him more often or looked at the cup and said 'yes', while they said 'no' to the other cups. DeLoache et al. (1985) and Sommerville et al. (1983) found in their research that 2-year-old children also use mnemonic strategies. However, the strategies that were used in these studies were fairly task-specific.

The idea is that the mnemonic strategy of rehearsal is developed once children go to school and rely increasingly on language. Younger children do not do this yet. They are capable to use the strategy, but they simply do not want to use the strategy. This is called production deficiency. Flavell et al. (1966) found that 10% of 5-year-olds, 60% of 7-year-olds and 85% of 10-year-olds used the rehearsal strategy.

Naus et al. (1977) instructed some 8- and 10-year-olds to rehearse the words as they normally would and some to rehearse the words with a specific method (repeat the word that was mentioned, together with two other words). The children with the learned strategy performed better on the memory test, suggesting that the quality of the rehearsal is a greater predictor of success than the amount of rehearsal.

Rehearsal first appears when it is encouraged by training or by task-specific factors. It is used at a later age as a mnemonic strategy.

The fact that young children do not own the rehearsal strategy is probably more a result of production deficiency rather than a lack of competence. Children are more likely to use rehearsal as metacognitive skills develop.

Using semantic categorization as a mnemonic strategy has approximately the same developmental pattern as the development of the use of repetition.

This was investigated in a study by Schneider (1986). He found that 10-year-old children had categorized words previously and that 60% did so spontaneously. In 7-year-olds this was just 10%. Schneider suggested that this is because organizing in categories in children aged 7 depends on the degree to which items were associated. The high associativity ensures that clusters are formed. So, it does not happen because the child consciously chooses to make these clusters. Bjorklund and Bjorklund (1985) also found that children unconsciously use strategies such as the seating arrangement in their classroom to recall the names of their classmates, but when asked how they remembered this, they are not aware of this. If they are given an instruction, they can consciously apply the strategies. High associativity automatically leads to the structure of recall.

Scheinder and Sodian (1988) found a conscious use of a categorization strategy among 4-year-olds. They used a simpler task that was easier to understand from the point of view of a 4-year-old. They claimed that children can make use of organizational strategies in the past.

Children use different mnemonic strategies and the more they use them, the better the recall of information. DeMarie and Ferron (2003) found that younger children (from the age of 5) already use different strategies and the number of strategies used increased with age. In addition, the use of strategies was significantly correlated with successful recall. Memory capacity was no predictor of successful recall. From this it was concluded that the use of strategies by children is an important factor in the development of memory.

Schneider et al. (2004) also investigated the use of different strategies in 6-year-olds. They showed children twenty items from five semantic categories and gave the children three minutes to do whatever they wanted to remember the items (categorize, repeat, etc.). The memory capacity was also measured. The use of strategies and memory capacity increased with age and there was a better recall among children who used more strategies. Schneider et al. were interested in individual differences and found that memory development is characterized by a rapid transition from nonstrategic to strategic behavior for most children. There is no gradual increase in the use of strategies. Children who use strategies early on also use them more when they are older.

The level of prior knowledge has an impact on coding and storing incoming information. Experts organize their knowledge differently from novices because of the knowledge they already have. This is called the novice-expert distinction. Experts are usually older than novices, but it is interesting to study if experts are younger than the novices.

Chi (1978) tested the memory for a position of chess pieces on a chessboard with experts in chess aged 6-10 years and adults who could play chess, but were novices. They were allowed to look at the set-up for ten seconds and then had to place the pieces in the same way.
The experts had placed 9.3 chess pieces at the correct location and the adults 5.9. To locate all the pieces properly, the experts needed 5.6 trials and the novices 8.4. Chi thought this was because the experts could see more different meaningful patterns in the set-up compared to the beginners, because of their prior knowledge. In a follow-up study by Schneider et al. (1993b) it was found that for the experts, it did not matter whether there was a meaningful position of chess pieces in order to remember more and that the effect completely disappeared if the set-up was not placed on a chessboard. This leads to the conclusion that expertise involves important quantitative differences, such as the geometric pattern of the chessboard. Experts also organize their information better than novices (Chi et al., 1989).

In some areas (such as chess and sports), expertise is a more important predictor of memory performance than general cognitive skills and intelligence. An unanswered question remains whether changes in the memory of experts are both qualitative and quantitative or quantitative alone.

What are the insights from cognitive neuroscience?

Cognitive neuroscience has focused on perceptual learning when studying implicit memory. The debate is centered on whether perceptual learning is domain specific or not and how this expertise is acquired.
Perceptual learning of faces is specialized in the fusiform gyrus. The general acquisition of expertise is also part of this area. Both cause an N170 wave in this area. Studies with autistic children show that their field of expertise induces activity in the fusiform gyrus but no activity to human faces, whereas a typically developed child shows the exact opposite.

With explicit memory, the frontal lobes (especially the prefrontal cortex) and the medial temporal lobes (such as the hippocampus) play a crucial role. Adults with damage to the medial temporal system have intact short-term memory, but cannot create new long-term memory memories. They do have access to memories from before the neural damage, which suggests that the medial temporal lobes are responsible for the consolidation of memories and not for long-term storage. The storage depends more on association cortices, and the prefrontal structures are important in the retrieval of memories.

The hippocampus plays a role in consolidating memories and recollecting them, but there is still debate about how this happens. It is stated that the hippocampus is important in remembering both episodic knowledge and semantic knowledge, but research by Vargha-Khadem et al. (1997) on three people who suffered from damage to the hippocampus in their childhood contradicts this. The research showed that the children had damage to their episodic memory, but that their semantic knowledge was intact.

Bauer did research on explicit memory and concluded that earlier encoding and consolidation of information retrieval are the significant sources of developmental changes in explicit memory at an early age.

The areas that play a role in working memory are the frontal and parietal areas for both children and adults. The visuospatial working memory ensures bilateral activity in the superior frontal sulcus and the intraparietal cortex.

What happens during the early development of memory?

Recall is the active retrieval of information from previous events. Bauer et al. (1987) concluded from their research that children aged 17-23 months remember events in a certain order. In their experiment they asked children to give a teddy bear a bath in a certain order, and even after six weeks the order was repeated correctly, without prior modeling. Since taking a bath is a familiar ritual in childhood, Bauer et al. investigated whether they got the same results if they let the child perform an unfamiliar task, such as building a rattle. They found the same results and stated that these young children were sensitive to causal relationships and that this means that young children have representations of purposeful ordering, just like adults.

What are cognitive skills? - Chapter 9 (1)

In this chapter metacognition, reasoning and executive functioning are discussed. Metacognition is knowledge about cognition. Flavell (1979) defined this concept as any cognitive activity or knowledge that takes an aspect of cognitive activity as its cognitive object. An example of this is the conscious use of cognitive strategies to increase your performance. Meta-memory (knowledge about memory) and executive functioning are separately studied aspects of metacognition. Executive functioning is the ability to regulate behavior and thoughts, planning behavior and inhibit inappropriate behavior. To develop metacognition, children must be able to treat cognition itself as an object of cognition. Metacognition and meta-representation are closely linked. However, metacognition is focused on knowledge about someone's own mind and not about that of others, such as meta-representation.

What is metamemory?

Metamemory is knowledge about your memory. When children learn more about this, they should become aware of what they can remember well and what is more difficult to remember and they learn to use mnemonic strategies that will benefit their memory. Improving metamemory will therefore result in fewer deficiencies in memory tasks.

Intentional memory is required, which is about procedures (procedural metamemory), requiring recognition that memory activity is necessary ('metastrategic knowing') and knowing the kinds of activity that may enhance performance (knowing 'how'). Wellman et al. studied metamemory variables and distinguished three different areas: knowledge about tasks (must a lot be remembered or not?), knowledge about persons (the self-concept of the child's mnemonic strategies) and knowledge about strategies (knowing that rehearsal improves memory performance).

The different metamemory variables are tasks, persons and strategies. Wellman (1978) investigated the concept of 5- and 10-year-olds for the different variables. He showed them pictures. For some pictures, a single variable was used, such as a picture of a boy who had to remember eight items and a picture of a boy who had to remember 18 items. Both the 5- and 10-year-olds understood that the boy who had to remember 18 items had a more difficult task. If there were two variables, for example, that one boy had to remember 18 items over a long time and another boy had to remember 18 items over a short time, the 5-year-old children performed worse. They could only judge performance based on one relevant variable. Wellman concluded that an important aspect of metamemory development is the ability to consider the relationship between various metamemory variables.

Justice (1985) investigated children becoming aware of the usefulness of different strategies in certain tests. They showed children a video of Lee, who had to remember a set of 12 items. The strategies he could use were rehearsal, categorization, naming and looking. Children aged 7, 9 and 11 had to judge what was the best strategy. The children aged 9 years and older rated categorization as the most effective strategy, which was the best strategy indeed.

Justice et al. (1997) also examined whether children understood what they were doing when they applied a strategy. If children had mentalistic explanations for using a strategy ("it helped get them in my mind"), they also performed better on a memory test, regardless of age.

What is self-monitoring and self-regulation?

Self-monitoring is another aspect of metamemory. It is the ability to keep track of where you are with respect to your memory goals. Self-regulation is another related concept. This is the ability to plan, manage and evaluate your own memory behavior. These are both forms of executive functioning. A sufficient degree of self-monitoring is necessary for successful self-regulation. Self-monitoring is measured in terms of ease-of-learning judgments, feeling-of-knowing and judgments-of-learning (assessment of learning over a short-term and long-term period). The latter is to estimate what has been learned immediately after a list of items has been studied and again after a few minutes.

Dufresne and Kobasigawa (1989) investigated whether children were able to adjust their study time to the difficulty of items they had to learn. They allowed children aged 6, 8, 10 and 12 to learn easy word pairs (dog-cat) and difficult word pairs (frog book). They recorded how much time the children spent learning the more difficult word pairs and learning the easier word pairs. Children aged 10 and 12 spent significantly more time on the more difficult word pairs. Also, the 8-year-olds spent slightly more time on the more difficult word pairs. The 6-year-olds did understand the difference between the more difficult pairs and the easier pairs, but did not yet have any metamemorial knowledge of how to study the more difficult pairs. They concluded that younger children are able to use self-monitoring, but that they do not necessarily use it to improve their memory performance.

What does ease-of-learning judgments mean?

The results from the research by Dufresne and Kabasigawa can be derived from the fact that children are not yet able to make good predictions about their memory performance. The ability to do this is called the ease-of-learning judgments. This is tested by having participants estimate how much they think they will remember and then test what they actually remember. Young children are not good at making these assessments. This is not necessarily due to problems in self-monitoring. Visé and Schneider (2000) investigated this and concluded that it was because young children think that if they put a lot of effort into something, they can do it. Their wishes are the same as their estimates. It is therefore not so much a matter of a lack of self-monitoring. Visé and Schneider concluded that the ease-of-learning assessment is a better index of self-control for older children than for younger children.

What does jugdments-of-learning mean?

The judgment-of-learning tasks provide a better indication of the ability of children to monitor their own performance than the ease-of-learning judgments. The participant is asked to estimate his or her performance immediately after learning or a few minutes later. For adults, the assessment is always better if a few minutes have passed. This also appears to be true for children aged 6, 8 and 10 years (Schneider et al., 2000b). Children are able to make good judgments of their learning if there is some time inbetween the learning and asking for a judgment. Scheider et al. stated that age development in judgment-of-learning was negligible and therefore self-monitoring did not really develop with age. They suggested that the development of self-regulation ensures the development of children's metamemory, rather than the development of self-monitoring. This is also in accordance with the previously discussed results of Defresne and Kosibagawa.

What does feeling-of-knowing mean?

Feeling-of-knowing is based on the judgment of how much information can be retrieved and whether this information is correct or incorrect (i.e. how much you know). There does not seem to be a better performance as children get older. For adults, feeling-of-knowing judgments are the same for correct information as for incorrect information. The judgments are lower for errors due to the omission of information. We also see these patterns in children. Children therefore have knowledge about their mental system and can use this knowledge to make predictions about their performance in the future.

The conclusion from the studies into ease-of-learning judgments, judgments-of-learning and feeling-of-knowing judgments is that self-monitoring in young children is fairly accurate, because children perform about the same as adults. Self-regulation is what continues to develop into adolescence and is therefore also responsible for the development of meta-memory.

What is source monitoring?

Source monitoring is the attribution of the correct source of one's memory, knowledge and beliefs. Drummey and Newcombe (2002) investigated this and had the experimenter and a doll each tell five new facts to a child aged 4, 6 or 8 years. After a week, they were asked how they got the knowledge. The youngest children made the most mistakes. If they did not know and had to make a choice between a number of sources (such as the parent), they attributed it to the parent, the TV or a teacher 60% of the time. The older children sometimes just didn't know for sure whether they had heard it from the experimenter or from the doll. They still remembered that they had learned it in the experiment.

Ruffman et al. (2001) showed that if there is less time between acquiring the knowledge and naming the source, the 8-year-olds perform better on source monitoring. Children who cannot recall the source of their knowledge are more vulnerable to suggestions.

Source monitoring develops between 4 and 8 years and depends on the nature of the material to be remembered and its salience to the child. Research shows that children who have better metamemory perform significantly better on memory tasks. The relationships are bi-directional: metamemory influences behavior and this in turn ensures better metamemory.

How are executive functioning and meta-memory related?

Traditionally, the central executive system, executive function and inhibitory control were studied separately. The central executive plays a central role in cognition through planning and the monitoring of cognitive activity. Inhibitory control is the ability to inhibit responses to irrelevant stimuli while trying to achieve a cognitive represented goal. These three concepts are now being studied together in executive functioning and metacognition. The development of conscious control over thoughts, emotions and actions is studied in this field.

The frontal cortex is important for working memory, for controlling behavior and for inhibiting inappropriate behavior. Children are not good at the last two aspects, and people with damage to their frontal cortex have difficulties with this too. It seems that people with damage to their frontal cortex realize that, for example, when sorting cards, they sort by color, whereas they have to sort by shape. However, they cannot inhibit their behavior.

According to neo-Piagetans, the working memory plays an important role in cognitive development, because it increases with age, which is accompanied by more and better processing.

The Dimensional Change Card Sort Task (DCCS) is a test in which the participant must sort cards according to different rules or dimensions. For example, you are asked to sort by color and then by shape. Frye et al. (1995) used this test in their research on 3-, 4- and 5-year-olds. The children first had to sort five trails in one sorting strategy (e.g. shape), then five trials in another sorting strategy (e.g. color) and then five trials in which they had to keep changing (so the first trial by color and the second by shape and the third by color etc.).

Children aged 5 years had no problem with this, but the 3- and 4-year-olds had difficulty with switching their sorting strategies between the first two sets of five trials. In the third set of five trials they performed at chance level. Fyre et al. checked whether the poor performance was due to the fact that the shapes they used were abstract and unfamiliar, by using pictures of boats and rabbits, but that turned out not to be the explaining factor. The conclusion was that the ability of children to make judgments on one dimension while ignoring the other dimension emerges between 3 and 5 years.

Another explanation for the results of Fyre et al. is that children cannot inhibit the rule used before the strategy-switch, even though they are aware of the other rule. In order to make this conclusion it was necessary to test if children were aware of the second rule and this was done by Zelazo et al. (1996). 89% of the 3-year-olds could repeat the second line verbally and was therefore aware of it. The children were also unable to switch sorting behavior if the rule changed after just one trial, suggesting that it was not a matter of getting used to not being able to ignore the first rule.

In another study (Jacques et al., 1994) the task was performed by dolls, so that an incorrect motor response from children was excluded. Here, children rated the behavior as incorrect when the dolls played according to the second rule and also if the dolls played according to the first rule in the second five trials, in which the second rule had to be applied. This data supports the idea that the important factor in not being able to follow the sorting rules is a representational conflict rather than the inhibition of a response.

Krikham et al. (2003) had children rename each card before the switch: so at the beginning they were asked about the color of the card and before switching the sorting rule about the shape of the card. The 3-year-olds performed better compared to the standard procedure (78% instead of 42%). If they put the cards from the previous trial with the picture face-up, so that the child could still see the rule of the previous trial, the 4-year-olds scored worse (57% instead of 92%). This led to the conclusion that the critical feature was to inhibit a mind-set that no longer applies. Young children have difficulty in flexibly shifting their focus between representations that are in conflict with each other.

Despite all the different explanations about the cause of why young children do not function well on the DCCS, the test has been considered a good measurement for testing executive functioning in young children.

What is the difference between cool and hot executive functioning?

A distinction is made between cool executive functioning and hot executive functioning. Cool executive functioning refers to purely cognitive tasks and hot executive functioning also includes making decisions about events that have emotionally significant consequences. Hot executive functioning was tested on children aged 3 and 4 by using a gambling test. A card with a happy face made sure the children won M&M's and with a sad face they lost M&M's. Two decks of cards were used: from the first deck you could win large amounts of M&Ms, but also lose large amounts. In the second deck, luck and loss were less disastrous. Over the 50 trials it was smarter to take cards from the second deck. Children aged 3 took significantly more from the first deck and children aged 4 took significantly more from the second deck. Hot executive functioning therefore develops in a similar way to cool executive functioning: it develops between 3 and 4 years. Children must also learn to inhibit their response to the more conspicuous deck in order to achieve better performance.

How does control of inhibition work?

There are two types of measures to test the control of inhibition. In one task, children must delay gratification of a desire (delay test; for example, wait with eating the candy until the experimenter rings with a bell) and in the second task, children must respond in a way that conflicts with their first impulse response (conflict test; for example, when a sun is presented they must say "night" and if a moon is presented they must say "day"). From a series of delay of gratification tests, Kochanska et al. (1996) concluded that older children were better at controlling inhibition and that girls performed better than boys.
Diamond and Taylor (1996) conducted research with the conflicting tests and used the day and night test mentioned above. They also used a test in which the experimenter tapped on the table with a stick. If he tapped once, the child had to tap twice, and if he tapped twice, the child had to tap once. Children aged 3-7 were tested and the older the children, the better the performance. The performance in later trials, however, decreased. The conclusions were that children between 3 and 6 exercise inhibitory control. It was suggested that the errors were caused by errors in inhibition (more often tapping the stick than once or twice), forgetting a rule or being unable to switch rules. The researchers also thought that growing control over inhibition is related to the development of the frontal cortex.

Hughes (1998) conducted a remarkable research with various tests on inhibitory control, working memory and the flexibility of attention. She measured this in children aged 3 and 4 years. She concluded that the executive performance of pre-schoolers can be measured and that there were age-related developments in the executive performance of these children. According to her, executive functioning was a construct that covers many facets.

Is planning an independent aspect of executive functioning?

Another aspect of executive functioning is planning. Carlson and Moses (2001) focused on planning. They gave a battery of tests for executive functioning to 107 3- and 4-year-olds. Both delay tests and conflict tests were used and these tests measured the inhibitory control. They also included a planning test. In this test there was a keyboard with four keys that had different colors. The children were only allowed to touch one key at a time with their index finger. They had to play certain sequences until the experimenter said "stop." The number of sequences that they played without skipping a key or without touching a key twice was scored.

The results for the inhibitory control showed that the score on the battery was correlated with age, gender, verbal performance and the scores that parents attributed to their children for inhibitory control. The researchers concluded that inhibitory control could indeed be measured regardless of age, gender and verbal performance. The task that tested planning was of poor quality, because no action had to be planned when touching the keyboard and therefore there was no real cognitive activity in this task.

That is why Carlson et al. (2004) did a follow-up test with three planning tasks: the Tower of Hanoi, the truck-loading task and the kitten-delivery task. All three tasks involved reasoning according to the 'if-if-then principle'. An example for the truck-loading task would be: 'if the invitations can only be picked up from the top of the stack and if the pink house is last, then the pink invitation must be loaded first'. The results were that children aged 3 and 4 years performed roughly the same on the Tower of Hanoi and the kitten-delivery task, but that older children performed better on the truck-loading task. The performance on the Tower of Hanoi test was significantly correlated with the truck-loading task, but not with the kitten-delivery task. When the relations between planning and inhibitory control were explored, there were no significant correlations if the results were controlled for age and verbal performance.

The conclusions were that planning and controlling inhibition were two different constructs in executive functioning and that they both develop at the age of 3 and 4 years.

Again, it appears that the core factor of meta-memory is self-regulation. Efficient planning and a efficient degree of inhibitory control are important for effective self-regulation.

What is the relation between executive functioning and the theory of mind?

Many studies have shown that there is a relationship between executive functioning (metacognition) and the theory of mind (metarepresentation). Surprisingly, various studies show that the extent to which children make use of planning is not significantly correlated with performance on theory of mind testing as the false belief test. Carlson et al. (2004) suggests that this may be because conflict inhibition (rather than delay inhibition) is the most important aspect of the relationship between executive functioning and theory of mind. This suggests that the common link is the need to control conflicting representations. However, their research did not show to what extent cognitive development explained individual differences in performance on executive functioning and the theory of mind.

Lang and Perner (2002) argued that there was a correlation between the theory of mind and executive functioning because the understanding of beliefs, the understanding that a reflex is involuntary, and executive functioning all depend on one developmental factor; the understanding of mental states that are causally responsible for action. However, this idea is already undermined because the concept of false beliefs is already present before the age of 4 years and therefore before executive functions are properly acquired.

What is the role of the frontal cortex and the prefrontal cortex in executive functioning?

Executive functioning is associated with the frontal cortex. This continues to grow for a relatively long time, which could explain why children only develop executive functions at a later age. Bunge et al. (2002) investigated this on the basis of fMRI with children from 8 to 12 years and adults. They had the participants do a flanker test in which the participants had to indicate to which side the central arrow pointed. There was a neutral condition where the direction was clear, a congruent condition where the other arrows pointed to the same direction as the central arrow and an incongruent condition where the rest of the arrows pointed to the other direction as the central arrow. There was also a no-go condition where the arrow was surrounded by crosses and the participants were not allowed to press.

The children performed worse than adults in the incongruent and no-go conditions, but did give the right answer in 90% of the cases. Under these conditions, the brains of children were associated with activity in the left ventro-lateral prefrontal cortex and in the brains of adults with activity in the right ventro-lateral prefrontal cortex. If the children who performed the same as adults were studied separately, there was more activity in the left cortex, in contrast to the expectation that the activity would rather take place in the right cortex.

Durston et al. (2002) used a more child-friendly go/no-go test with Pokemons. They showed that children aged 8 showed activity in the right ventro-lateral prefrontal cortex in some cases. This may be because this task triggers much more motivation among children. In adults it is assumed that the right inferior frontal gyrus plays a critical role to the inhibition of responses in go/no-go tests.

Schroeter et al. (2004) used the Stroop test (in which a participant must name the color of the word and not the word itself: so for example when 'green' is written in red letters, then the participant must say red and not green) and fNIRS (functional near-infrared spectroscopy) in their research. In both children and adults, there was activity in the left lateral prefrontal cortex and not in the right. This is probably because this side also plays a role in language. It was found that the activity in the dorso-lateral prefrontal cortex increased with age.

Luna et al. (2001) used the oculomotor response suppression task. In this task, the participant had to look at the center of his visual field. If a red dot appeared on one side, the participant had to suppress the reflex to look at the side of the dot and look at the other side instead. If a green dot appeared, the participant had to look at the dot. Children aged 10 and 15 and adults were tested. The 10-year-old children made many mistakes. Only in adolescents there was a large brain activity in the dorso-lateral prefrontal cortex. Children and adults do not show this pattern.

The studies prove that the maturation of the frontal cortex is important for executive functioning. However, all studies are correlational studies and it is therefore not clear whether executive functioning is caused by the maturation of the prefrontal cortex or whether executive functioning causes changes in the brain structure.

What is the connection between reasoning and metacognition?

Inductive reasoning is the generalization of specific examples, making an inference from particular premise and reasoning according to analogies. Deductive reasoning is about reasoning about logical problems that have only one correct answer. Reasoning with the help of syllogisms is an example of deductive reasoning. Both inductive and deductive reasoning are influenced by similar factors and are the subject to similar heuristics and biases in children and adults.

Analogical reasoning is the ability to reason about novel problems based on the relationship between this novel problem and a familiar problem. This is a form of inductive reasoning. For reasoning by analogy, it is necessary that the correspondences between the novel problem and the familiar problem is noticed. Most research into analogical thinking in children looks at whether children are able to recognize the relationships between familiar problems and novel problems and whether they can use reasoning about relationships to solve analogies.

Whether children have the cognitive ability to map relations has been extensively investigated on the basis of analogies. An example of such an analogy is: bird is to nest as dog is to ...? (doghouse). Children aged 4 can already make such analogies as long as they are set in a well-known domain, for example depending on a task in the form of a game in which they have to complete series of pictures. Children are good at making these analogies. Even if they make a (correct) analogy that was not the original intention of the research and they have to choose the correct picture, they choose the right analogy. If the analogies are about causal relationships such as slicing (apple is to sliced ​​apple as cake is to sliced ​​cake) children of 3 already understand this.

This is difficult to investigate in children younger than 3, as it is an abstract test. In these children, the use of ingenious problem analogies is necessary to test their knowledge of analogies. With problem analogy, the child must solve a problem and this problem can be solved in the same way as a previously presented and solved problem.

What are important conclusions from research by Singer-Freeman?

Singer-Freeman et al. (2005) investigated 2-year-olds could make analogies if a similar problem had first been solved by the experimenter. For example, they first showed children how to attach a rubber band to two plexiglass poles so that an orange could roll across. The child was allowed to play with the orange for a while and then they were asked if they could let a bird fly. The child received a model with a tree on one side and on the other, an elastic band and a toy bird. The idea was that the child should do the same as the experimenter had done with the rubber band.

In the control condition (children who had not had an example) 6% of the 2-year-olds came up with an accurate solution and in the example condition 28%. If hints were given to make the analogy, 48% came to the right solution. These results are similar to the spontaneous results found in adults and 4-year-olds. When children notice the relationship between different problems, it is possible that they will transfer their knowledge and therefore perform metacognition.

What are important conclusions from research by Brown?

Brown et al. (1986) investigated problem analogy on the basis of the 'Genie problem'. In this problem, a genie moves from one location to another and must take his jewelry with him without damaging the jewelry. The solution is to roll them in his flying carpet. The children were presented with this solution and they were asked questions about the situation, so that they received metacognitive support. Then the 'Easter Bunny problem' was presented in which an Easter Bunny is a bit late with delivering easter-eggs and therefore gets help from a friend. He must bring the eggs to the friend's house without breaking them. 70% of children aged 4 and 5 years came spontaneously with the right solution based on the analogy. For children in the control condition (who had not received extensive questions about the situation) this was only 20%. The conclusion was that children find it easy to recognize similarities in problems as long as they still had the relational structure of the previous problem in their memory. This is reinforced by the metacognitive support.

In further research, Brown investigated whether children were able to figure out that they could use analogies through a series of problem solving problems. He presented an A - B - A - C sequence of problems. Children were usually unable to solve problem A. Problem B was easier and when it was solved problem A was presented again and the experimenter gave hints about the similarities between problem A and B. After successfully solving problem A, problem C was presented which also had a relationship with problem A and B. Children in the control group received no hints about the relationships between problem A and B and thus had not obtained any meta-knowledge about the use of analogies in solving the problems.

98% of the 7-year-olds from the meta-cognitive group were able to solve problem C and 38% of the children in the control group. Related work by Brown showed that 3-year-olds from a meta-cognitive group also performed better. This study shows that very young children select meta-knowledge that facilitates inductive reasoning. This inductive reasoning is facilitated if they follow hints, as long as they have rich conceptual representations of the domain being studied and as long as they are interested in the subject. As children learn more about the world, the types of analogies they are able to make become deeper or more complex.

A problem that can be solved by deductive reasoning has only one right solution, which can be found by studying the conditions. An example is: all dogs have tails. Guus is a dog. Does Guus have a tail? Yes! This is an example of a syllogism. It does not matter whether the conditions seem to be true or not, because the conditions can only provide one possible answer.

What are important conclusions from research by Dias and Harris?

Dias and Harris (1988) had children aged 5 and 6 solve syllogisms and had one group solve these problems in a game condition, where they could use toys that depicted the syllogisms. A second group simply had to listen to the syllogisms without being allowed to use toys. The children in the game condition performed well in solving the syllogisms when the syllogisms were about known facts, about unknown facts and about problems that contradicted facts. The children in the listening condition only performed well if the syllogisms were about known facts.

In a follow-up study the children in the game condition were offered syllogisms that conflict with factual information and it was said that it was possible on another planet (ie cats that bark). Again they performed better than the control group and the conclusion was that children can reason deductively, even in conditions that are in conflict with facts, but only if the problems were presented in a context of play.

What are important conclusions from research by Leevers and Harris?

However, this conclusion was contradicted by Leevers and Harris (2000) who asked four-year-olds to think carefully before they answered. If this was emphasized, the children could also perform well on the syllogisms that presented conditions that conflicted with the facts. Even if they were tested again after a while without being instructed to think about it, they performed well. Leevers and Harris concluded that this achievement was because they encouraged the children to process the conditions mentally.

Another measure of deductive reasoning is the selection task. This method focuses on 'if p, then q' reasoning. For example: if the envelope is sealed, it has a stamp on it. Conclusions must then be drawn from the situations presented. This can be p (a sealed envelope, shown face-down), non-p (a non-sealed envelope, shown face-down), q (a stamp on the envelope, shown face-up) and non-q (no stamp on the envelope, shown face-up). The participant has to decide in which situation the least evidence is needed to comply with the rule. The correct answer in this example is p or non-q. When it comes to familiar situations, adults perform well on this task, but when it comes to novel situations, they perform well in only 10% of the cases.

Light et al. (1989) showed that 6- and 7-year-olds also show successful deductive reasoning. Children could manage an abstract version of the familiar situation (so if cars and trucks were used first and were later replaced with triangles and cubes).
Harris and Nunez (1996) used a simpler task with 3- and 4-year-olds, in which they only had to choose the condition that depicted a violation of the rule. In the example 'if Nancy goes outside, she has to put on her coat', the picture that presents that Nancy goes outside without a coat is the condition that is not allowed. Children were able to select the correct pictures, also in unfamiliar conditions.

In this chapter metacognition, reasoning and executive functioning are discussed. Metacognition is knowledge about cognition. Flavell (1979) defined this concept as any cognitive activity or knowledge that takes an aspect of cognitive activity as its cognitive object. An example of this is the conscious use of cognitive strategies to increase your performance. Meta-memory (knowledge about memory) and executive functioning are separately studied aspects of metacognition. Executive functioning is the ability to regulate behavior and thoughts, planning behavior and inhibit inappropriate behavior. To develop metacognition, children must be able to treat cognition itself as an object of cognition. Metacognition and meta-representation are closely linked. However, metacognition is focused on knowledge about someone's own mind and not about that of others, such as meta-representation.

How do reading skills and mathematical skills develop? - Chapter 10 (1)

How are numbers represented mentally?

Cognitive neuroimaging research by Dehaune and his colleagues suggested that there are three coding systems for numbers in different areas of the brain:

  1. A visually based code for Arabic numbers, located in the fusiform gyrus.

  2. A linguistic system for storing number facts. For example, the multiplication tables, which are practiced so often that they are stored linguistically. This happens, for example, in the same way as how the days of the week are stored. This is found in the left lateral language areas, such as the left angular gyrus.

  3. A general number of sense, located in the parietal lobes.

Dehaune also argued for an analog magnitude representation in the horizontal intraparietal sulcus. This is an internal mental representation of continuous quantities. This indicates that the brain uses a kind of internal continuum (e.g. a number line) when comparisons between continuous amounts are made. This continuum is an analog of the external stimulus. In the case of numbers, there is an analog representation of numbers. These are not mentally discrete entities that indicate exact quantities, but they indicate estimates of quantities. The greater the quantities, the less precise the representations for these numbers.

Cognitive and neuroimaging research supports the idea that different properties from the physical world (the spatial world) are represented in the brain in an analog form. Thus, Weber's law, among other things was discovered. This law states that our ability to make physical discriminations is sensitive to ratio. The threshold to distinguish between different numbers of stimuli becomes higher as the stimulus intensity becomes larger. For example, if an adult, when seeing a number of dots shown briefly, has to decide whether there were 12 dots, performance is better for 4 or 20 dots than when 10 or 11 dots are presented.

The symbolic distance effect is also a form of analog representations. This means that an adult has to think longer if a number is greater than or less than 5 if the numbers are close to 5 (like 4 or 6) than if they are distant to 5 (like 1 or 9). The closer the ratio gets to 1, the harder it is. Infants are also ratio sensitive when it comes to discriminating quantities. Infants perform better with comparisons of, for example, 8 and 16 (ratio 1:2) than with 8 and 12 (ratio 2:3 and therefore closer to 1). This happens with numbers larger than 3 and therefore shows reliance on the analog magnitude representation. Analog representations are not precise and therefore two things can be said about the representations of numbers in children: (1) the representations of numerically close quantities is similar and (2) the precision of coding becomes worse as the quantities get larger.

Numbers smaller than 3 and 4 use a different mechanism: subitizing. This means that children can distinguish the differences between very small sets of numbers without counting. The distinction of small quantities is therefore not processed via the analog magnitude representation system.

Older children also use the analog magnitude representation system to distinguish between different quantities. Huntley-Fenner and Cannon (2000) did an experiment in which they showed two rows of black squares to children aged 3, 4, and 5 years who had to decide which row contained the most squares. The rows varied from one to fifteen squares and made use of the ratios 1:1, 1:2 and 2:3 between the different rows. Most children did not use counting and if they did, they only did so in a minority of trials.

If the children did count, they did not perform better than if this was not done. Children were more successful in correctly distinguishing the largest row if the ratio was distant from 1.

Barth et al. (2005a) also checked the perceptual variables in a similar experiment (for example: different quantity, but spread over an equally large area). The 5-year-olds scored significantly above chance level and showed a reliance on an analog magnitude representation for number.

In a follow-up experiment (2005b) the size of the dots, the total contour length, the summed dot area, and the density were negatively correlated with number in half of the trials. In the other half of the trials, these aspects correlated positively with number. If children base their assessments on perceptual variables, they would perform worse on the negative correlating trials. The 5-year-old children performed significantly well and also showed an effect of ratio. Barth et al. concluded that the children's assessments were determined by an abstract and non-linguistic mechanism.

Jordan and Brannon (2006) showed children pictures of a number of displays in the first phase of their experiment (for example, 2 dots). Then two other displays appeared. Of these displays, one was similar to the previous display (also consisted of two dots) and the other did not (consisted of, for example, of eight dots). The child had to identify the display that corresponded to the display shown before. In the second phase, another display was shown and then two new displays were shown, only these displays did not match the quantity of the display shown before. The child had to choose the display that most closely resembled the quantity of the picture shown before. The children were not told this explicitly; it was only told that the children had to play the same way as in the first phase. The choices in the second phase were always 2 and 8 or 3 and 12. For example, the researchers could observe whether children rated smaller targets as small and larger targets as large. The amount for which children are equally likely to choose the small estimate (2 or 3) as the large estimate (8 or 12), should be the average. The researchers concluded that children did indeed choose 8 and 12 as the estimate of the stimulus and that the children chose the small estimate as often as the large estimate; the geometric mean. So they were relying on analog magnitude representation.

Generally, it is stated that the analog magnitude representation is located in the intraparietal cortex in the brain. Deheane et al. (1999) investigated brain activity in adults using fMRI to make exact additions (e.g., 4 + 5 = 9) and approximate additions (e.g., 4 + 5 = 8). During exact additions there was a large activity in the left lateral area in the inferior frontal lobe (the language area). During approximation additions, there was greater activity in the bilateral parietal area that plays a role in visuospatial processing. Exact calculation therefore depends on numerical facts in the language area and the approximate calculation depends on visuospatial parietal networks.

Pinel et al. (2001) found that there was a numerical distance effect and that there were faster responses to the numbers that were more distant from the target number in adults. There was also high activity in the parietal cortex at small distances from the target.

Cantlon et al. (2006) investigated where the analog magnitude representations is located in children's brains. They investigated the neural activity of their task in both children and adults. If a difference in quantity was presented, this resulted in increased activity in the right intraparietal sulcus in children. The activity in adults was measured in the same area, but bilaterally. If differences in shape (squares instead of circles) were presented, activity was measured in the visual areas, such as the fusiform gyrus, in both groups.

Cantlon et al. concluded that the intraparietal sulcus for non-symbolic numerical processing is developing early.

Temple and Posner (1998) investigated the distance effect of 5-year-old children by comparing their EEG with that of adults. They let the children decide whether a number (1, 4, 6, or 9) was larger than or smaller than 5 and measured the response times and EEG. The EEG waves were approximately the same in children and adults and the neural distance effects also occurred at about the same time (about 200 ms after showing the stimulus). Children did have a longer reaction time than adults. The activity was measured in the parietal area for both adults and children.

How does counting develop?

The development of a symbolic number system probably starts with counting. Children learn this from an early age, but that does not necessarily mean that they also understand what they are doing. At first, it is probably purely linguistic. The realization comes gradually. Children learn the principle of cardinality. This means that they understand that all sets with the same number contain the same number of objects. They also learn the principle of ordinality; that numbers are presented in an ordered scale of magnitude (4 is always greater than 3).

Gelman and Gallistel (1978) stated that children from the age of 2 already use numbers in systematic ways. They partly understand cardinality, ordinality and one-to-one correspondence (that when counting, each object may only be counted once). However, they are not yet fully aware of it.

Saxe (1977) did systematic research for the first time on children aged 3, 4 and 7 to study the development of counting. All children could count to 9 and he had children perform a number of tasks (for example, show a number of beads and then ask the child to lay down the same number of beads as in the example). Counting becomes more accurate between ages 3 and 4 years. Children also go from "prequantitative" counting to "quantitative counting". With prequantitative counting, counting was not used to produce the same number. A child counts the objects (twice) until he reaches the correct number. In quantitative counting, the child uses one-to-one correspondence.

In a longitudinal study, Saxe discovered that the change from prequantitative to quantitative counting was related to increasing accuracy of counting and thought that the acquisition of one-to-one correspondence was the basis for understanding the logic of counting. He did speculate that this may not apply to the counting of small numbers.

Wynn (1990, 1992a) did valuable research into counting small numbers. Children were asked to give one, two, three, five or six dinosaur toys to a doll. He then asked them to check the given amount (by counting) and had them correct it if it was wrong. The children aged 2.5, 3 and 3.5 years responded in two ways: (1) they counted the toys, or (2) they grabbed a handful for the dinosaur. The children who simply grabbed were accurate if they had to grab one, often also two, but more often the number of grabbed toys did not even come close to the requested number. Wynn concluded that these children did not understand cardinality.

Around the age of 3.5, just grabbing a handful of toys is replaced by counting. Wynn suggested that this is because children at this age understand more about counting. If they understand that one, two and three are a sequence that follows one another, the understanding of all other numbers will also improve.
A longitudinal study by Wynn (1992) showed that children who understand the cardinality of numbers use counting during the give-a-number task and two related tasks: a how-many task and a point-to-x task. Children who understand the cardinality of only small numbers could not distinguish different numbers. But even young children already understand that number words refer to specific quantities.

There is currently a debate about whether younger children regard all larger quantities as "a lot". It is suggested that if children are frequently exposed to quantities larger than four, they are more likely to experience a conceptual change. First, they have a representation of one, two, three, four and everything larger than 4 is called 'a lot'. Le Corre et al. (2006) state that learning to count as representations of quantities requires a construction of a new representational format. Children must recreate the construction for themselves. This depends on the extent to which they are exposed to the verbal count sequence.

In an experiment by Sarnecka and Gelman (2004), children were given the six-versus-a-lot task in which it was examined whether children see the name 'six' as an exact amount or simply as another name for 'a lot'. The results showed that children aged from 2 years and 7 months to 3 years and 6 months did not treat the term 'a lot' as specific, but the term 'six' was treated specifically. In contrast to what Le Corre et al. (2006) claimed, children do have an idea of ​​the exact amount of larger numbers.

Before children understand the sequence of a series of numbers, experience with the language of counting provides social and cultural support for cognitive development. Based on this, children can form a cognitive structure for counting. We must, however, bear in mind that the claim that learning to count encourages the development of a symbolic numerical system does not mean that language and numeric are independently linked. Patients who have deficiencies in their linguistic skills may still have excellent math skills.

Also, the claim that learning to count encourages the development of a symbolic numerical system does not mean that children have no concept of large quantities before learning to count. Gelman and Butterworth (2005) argue that if children understand the concept of "one", they can add "one" infinitely. However, they state that it is useful for learning enumeration and arithmetic. Furthermore, it is likely that if counting also provides a concrete representation of the ordinal aspect of numbers, this is very important for the development of a child.

What are cultural differences in the language of numbers?

In America words like 'eleven', 'twelve' etc. are used, but in Asian languages ​​these numbers are labeled 'ten-one', 'ten-two' etc. Words like 40 are labeled 'four-tens'. Miura et al. (1988) investigated whether this difference in language had an effect on the representation of figures in children. They allowed children aged 6 and 7 to construct quantities with blocks. For example, there were blocks that represented ten units. Children from Asian countries performed better and had a different cognitive representation for the figures. This was attributed to the difference in language. However, if American children were first shown how to use the units in an exercise phase, they performed statistically equivalent as Asian children in a test phase.

Hodent et al. (2005) also argued that language could play a role in numerical representation. In French, for example, 'un' means one, but it is also used to distinguish singular and plural. It can therefore also refer to 'something'. This does not occur in English. They tested 2-year-old French and English children and predicted that French children would not judge 1 + 1 = 3 as strange, because 1 + 'something' could be three. They also predicted that if a sum would start with 'un' (e.g. 1 + 2 = 4), it would be assessed differently than if a sum would not start with 'un' (e.g. 2 + 1 = 4). The French children would then understand that in the latter case 'un' represented a number and that 2 + 1 = 4 is therefore incorrect.

Indeed, the French children were not surprised by 1 + 2 = 4, but they were surprised by 2 + 1 = 4. English children both thought for both computations that it was strange. This indicates that language plays a role in the cognitive representation of numbers.

The opposite idea is that it is not necessary to have a system for the names of numbers in order to have an idea about large quantities. This has also been supported by studies on linguistics. Some tribes in the Amazon do not have words for quantities larger than five but can perform well on a test if they have to indicate the largest amount. This suggests the use of Weber's law. The tribes therefore also appear to use the analog magnitude representation.

From a different perspective, learning to count is very important for the development of a symbolic numerical system. It is suggested that children already start to look at number words as references to a specific quantity, even though they do not know exactly to which quantity the word refers.

How are numbers represented mentally?

Cognitive neuroimaging research by Dehaune and his colleagues suggested that there are three coding systems for numbers in different areas of the brain:

  1. A visually based code for Arabic numbers, located in the fusiform gyrus.

  2. A linguistic system for storing number facts. For example, the multiplication tables, which are practiced so often that they are stored linguistically. This happens, for example, in the same way as how the days of the week are stored. This is found in the left lateral language areas, such as the left angular gyrus.

  3. A general number of sense, located in the parietal lobes.

How do the theories of Piaget and Vygotsky relate to each other? - Chapter 11 (1)

Why are the theories of Piaget and Vygotsky important to cognitive development?

According to Piaget, logical reasoning was constructed by active experiences with the external world. A child constructs cognitive "schemes" based on these experiences. The child constructs his or her own knowledge from active experiences with the external world. According to Vygotsky, the social world plays a key role in cognitive development and adults are important in mediating the development of a child by supporting them to acquire cultural knowledge. Connectionism has also been proposed as a model of complex cognition. However, none of the theories are extensive enough to explain everything and there is probably a collaboration between the social world and a child's own knowledge construction. The biological principles must also be observed. Research into how the brain works is done in a new theoretical framework called neuroconstructivism.

What has Piaget contributed to cognitive psychology?

Piaget was originally a biologist. His theory claims that knowledge structures, or schemas, adapt themselves to their environment. The cognitive system seeks for balance and cognitive development takes place on the basis of two processes: accommodation and assimilation. With assimilation new experiences are incorporated into existing schemes. With accommodation the cognitive schemes are adapted when new experience does not fit reality. Hereby knowledge becomes broader and deeper.

Piaget has proposed a stage model of cognitive development. He argued that children go through four major changes in cognitive schemes until they are adults. The acquisition of each new phase does not necessarily have to take place across all domains at the same time and variability could also take place within the different stages. The stages that Piaget proposed were:

  1. The sensory-motor period: 0 - 2 years. Sensory-motor cognition is based on the infant's physical interactions with the rest of the world.

  2. The period of pre-operations: 2 - 7 years. This is the beginning of representational thinking, which also includes the internalization of actions on the mental level.

  3. The period of concrete operations: 7 - 11 years. If the above-mentioned internalizations, which are called compositions, become mentally reversible, concrete operational cognition develops.

  4. The period of formal operations: 11 - 12 years. In this stage, certain concrete operations are connected with each other and this is the beginning of scientific thinking.

The ages that Piaget mentioned were not fixed and there may be some variation.

What changes occur during the sensory-motor period?

One of Piaget's major assumptions is that a child constructs his thoughts through actions. This has also been supported by cognitive neuroscience. A baby is born with a number of motor skills (such as sucking and grasping). Perceptual possibilities (which are also present at birth) enable the infant to form hypotheses about the world. A child is always busy interpreting and reinterpreting perceptual information. Piaget distinguished six sub-stages in the sensory-motor period:

  1. Modification of reflexes: the baby can, for example, adjust his sucking reflex to the contours of the mother's nipple. In addition, the baby assimilates his or her sucking response to all sorts of other objects and makes a distinction between objects that do and do not provide food.

  2. Primary circular reactions: circular reaction is repetitive behavior to maintain a certain sensory experience. The behavior is called primary because the first behaviors are self-directed. An example is thumb sucking.

  3. Secondary circular reactions: these behaviors are aimed at the rest of the world. An example is always dropping something without the child getting bored.

  4. Co-ordination or circular reactions: a child can perform a series of actions to achieve a goal. An example of this is pulling on a blanket so that the toys on the blanket move towards the baby.

  5. Tertiary circular reactions: the children have a better ability to recreate events in the world. They can perform various trial-and-error actions to determine the results of certain actions. The focus is not on the repetition of the behavior but on the different outcomes. This can be described as hypothesis-testing behavior.

  6. Interiorization of schemes: a child can then determine the consequences of certain actions and come up with a series of actions that enable him to achieve a certain goal. This no longer requires trial-and-error actions. So, there is a cognitive representation of actions and their consequences. According to Piaget, this phase is the beginning of conceptual thinking.

The best-known example of how Piaget thought that sensory-motor information led to the development of conceptual thinking was his analysis of object permanence. He studied this by examining the search behavior of babies. According to Piaget, there was a full understanding of object permanence if invisible movements to a new location could be solved and the child could retrieve it again without knowing where the object was hiding. This ability develops at around 18 months. According to Piaget, a child at this age had a cognitive representation of the object, detached from motor actions and sensory perception.

Piaget was wrong about the age at which children develop cognitive representations. However, he was right about the fact that sensory-motor responses are an important source of information for a child and that these responses are important for the acquisition of knowledge through statistical learning, learning by imitation, learning by analogies and learning based on explanation. Action, rehearsal and recreation of sensory-motor experiences are important for the development of an infant's cognition.

Piaget also argued that sensory-motor behaviors were made representative through interiorization. For example, his own children imitated certain behaviors from the physical world with their own bodies. This also seems to be true. Furthermore, according to Piaget's theory, analogies are important in the generalization of sensory-motor schemes to new objects. If a new physics concept was understood, there was a rapid analogic shift. Recent research shows that infants aged three months are indeed making analogies for new, unfamiliar objects.

What changes occur during the pre-operational and the concrete operational periods?

Concrete operations are the set of logical concepts that describe groups of objects and their relations. The key concrete operations were the children's understanding of relations between objects with different lengths or heights (transitivity), the children's understanding of different classes of objects and their part-whole relations (class inclusion) and the children's understanding of addition, subtraction and equivalence (conservation). These three concepts and seriation (the understanding that objects can form ordered series on the basis of their physical or psychological characteristics) have been studied extensively in later research.

Piaget's explanation for these developments in logic was that during the pre-operational period the schemes change from sensory-motor to schemes based on mentality. Typical for the concrete operational period was that operations could also be argued reversibly (2 + 2 = 4 is similar to 4 - 2 = 2).

The main characteristics of pre-operational thinking were that it was egocentric (perception and interpretation in terms of the self and impossible from someone else), centration (focused on one aspect of an object or event and ignore other objects) and that there was a lack of reversibility (inability to mentally reverse a series of events or steps of reasoning). In the concrete operational period, the egocentric aspect gradually decreases, children focus on several aspects of a situation or object and they are able to apply reversibility. They also understand that every operation on an object can also imply its inverse.

Transitivity is the understanding that the relationship between two items that cannot be directly compared may still be inferred by reference to a third or more intermediate objects. An example of this is: A is greater than B, B is greater than C: who is greater, A or C? There are transitive relations between all entities that can be organized into an ordinal series. The given example can only be solved if concrete operational reasoning is possible. It was assumed that this form of reasoning only occurs from the age of 6 - 7 years, but Bryant and Trabasso (1971) proved that children could already make transitive conclusions if they are trained to remember the conditions (so for example, that A is always greater than B). They used rods of different sizes and different colors. In the training phase pairs of rods were presented and one inch of each rods was shown. The child was asked which rod was taller (or shorter). After answering, the solution was shown. During the testing phase, the children had to do the same but received no feedback about the real length of the bars. They were trained with four pairs of rods, but in the testing phase other pairs were presented to the child. Children aged 4, 5 and 6 years scored well in the critical B taller/shorter than D comparison in 78%, 88% and 92% of the cases respectively.

However, this task was criticized because a spatial cue would be given to the children so that they could get a hint about which rod was taller and therefore made their decision based on this and not because they had mastered transitivity. That is why Pears and Bryant (1990) eliminated the training phase and presented pairs of blocks of different colors. They then had to built a tower in which they had to apply the rules presented by the blocks (for example green is always above white). They concluded from the results that children aged 4 already have the ability to make transitive inferences.

Conservation is the understanding that some aspects such as weight, quantity and volume do not change if some aspects, such as the shape, do change. This can be measured, for example, by showing the child two vases with the same amount of water. In the presence of the child, the water from one of the vases is then poured into a wider vase of water, so that the child can see that the second vase contains the same amount of water. However, it seems to be less, because the vase is wider and the water level is therefore lower. Children under the age of 7 did not understand that the second vase contained the same amount of water as the first vase.

Elkind and Schoenfeld (1972) re-examined the results and came to the conclusion that children aged 4 have some understanding of conservation. However, most who have replicated Piaget's study have replicated Piaget's findings; full understanding of conservation emerges at the age of 7 years.

An aspect of the original conservation task has been criticized by many. This is the fact that the experimenter asks the same question twice. At the start of the experiment, if the two similar vases are presented, the experimenter asks if there is more, less or the same amount of water in one vase. After the transfer this is asked again. However, if someone asks the same question twice, the answer should usually differ from the first answer.

This could mislead children. McGarrigle and Donaldson (1975) omitted this aspect in their 'naughty teddy' study. The children were asked which row contained the most objects or whether it was the same amount, but before the child could answer, teddy came and altered the length of one row. The 4- and 5-year-old children did give correct responses. However, in this investigation it may be that the excitement that arises from teddy's interruption can influence the results.

Siegler (1995) conducted a study in which he used different types of training. His results showed that the concept of conservation occurred gradually and not suddenly. There were also major differences in the extent to which children benefited from the training in which they had to explain the reasoning of the experimenter. Siegler's conclusion was that children used different types of reasoning during a transitional period.

Class inclusion is the knowledge that objects can belong to one or more categories (lop-eared rabbits belong to the category lop-eared rabbits but also to the category of rabbits in general). Piaget tested this, for example, with regard to six flowers, two of which were white and four were red. He asked the child if there were more red flowers or more flowers. Children younger than the age of 6 responded that there were more red flowers because they are unable to deal with parts or with the whole. They could not think about the red flowers as two different things at the same time that belong to a different category; red flowers and flowers.

Markman and Seibert (1976) criticized Piaget's research because Piaget's question sounds strange, confusing the children and thinking that they should make a part-part comparison. In their research they used two conditions: one such as Piaget's original question and one in which the question was formulated differently. For grapes it was told: here is a bunch of grapes and there are green grapes and blue grapes. Who would get more to eat, someone who ate the green grapes or the bunch of grapes? In the standard Piaget condition, people were asked: who would get more to eat, someone who ate the green grapes or the grapes? Children aged 4 and 5 scored better in the adjusted condition (70% correct) than in the original condition (45% correct). Even 4-year-olds therefore understand class inclusion.

If a word is used that indicates a large amount and the child can observe a condition in which there are fewer, they still respond that this condition is the largest. This applies to 5- and 6-year-olds and was found in a study by Dean et al. (1981). For example, they said that there were a number of red ants and some green ants and that the red ants were an army and the green ants were not. Which group contained the most ants? While the green ants were clearly in the majority, the children opted for the red ants. Language can therefore distort our cognitive responses. Even adults can still be misled by this.

Research by Goswami and Pauen (2005) shows that children can use analogies when assessing class inclusion.

Logic of concrete operations are understood by children aged 4. Children reason in the same way as adults, but are rather misled by interfering aspects such as contextual variables, because they are worse at inhibiting irrelevant information. So, it seems that sensory-motor cognition is not redeveloped in the mental realm, as Piaget claimed. Rather, sensory-motor knowledge is supplemented by experience. This experience is sometimes active and sometimes acquired by language.

What changes occur during the formal operational period?

According to Piaget, children between the ages of 11 and 12 reach this period when they are able to take the results of concrete operations and to formulate hypotheses about logical relationships. Piaget called this operating on operations or second-order reasoning. Formal operational thought is the ability to apply formal systems such as propositional logic to elementary operations that describe categories of objects and their relations. It is also referred to as scientific thinking.

Piaget tested this thinking by having children manipulate independent variables and see if they could discover the correct rule. For example, he used a task with a pendulum, in which children need to discover the period of the pendulum. This depends on the length of the pendulum. Children begin this task with the conviction that the weight of the end of the pendulum affects the oscillation of the pendulum. They must keep the length constant and vary with the weight to find out that weight is not the cause. However, children younger than 11 - 12 years do not keep the other variables constant (in this case the length of the pendulum) and thus cannot find the correct solution.

Other research shows that talking to peers about the problem before a task is started ensures that children aged 9 years get a better understanding of a scientific task that they must perform (Howe et al., 2000).

Research shows that children can use analogies in their reasoning and test hypotheses prior to adolescence. Even adults sometimes make mistakes in hypotheses testing, so Piaget's prediction about age is incorrect.

It is also suggested that scientific thinking can be promoted through training. The key factors to be investigated for changes in the development of reasoning are familiarity with the problem, contact, metacognition and language.

However, Piaget's idea that cognitive structures resemble mathematical systems was insightful. Recent research supports this idea. However, the way Piaget described it may be incorrect. More emphasis should be put on machine learning algorithms like causal Bayes' nets and explanation-based learning.

What has Vygotsky contributed to cognitive psychology?

While Piaget focused on the individual child who constructed his or her own knowledge in cognitive development, Vygotsky emphasized the social context, culture and language that play an essential role in cognitive development. Language is the first symbolic system to which children can respond psychologically and therefore plays an important role in cognition. Vygotsky argued, and research supports this, that children develop cognition rather than language. A milestone for the child is when language and cognition come together and this develops in egocentric or private speech. A child can then organize his thoughts and this later develops into inner speech. Vygotsky saw thoughts at an early age as prelinguistic and early language as preintellectual, with purely social functions.

According to Vygotsky, language is just as important as action in attaining goals. The language allows children to disconnect themselves from the present situation to generate plans and to solve problems. They can also control their own behavior by, for example, expressing their intentions. Moreover, language enables children to ask adults to help them.

In addition to language, Vygotsky has identified a number of sign systems or cultural semiotic systems that make symbolic representations of knowledge possible. Examples are pictures, diagrams, writing and reading. These are also psychological tools for organizing cognitive behavior. Cognitive development is therefore dependent on contact with the environment. First this consists of contact with parents and teachers and later the knowledge is internalized. Sign systems can play a major role in a person's intellectual development because they can expand cognitive skills. For example, you can remember something better if you write it down. This was called mediated cognition by Wertsch (1985). These sign systems have a social nature because they are taught by the environment. However, there is always an interaction between social and cognitive processes.

Language and the use of psychological tools are taught in infancy. Psychological tools are the tools that humans have invented, such as gestures and symbols. The combination of psychological tool use and language changes psychological functioning to a higher level of cognitive activity in the child. These high levels of cognitive activity are specific to humans.

Vygotsky did not agree with Piaget's idea that children develop without any influence from learning at school. Children have prior knowledge before they go to school. For example, they already have experience with quantities before they are taught arithmetic.

When children go to school, they enter the zone of proximal development. This is the difference in what children can do on their own and what they can do with the support of adults or during collaboration with more able peers.

This idea has had a lot of impact in education. It proposes that children do not necessarily have to taught at their own developmental level, as the child's developmental level may change. It is beneficial if children are taught with regard to their zone of proximal development.

Vygotsky made a distinction between scientific concepts and spontaneous concepts. Spontaneous concepts are the concepts that a child generates through experiences and observations in the world and these concepts do not necessarily have to be true. Scientific concepts are acquired consciously and with effort in school and can transform spontaneous concepts and raise them to a higher level. Vygotsky himself has found no evidence for this, but neo-Vygotskians have. However, these investigations have been criticized. Vygotsky's idea that school and the zone of proximal development can change knowledge is very important. Neo-Vygotskians have also emphasized the joint activity of children and adults. If children are only helped verbally, this is not enough for to optimize learning.

An example of a successful program based on the ideas of Vygotsky is reciprocal teaching; a teacher models to a small group the optimal strategies for reading comprehension. These strategies include summarizing and predicting what will happen next. First the teacher leads the children and then he gradually withdraws and lets the children go their own way.

Neo-Vygotskians mainly focused on 'theoretical learning' in their research. This is an alternative to the constructivist idea of ​​Piaget. Theoretical learning about precise definitions of scientific concepts needs to be taught to children and they are not required to discover it themselves. Then they start working on it and internalize it. This leads to cognitive benefits because children eventually develop a general strategy that can be used when faced with a new problem.

According to Vygotsky, playing is also of great importance for child development. The world of imagination has a crucial psychological function in development, because it enables to fulfill the desires of a child, which is otherwise not possible (for example, being a mother). Vygotsky perceived play as a human form of cognitive activity. Play is characterized by creating an imaginary situation.

Play is not purely symbolic because it also has a role of motivation and it involves rules of behavior. Children try to be someone while they play and behave in the way they think they should be like. This can go unnoticed in real life, but will become a rule of behavior in play. If two sisters play that they are sisters, they will behave like sisters in the play, but in real life they are not aware that they also behave like sisters.

According to Vygotsky, children think things determine actions: a door must be opened, a shopping cart must be pushed. Children let go of this while playing: for example, they use a stick as a horse. In play they separate the perceptual cues from the status of the object. According to Vygotksy, they learn to think more mature in such a way that the meaning of objects is free from the situational boundaries.

According to Vygotsky, the rules of play ensure that children keep their impulses under control and, according to him, this facilitates the development of self-regulation. This is also supported by a child's playmates, because they can correct someone who does not follow the rules. Play also ensures that children can separate meaning from action. Finally, play may have a role in the zone of proximal development. Adults can assist in this (also a form of the zone of proximal development) by discussing real events with the child, such as a visit to the doctor.

Vygotsky's theories have had a greater impact on education than on developmental psychology. Neo-Vygotskians note that some of Vygotsky's theories have involved misunderstandings of Vygotsky's ideas in education. For example, Vygotsky suggested that cognitive development takes place in a socially meaningful activity, but also emphasized that teachers also have to teach children directly the acquired knowledge and should not always have them discover this for themselves.

Vygotsky further emphasized the importance of language and argued that it is just as important as action. In the West however, more emphasis is placed on how children create knowledge for themselves, through action. The emphasis that Vygotsky places on the role of interpersonal communication does have parallels with Western psychology as well as the role of play and imagination. However, there are no Western studies into the basis of play in self-regulation and metacognition. Western research programs tend to focus on non-playing tasks when examining metacognition.

What is neuroconstructionism?

Neuroconstructivism explains the mechanisms of cognitive change through biological constraints of neural activation patterns that compromise mental representations. According to neuroconstructivism, experiences are the key element of brain development, since they can change the brain's hardware. This causes changes in the nature of mental representations, which in turn leads to new experiences and further changes to the neural systems.

Neuroconstructivism distinguishes a number of constraints (limitations of development). An example of such a constraint is the biological action of genes. A gene cannot turn itself on or off, this is driven by certain signals. These signals can originate within the cell, outside the cell or outside the organism. The genetic activity therefore depends on neural, behavioral and external events. Neuroconstructivism therefore states that it is important to understand the interactions of the signals and the genetic activity in order to explain the development of cognitive disorders (such as depression).

A second constraint identified by neuroconstructivism is encellment. This is the fact that the development of neurons is constrained by their cellular environments. A similar restriction is enbrainment; the fact that the functional properties of brain parts are constrained by the co-development of other regions, for example by feedback processes and top-down interactions. These interregional interactions can influence the development of neural structures involved.

A third and final constraint is ensocialment. Social aspects of the environment have effects on social and behavioral development, for example by influencing gene expression.

What is connectionism?

While neuroconstructivism is concerned with development, connectionism is concerned with learning. Connectionists build computational models of cognition that represent a neural network.

A major contribution from early connectionist models was the in-principle demonstration that a simple network could learn the structure of the input (in this case the structure of a language).

As time went on, more and more in-principle effects were demonstrated, for example, that bottom-up processing could generate effects that were appropriate for behavioral data to require top-down processing.

Why are the theories of Piaget and Vygotsky important to cognitive development?

According to Piaget, logical reasoning was constructed by active experiences with the external world. A child constructs cognitive "schemes" based on these experiences. The child constructs his or her own knowledge from active experiences with the external world. According to Vygotsky, the social world plays a key role in cognitive development and adults are important in mediating the development of a child by supporting them to acquire cultural knowledge. Connectionism has also been proposed as a model of complex cognition. However, none of the theories are extensive enough to explain everything and there is probably a collaboration between the social world and a child's own knowledge construction. The biological principles must also be observed. Research into how the brain works is done in a new theoretical framework called neuroconstructivism.

Join World Supporter
Join World Supporter
Log in or create your free account

Why create an account?

  • Your WorldSupporter account gives you access to all functionalities of the platform
  • Once you are logged in, you can:
    • Save pages to your favorites
    • Give feedback or share contributions
    • participate in discussions
    • share your own contributions through the 7 WorldSupporter tools
Follow the author: Vintage Supporter
Promotions
vacatures

JoHo kan jouw hulp goed gebruiken! Check hier de diverse bijbanen die aansluiten bij je studie, je competenties verbeteren, je cv versterken en je een bijdrage laten leveren aan een mooiere wereld

verzekering studeren in het buitenland

Ga jij binnenkort studeren in het buitenland?
Regel je zorg- en reisverzekering via JoHo!

Access level of this page
  • Public
  • WorldSupporters only
  • JoHo members
  • Private
Statistics
[totalcount] 1
Content categories
Comments, Compliments & Kudos

Add new contribution

CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Image CAPTCHA
Enter the characters shown in the image.