Preoperational Stage

The preoperational period is divided into two stages:

Symbolic Function Substage 

This substage typically spans ages two to four. It is marked by the emergence of a child’s ability to use symbols to represent objects, people and events that are not physically present.

Children use symbols in language, drawing, and pretend play. This unlocks more complex, abstract thought.

The early preoperational period (ages 2-3) is marked by a dramatic increase in children’s use of the symbolic function. This is the ability to make one thing, a word or an object, stand for something other than itself.

The symbol need not resemble the real thing.

Symbolic representation allows children to think about objects, people, or events that are not physically present. This ability lays the foundation for language development, imaginative play, and later academic skills like reading and mathematics.

Concrete props help at first.

Early in this stage, children might use objects that closely resemble what they’re representing (e.g., a toy phone for a real phone). As they progress, they become more flexible, using objects that are less similar to what they’re representing (e.g., a banana as a phone).

By the end of this stage, children can use purely mental symbols without physical props.

Manifestations of symbolic representation:

• Language use: Words become symbols for objects, actions, or ideas. For example, the word “dog” represents the concept of a dog, even when no dog is present.
• Object substitution: Children use objects to represent other things in play. A block might become a phone, or a stick might become a magic wand.
• Drawing: Children begin to create marks on paper that represent people, objects, or ideas, even if these aren’t recognizable to adults.
• Gestures: Children use movements to represent actions or objects, like pretending to drink from an imaginary cup.

However, Piaget (1951) argues that language does not facilitate cognitive development, but merely reflects what the child already knows and contributes little to new knowledge.

He believed cognitive development promotes language development, not vice versa.

2. Early Forms of Animism

Animism refers to the tendency of young children to attribute life-like qualities, consciousness, or intentions to inanimate objects. As children begin to use symbols and engage in pretend play, they often start attributing life-like qualities to their toys and other objects.

For example, inanimate objects such as toys and teddy bears have human feelings and intentions. No object is truly ordinary to them.

A child might believe their stuffed animal has feelings and needs to be comforted. They might talk to toys as if they can hear and understand.

Animistic thinking often enhances pretend play, as children can more easily imagine their toys as living characters. Anything might be alive.

This, in turn, can make pretend play more complex and engaging for the child. It allows children to practice perspective-taking and empathy, even if applied to inanimate objects.

Characteristics of early animism:

• Broad application: At this stage, children might consider almost anything to be alive, especially if it moves or seems to have a purpose.
• Emotional connection: Objects that are important to the child (like favorite toys) are more likely to be perceived as having feelings or intentions.
• Lack of consistent criteria: The child’s judgments about what is “alive” are often inconsistent and based on immediate perceptions.

3. Pretend (or symbolic) Play

Pretend play involves children using their imagination to create scenarios, roles, and environments that differ from their immediate reality.

It’s a clear manifestation of the child’s growing ability to use mental representations and symbols.

Toddlers often pretend to be people they are not (e.g. superheroes, policeman), and may play these roles with props that symbolize real-life objects.

Children may also invent an imaginary playmate.

“In symbolic play, young children advance upon their cognitions about people, objects and actions and in this way construct increasingly sophisticated representations of the world” (Bornstein, 1996, p. 293).

Piaget (1945) saw make-believe as the child’s way of bending reality to fit their own wishes and schemes. In play, there is little pressure to match how things really are.

The symbol is detached from the object. So one prop can carry many meanings. A stick becomes a spoon, a horse and a sword in turn. Children also replay emotionally charged events, such as a visit to the doctor, and work through fears in a safe, self-directed way.

Play follows a reliable progression. Children move from exploring objects, to using them functionally, to symbolic and eventually planned, sequenced pretend episodes (Tamis-LeMonda & Bornstein, 1996).

The link with language runs both ways. Pretending means holding a represented meaning against a perceived reality, the same dual representation that language demands. A larger vocabulary then supplies more symbols to play with.

Characteristics of pretend play:

• Object substitution: Children use objects to represent other things (e.g., a cardboard box becomes a spaceship).
• Role-playing: Children take on different personas, often mimicking adult roles or fantasy characters.
• Imaginary scenarios: They create and act out fictional situations.
• Symbolic actions: Children perform actions that represent more complex activities (e.g., moving a toy car along the floor to represent driving).

Role in overcoming egocentrism:

While children in this stage are still largely egocentric, pretend play provides opportunities to begin seeing things from other viewpoints.

As play becomes more social, children must negotiate roles and scenarios with peers, gradually developing a more nuanced understanding of others’ perspectives.

4. Parallel Play

Parallel play refers to a stage of play where children play alongside each other, but not directly with each other. They may be in close proximity and even engaged in similar activities, but there’s little to no interaction between them. They rarely play together.

Parallel play is typically observed in toddlers, around 2-3 years old, though it can extend into the early preschool years. It allows children to observe and learn from peers without the pressure of direct social interaction. Yet each child stays in a private world.

Speech at this stage is egocentric: its main job is to externalize the child’s own thinking, not to communicate with a partner. This talk is not for others.

During parallel play, children often engage in self-talk or narrate their own activities. This self-directed speech helps develop language skills and internalize thought processes. As yet, the child has not grasped the social function of either language or rules.

Characteristics of parallel play:

• Children play independently, even when in a group setting.
• They may use similar toys or engage in similar activities.
• There’s minimal social interaction, though they may occasionally glance at or imitate each other.
• Children are more focused on their own play than on what others are doing.

Intuitive Thought substage

The intuitive thought substage, occurring between ages 4-7, is part of Piaget’s preoperational stage of cognitive development.

During this period, children rely heavily on intuition and perception to understand their world.

Children in this substage become intensely curious and ask endless “why?” questions. They reason with growing confidence.

Yet they cannot say how they know what they claim to know, because their conclusions rest on intuition and perception rather than on articulated logical rules.

So a four- or five-year-old may give a wrong answer to a conservation problem with complete certainty.

They exhibit characteristics such as egocentrism, centration, and animism.

While their thinking becomes more symbolic and language use expands, they still struggle with logical reasoning, conservation, and reversibility of actions.

1. Centration

Centration refers to the tendency of young children to focus on only one aspect of a situation or object while ignoring other potentially relevant features.

It’s the inability to consider multiple aspects simultaneously.

When a child can focus on more than one aspect of a situation, at the same time, they have the ability to decenter.

For example, children might categorize objects based on a single feature (e.g., color) while ignoring other relevant attributes (e.g., shape or size).

Alternatively, a child might judge fairness on a single factor alone. Other important considerations go unseen.

The inability to mentally reverse actions (irreversibility) is often a result of centration on the end state of a process.

During this stage, children struggle to think about more than one aspect of a situation at the same time. Social situations are no different. Decentering is just as hard there.

Key characteristics of centration:

• Single-focus attention: Children concentrate on one salient feature while neglecting others.
• Lack of reversibility: Difficulty in mentally reversing actions or considering alternative perspectives.
• Perceptual dominance: Judgments are often based on how things appear rather than logical reasoning.

Class inclusion: a second example of centration

Piaget’s class-inclusion task is a second classic test of centration. Shown five roses and three tulips, a child is asked whether there are more roses or more flowers. The preoperational child typically answers “more roses” (Piaget, 1952).

The error shows the same failure to hold two things in mind at once: the subclass (roses) and the whole class that contains it (flowers).

The wording may also understate competence. The word “flowers” is ambiguous, because it can name the whole class or just the other subclass, the tulips.

Politzer (2016) removed the ambiguity by naming both subclasses before asking the comparison. Children as young as five then succeeded. That is three to four years earlier than the standard task suggests.

As children move towards the concrete operational stage (around age 7), they gradually develop the ability to decenter – consider multiple aspects simultaneously.

2. Egocentrism

Egocentrism refers to the child’s inability to see a situation from another person’s point of view. The egocentric child assumes that other people see, hear, and feel exactly the same as the child does.

It’s not selfishness, but rather a cognitive limitation in perspective-taking.

In the developmental theory of Jean Piaget, this is a feature of the preoperational child. Children’s thoughts and communications are typically egocentric (i.e., about themselves).

For example, egocentric children have difficulty understanding why others might be upset or happy when they are not. They experience challenges in empathy as they struggle to put themselves in others’ shoes.

Piaget’s Three Mountains task shows this clearly. Children struggle to describe how a scene looks from a different vantage point.

While it’s a limitation, egocentrism also serves a purpose in cognitive development. It allows children to solidify their own understanding of the world before considering multiple perspectives.

Key characteristics of egocentrism:

• Difficulty distinguishing between their own and others’ perspectives.
• Assumption that others see, think, and feel exactly as they do.
• Challenges in understanding that others might have different information or beliefs.

The development of “theory of mind” (understanding that others have different thoughts and beliefs) is a key factor in overcoming egocentrism.

3. More Developed Forms of Animism

In the symbolic function substage (ages 2-4), animism is broad and undifferentiated. The child believes that almost everything is alive and has a purpose.

During the intuitive thought substage (ages 4-7), animistic thinking becomes more nuanced and selective. Only objects that move have a purpose.

Children at this stage often use “if-then” logic: “If it moves, then it must be alive.” Their judgments are based primarily on observable features rather than underlying biological principles.

Movement is the key clue.

For example, a car might be considered alive because it moves. A toy robot that moves might be considered alive, while a stationary doll is not. This may lead to misconceptions about natural phenomena (e.g., thinking wind is alive because it moves things).

Even the wind seems alive to them.

Stationary living things (like sleeping animals) might be mistakenly classified as non-living.

As children approach 7-8 years old, they begin to refine their criteria further, leading to the next stage where only things that move on their own are considered alive.

4. Artificialism

Artificialism is the belief that natural phenomena and objects are created by human beings or a human-like entity for a specific purpose.

Children at this stage tend to view the world as centered around human needs and experiences, leading them to attribute human-like intentions and creations to natural phenomena.

For instance, a child might believe that the sun exists solely to provide light for people, unable to conceive of a world not designed with humans in mind.

Characteristics of artificialism:

• Attribution of human creation to natural phenomena.
• Belief in purposeful design of natural objects and events.
• Tendency to see the world as existing to serve human needs.

Children’s egocentricity, coupled with their inherent finalism (the belief that everything has a purpose), leads them to view nature as serving human needs. They reason that if the sun provides warmth and the moon lights the night, both must have been made for those purposes by people.

That is their logic.

Another crucial factor of artificialism is children’s limited understanding of causality. At this stage, their grasp of cause and effect is still developing, which leads them to create simplified, often human-centred explanations for complex natural processes.

Causality still puzzles them.

Without a mature understanding of scientific principles, children fill in the gaps with explanations that make sense to them, based on their limited experience. This might result in beliefs such as rain falling because someone waters the sky, or mountains existing because giants piled up rocks.

The world always has a reason, in their eyes.

Artificialism fades as children approach the concrete operational stage, where conservation and reversibility take hold. This new flexibility lets them decenter and think more objectively about the natural world.

Transductive Reasoning

Transductive reasoning is a third way preoperational thinking departs from logic: the child links two particular events just because they happened together, without any general rule connecting them. Cause and coincidence blur together.

A child who naps every afternoon might insist it “isn’t afternoon yet” simply because they haven’t napped. Or they may reason that a cat “is” a kind of dog, since both have four legs and fur.

This same faulty logic feeds directly into animism and artificialism. It is also why young children’s causal explanations can seem so idiosyncratic.

5. Irreversibility

Irreversibility is the inability to mentally reverse a sequence of events back to its starting point.

The film cannot run backwards in the child’s mind.

So the child struggles to see that actions can be undone, or that objects can return to how they were.

For instance, a ball of clay rolled into a snake shape cannot, in the child’s mind, be rolled back into a ball.

Examples of Irreversibility:

Conservation tasks: a child might believe that when water is poured from a short, wide glass into a tall, narrow one, there is more water in the taller glass.

The glass shape fools them.

They struggle to mentally reverse the pouring action and understand that the amount of water remains the same.

Spatial reasoning: a child might have difficulty understanding that a route from point A to point B can be reversed to return the other way.

Direction confuses them too.

Logical operations: if a child sees a row of coins spread out, they might think there are more coins than when the same coins are grouped closely together.

They can’t mentally reverse the spreading action to recognize the quantity remains unchanged.

Key Characteristics of Irreversibility:

• Non-Reversal: Difficulty in mentally reversing actions or transformations.
• No Conservation: Inability to recognize the conservation of quantity, mass, or volume.
• One-Way Process: Challenges in understanding that processes can be bidirectional.

That takes time.

As children move towards the concrete operational stage (around age 7), they gradually develop the ability to mentally reverse operations.

This development is crucial for more advanced logical thinking and problem-solving skills.

6. Social Play

Between ages 4 and 7, social play changes markedly.

This period marks a transition from the more solitary and parallel play of earlier years to increasingly interactive and complex social play. 

Children begin to engage in cooperative play, where they work together towards common goals or shared imaginative scenarios. This is a significant shift from the parallel play of earlier years.

Imaginative (pretend) play becomes more elaborate. Children take on specific roles (e.g., doctor, teacher, parent) and create complex scenarios. This type of play helps develop social understanding and perspective-taking skills.

Simple games with basic rules start to emerge. While understanding and following rules can be challenging due to egocentrism, children begin to grasp the concept of structured play.

Children start negotiating roles, rules, and scenarios in their play. This process, while often fraught with disagreements, is crucial for developing social skills and understanding others’ perspectives.

Play becomes an important context for learning to manage emotions, especially when dealing with winning, losing, or not getting a desired role in pretend play.

The Three Mountains Task

Jean Piaget used the three mountains task (pictured below) to test whether children were egocentric. An egocentric child assumes everyone else sees the display just as they do.

The model shows three mountains. Each has its own feature: snow on one, a house on another, a red cross on the third.

Picking the doll’s own view meant the child was not egocentric. Picking their own view instead, the one they themselves could see, meant they were.

Evaluation: Piaget’s tasks may have underestimated what children can actually do. Several factors could explain this: complicated language, unfamiliar materials, missing context, and children misreading what the experimenter wanted.

The materials mattered.

More recent studies have tried to ask clearer questions, using situations children can relate to more easily.

Evidence Check: Does the Stage Hold Up?

Verdict: Mixed or contested.

Piaget’s classic phenomena are secure. One independent rerun supports the claim that toddlers pass a traditional false-belief question once processing demands are cut.

The claim that infants already represent false beliefs is far less secure. The best pooled evidence shows a weak, shrinking and probably biased effect. The larger preregistered reruns of anticipatory-looking tasks mostly failed.

Toddlers: Processing Demands or Missing Competence?

Level of evidence: Experiment; independently replicated.

The standard age-4 false-belief milestone may over-state when children first represent what others falsely believe. The claim is that 2½-year-olds pass the traditional verbal question once the task’s processing demands are reduced (Setoh, Scott, & Baillargeon, 2016).

The strongest evidence comes from three studies:

• Setoh et al. (2016): Toddlers given response-generation practice trials in a low-inhibition version of the traditional task performed above chance. They failed when the practice was less effective or the task’s inhibitory demands were high.
• Grosso et al. (2019): An independent rerun in a different language found correct responding in 74% of 58 children aged 33 months. A concurrent measure showed partial competence in only 15% of them and full competence in 5%.
• Wellman et al. (2001), against the artefact view: A meta-analysis of 178 studies found performance clustered by age and task factors (multiple R of .74, R² of .55). It also showed a shift from below to above chance, read by its authors as genuine conceptual change.

Setoh et al. (2016) is a within-task manipulation: the same age group passes or fails depending on the demands. That is a strong design for separating performance from competence.

Its weaknesses are that its abstract does not give the sample size, preregistration is not reported, and it was run by a team committed to the processing-demands view.

Grosso et al. (2019) counts for more as independent evidence. But N = 58 children aged 33 months is small. It is also only one rerun.

Wellman et al. (2001) is the highest rung (k = 178). It was published fifteen years before the reduced-demand procedure, so it bears on the claim only by inference.

The most trustworthy single piece of evidence is the independent rerun, because it is the only test of the specific procedure by researchers with no tie to the original.

The evidence supports a narrow version of the claim. It qualifies the broad one.

Two labs found toddlers passing the traditional question under reduced demands. So the standard task’s age of 4 is not a hard limit on what a child can represent. The effect is real under these conditions.

But the same children failed a differently designed measure (Grosso et al., 2019). Task factors alone did not erase the preschool change either (Wellman et al., 2001).

In short, children under four can pass a traditional false-belief question when demands are sharply reduced. One independent rerun at 33 months found the same. Whether this reflects early competence or a developmental change that tasks merely expose remains disputed.

Infants: Contested, Not Established

Level of evidence: Meta-analysis or systematic review; failed to replicate.

The claim is that even 15-month-old infants act as though they anticipate behaviour on the basis of an agent’s false belief (Onishi & Baillargeon, 2005). Supporters add that meta-analytic evidence outweighs the failed replications of the earliest infant studies.

The evidence is one meta-analysis and four reruns. The anticipatory-looking reruns used a related but different paradigm, tested older children, and mostly failed:

• Barone et al. (2019): A meta-analysis pooled 56 false-belief conditions from infants under two years old (1,469 infants). Correct performance was about 1.76 times more likely than incorrect performance (β = 0.57, 95% CI 0.33 to 0.80).
• Powell et al. (2018): A study with 18-month-old infants that set out to build on earlier violation-of-expectation results failed to reproduce them. Its authors discuss whether infants track others’ knowledge and ignorance without representing the contents of their beliefs.
• Schuwerk et al. (2018): An anticipatory-looking rerun did not replicate the original finding in 2- to 3-year-olds (N = 48) or 25-month-olds (N = 78). Adults (N = 115) did show the effect.
• Kampis et al. (2021): A preregistered two-lab direct replication with two original authors tested 160 children aged 24 to 26 months. It found no evidence of action anticipation.
• Kaltefleiter et al. (2022): A large longitudinal replication of a separate multi-trial paradigm followed N = 185 children tested at ages 2, 3 and 4. It failed to replicate the combined score and found mixed results across its two conditions.

In the meta-analysis, heterogeneity was high. The funnel plot was asymmetric in a way that suggests unpublished small effects.

The effect also shrank with publication year (β = -0.11, 95% CI -0.16 to -0.06), and it was larger in violation-of-expectation tasks. Its authors say this casts doubt on the phenomenon’s robustness.

The meta-analysis sits on the top rung. But high heterogeneity, an asymmetric funnel plot and a declining effect are the signatures of publication bias. The pooled estimate is probably an overestimate.

The preregistered reruns are large, but they tested older children than the 15-month-olds of Onishi and Baillargeon (2005).

Violation-of-expectation reruns are fewer, and one with 18-month-olds failed (Powell et al., 2018). That paradigm is not secure either.

A philosophical analysis of this case argues that researchers disagree over which experiments count as replications, because there is no agreed theory of what the tasks measure (Lavelle, 2022).

The most trustworthy evidence is the preregistered 160-infant two-lab replication, because original authors took part and it still found nothing. It tests one paradigm only.

The evidence qualifies the claim heavily. The meta-analysis shows an above-chance signal in aggregate. But its authors attribute doubt to heterogeneity, bias and a declining effect.

So it cannot carry a verdict of robust infant false-belief understanding on its own. The effect may be smaller than claimed, or the tasks may not measure belief.

Saying the effect is not real would be wrong. Saying it is established would be equally wrong.

In short, some infant looking studies suggest sensitivity to an agent’s false belief, but the pooled effect is modest, shrinking and probably affected by publication bias. Reruns of anticipatory-looking tasks with two-year-olds have mostly failed, so infant false-belief understanding is contested rather than established.

Critical Evaluation

Policeman Doll Study

Martin Hughes (1975) argued that the three mountains task confused children for the wrong reason. Matching the doll’s view to a photograph, he said, was simply too hard a task, not a true test of egocentrism.

Hughes devised a task which made sense to the child.

He showed children a model comprising two intersecting walls, a “boy” doll and a “policeman” doll. He then placed the policeman doll in various positions and asked the child to hide the boy doll from the policeman.

Hughes did this to make sure that the child understood what was being asked of him, so if they made mistakes they were explained and the child tried again. Interestingly, very few mistakes were made.

The experiment then began. Hughes brought in a second policeman doll, and placed both dolls at the end of two walls, as shown in the illustration above.

Most succeeded easily.

The child was asked to hide the boy from both policemen, in other words, he had to take account of two different points of view. Hughes’ sample comprised children between three and a half and five years of age, of whom 90 percent gave correct answers.

Age four was enough.

Even when he devised a more complex situation, with more walls and a third policeman, 90 percent of four-year-olds were successful. This shows that children have largely lost their egocentric thinking by four years of age, because they are able to take the view of another.

Hughes’ experiment allowed them to demonstrate this because the task made sense to the child, whereas Piaget’s did not.

The ‘Turntable’ Task

In Borke’s (1975) test of egocentrism, the child is given two identical models of a three-dimensional scene. Several different scenes were used, including toy people, animals, and a mountain model similar to Piaget and Inhelder’s.

One of the models is mounted on a turntable so it can easily be turned by the child. The turntable made this easy.

After a practice session with the materials, a doll is introduced to test perspective-taking. In Borke’s study this was Grover, the “Sesame Street” character, a figure the children already knew well.

Grover felt familiar.

The Grover doll was placed “looking” at the model from one vantage point. The child then had to turn the other model until its view matched what Grover could see.

The task felt like a game.

Borke (1975) found, using the ‘mountains’ model, three-year-olds selected a correct view 42% of the time and four-year-olds selected the right view 67% of the time. With other displays, the three-year-old accuracy increased to 80% and the four-year-olds’ to 93%.

Cultural and Material Bias

Piaget’s materials were specific to a mid-twentieth-century European childhood. That is not incidental.

Swapping the mountains for familiar displays lifted accuracy from 42% to 80% in three-year-olds and from 67% to 93% in four-year-olds (Borke, 1975). The jump came from the materials alone.

Treating one style of test as culturally neutral is risky. It can turn a mismatch between task and child into an apparent cognitive deficit. The ages Piaget reported may say as much about the fit between task and everyday world as about a universal ceiling.

Culture shapes more than materials. Dasen (1975) tested children from a remote Central Australian Aboriginal community.

They showed the same shift from preoperational to concrete operational conservation that Piaget found in Swiss children. It consolidated several years later.

On a spatial-reasoning task, the pattern reversed. The same children reached the equivalent level earlier than European comparison samples.

Dasen linked the crossover to practice. Exact quantity is rarely important where water is vital but rarely counted. Finding the way back to a waterhole is a survival skill rehearsed from early childhood.

The sequence held in both domains. Its timing, though, tracked the skills each culture demanded, not a single biological timetable.

The “Naughty Teddy” Conservation Task

Piaget’s conservation tasks assume the child cannot yet grasp that quantity stays the same. But is that fair? The standard procedure may mislead the child.

Contemporary Research

A modern line of research asks whether children grasp false belief, the understanding that someone else can believe something untrue, earlier than Piaget’s tasks suggest. In the classic version of the test (Wimmer & Perner, 1983), a character leaves an object in one spot before someone secretly moves it.

Asked where the character will look, most children under four point to the object’s real, new location. That looks like textbook egocentrism.

This single study is not an outlier. The wider evidence, though, is mixed.

A meta-analysis pooling many infant false-belief studies found a modest above-chance signal before age two, with high heterogeneity and signs of publication bias (Barone, Corradi, & Gomila, 2019).

Some infants even seem to act on another’s false belief from as young as 15 months (Onishi & Baillargeon, 2005). Reruns of anticipatory-looking tasks with two-year-olds have mostly failed. The infant evidence stays contested.

Scott and Baillargeon (2017) read the evidence as a staged unmasking, not a sudden arrival at four. On this view, the capacity for false belief is present early, and executive function gradually makes it easier to express. That account remains disputed.

Real Limit or Task Artefact?

Weighing the evidence by quality gives a verdict that neither vindicates nor dismisses Piaget. The phenomena are real. The ages and the all-or-nothing framing are not.

Children really do fail the standard three mountains and conservation tasks at the ages Piaget reported, and the failures are systematic. What the challenge studies overturn is his reading of those failures as a global incapacity. The age of success varies with the task:

• Three mountains: Perspective-taking appears around age 7 on the unmodified task.
• Hiding and turntable tasks: It appears at around 3½ to 4 (Hughes, 1975; Borke, 1975).
• Reduced-demand false belief: Success appears as early as 2½ (Setoh, Scott, & Baillargeon, 2016).

A single ability cannot switch on at three different ages. A continuous competence, masked to different degrees by each task’s incidental demands, fits the pattern better.

The reduced-demand studies are the most probative. They hold the concept constant and vary only the demands.

The same children pass or fail depending on whether a tempting wrong answer must be inhibited or an extra memory load is added. That isolates performance factors more cleanly than comparing age groups on one hard task.

Two caveats stop this becoming an over-correction.

Hiding a doll from a pursuer, which needs only knowing whether another can see something, is easier than working out how a scene looks from another angle. Some of the age gap is therefore real task difficulty.

Conservation has also proved harder to rescue than egocentrism. Even with the pragmatic cues removed, full and justified conservation across number, mass, weight and volume still consolidates in the concrete operational years (McGarrigle & Donaldson, 1974).

The stage boundary is softer and earlier than Piaget claimed. It is not illusory.

Preoperational thought looks less like missing logic than like real, emerging competences that processing demands easily swamp. Whether infants already have such competences is still contested, as the Evidence Check above shows.

Limitations in the Child’s Thinking

Piaget focused much of this stage on the child’s limitations: mental tasks children seem unable to do. These include decentering, conserving, understanding seriation (arranging objects in a logical order), and handling inclusion tasks.

Take centration first. A preoperational child can only focus on one aspect of a problem at a time.

Suppose two rows of blocks are arranged so a row of 5 looks longer than a row of 7. The child can count both rows correctly.

But asked which row has more, they point to the longer-looking one, unable to hold length and number in mind together. This failure to decenter feeds directly into the child’s egocentrism.

Examples of Preoperational Thinking

Six everyday examples show how these limitations look in practice:

• Hiding by covering the eyes: A child “hides” in hide-and-seek by covering their own eyes. If I cannot see you, you cannot see me.
• Pretend play: A banana becomes a telephone, and a cardboard box becomes a spaceship.
• Animism: A child comforts a “hurt” table that was bumped into, or believes the moon follows them home.
• Artificialism: The sun was “lit” to give us light, or lakes were “dug” so people could swim.
• Failing to conserve: A tall, thin glass is judged to hold “more” juice than the short, wide glass it was poured from.
• Class inclusion: Shown five roses and three tulips, a child says there are more roses than flowers.

Key Study: The Conservation Tasks

Conservation is the understanding that a quantity stays the same when its appearance changes, as long as nothing is added or removed. Piaget tested this with several classic tasks.

That is the mark of the next stage.

Solving these problems signals the transition to the next stage. So what do conservation tasks tell us about preoperational thought in general?

Piaget drew four related conclusions:

1. Perception-Bound: The child is drawn by changes in appearance to conclude that something has changed.
2. Centred: Children notice a change in one aspect (like water level) without noticing that other aspects changed too.
3. State-Focused: Children judge based on how things look now, not on what happened to get there.
4. Irreversible: The child cannot mentally undo a change to see that the original state would return.

Rival Theories

Piaget’s account of the preoperational child is not the only serious explanation of these same limitations. Two rival frameworks reinterpret the same evidence in different ways.

Core-Knowledge Nativism

Core-knowledge nativism argues that infants are born with a small set of specialised systems, for objects, number, space and other minds. Logic need not be constructed at all.

Long before they can count, for example, infants already track roughly how many objects are in a set. They also notice when one goes missing, exactly the early, specialised skill nativism expects (Spelke & Kinzler, 2007).

Nothing needs to be learned first.

On this view, the challenge studies above are not surprises but exactly what an early-maturing “other minds” system predicts. False belief appears so early, and survives such heavy demand-reduction, because it belongs to a system present from infancy.

It does not wait for Piaget’s operational stage to arrive (Baillargeon, Scott, & Bian, 2016).

The capacity was there all along.

The two theories agree on what children can do once demands are lowered. They disagree on why: a slowly built general logic, or a specialised system that was there all along.

Piaget saw an absent structure; nativists, a fragile one.

Vygotsky’s Sociocultural Theory

Vygotsky’s sociocultural theory challenges a different part of Piaget’s picture: the child’s private, muttering “egocentric speech” during pretend play. Piaget read this running commentary as proof of a self-centred mind, unable yet to speak for another’s benefit.

Vygotsky (1962) disagreed. He reinterpreted the same speech as a genuinely useful tool the child uses to plan and regulate its own actions.

Far from vanishing, it simply turns inward, becoming silent inner speech as the child develops. So where Piaget treats egocentrism as a flaw development corrects, Vygotsky treats it as a tool the child is already putting to use.

The tool stays useful.

Private speech does not vanish outright with age; it grows quieter and more fragmented, exactly the trajectory Vygotsky’s account predicts and Piaget’s does not. The evidence favours Vygotsky here.

This social emphasis explains a pattern running through this article’s challenge studies. Children do best on Hughes’s, Borke’s and the Naughty Teddy tasks when a knowledgeable adult frames the task helpfully.

Neither rival theory claims Piaget’s observations were wrong. Both agree the child’s early logic is more capable than his tasks revealed; they differ only on where that capability comes from.

Frequently Asked Questions