The Mean Monkey Experiment (Deception) – Joan Peskin

The study of developmental social cognition fundamentally transformed during the late twentieth century as researchers shifted their focus from general Piagetian stages of intellectual development toward modular, domain-specific conceptual systems. At the vanguard of this epistemological pivot stood Theory of Mind (ToM)—the human capacity to impute unobservable mental states, such as beliefs, desires, knowledge, pretends, and intentions, to oneself and others, and to predict behavior based on those imputed states. While initial experimental paradigms focused predominantly on whether young children could passively detect false beliefs through abstract, verbally saturated narratives, developmental psychologist Joan Peskin recognized a critical empirical chasm. Standard clinical narratives often failed to capture how children strategically mobilize mental state understanding within dynamic, ecologically authentic, and competitive social exchanges where tangible personal stakes dictate behavioral outcomes.

In her groundbreaking 1992 investigation, “Ruse and representations: On children’s ability to conceal information,” published in Developmental Psychology, Joan Peskin introduced what is now colloquially and universally celebrated in developmental science as the “Mean Monkey” experiment. By engineering an interactive, game-like setting involving an antagonistic puppet agent programmed to relentlessly intercept and seize a child’s desired rewards, Peskin constructed a powerful behavioral crucible. The paradigm removed the heavy linguistic scaffolding of traditional transfer tests, directly challenging the child to either execute tactical deception by misrepresenting their internal desires or face the repetitive, exasperating forfeiture of their most coveted possessions. In doing so, the experiment untangled raw verbal ability from functional, strategic mentalizing.

The resulting empirical insights revealed a stark, profound developmental watershed between the ages of three and five. Younger children exhibited a persistent, counterproductive honesty, systematically directing their rival to their prized targets despite suffering constant, predictable losses. Conversely, older preschool children readily deployed deliberate, strategic falsehoods to systematically manipulate the puppet’s epistemic state, steering the antagonist toward disposable distractors and securing their personal prizes. Over the ensuing three decades, the Mean Monkey experiment has become not only an indelible pedagogical cornerstone across cognitive psychology curricula, but also an indispensable paradigm for investigating executive functioning, social communicative atypicalities such as autism spectrum conditions, phylogenetic divergence across primates, and the profound neurocognitive architecture that allows the human mind to construct counterfactual realities.

1. Introduction to Joan Peskin’s Seminal Deception Research

1.1 Historical Context of Theory of Mind Research in the Early 1990s

The cognitive developmental landscape of the late 1980s and early 1990s was profoundly shaped by the conceptual foundations laid by David Premack and Guy Woodruff in their pioneering 1978 paper, “Does the chimpanzee have a theory of mind?” Their central question—whether non-human primates could interpret the behaviors of conspecifics and human actors by attributing internal representational states—radically recalibrated developmental psychology. It challenged researchers to determine the precise ontogenetic point at which human children transition from naive realists, who equate subjective perception with objective external reality, to sophisticated mentalists who recognize that behavior is mediated by internal, potentially fallible representations of the world.

This theoretical momentum crystallized methodologically in 1983 when Heinz Wimmer and Josef Perner developed the classic “Maxi task,” an unexpected transfer paradigm designed to systematically assess false-belief attribution. In this foundational setup, young participants were told an illustrated story: Maxi places chocolate in a kitchen cupboard and exits the room; during his absence, his mother relocates the chocolate to a different drawer; Maxi returns, and the child participant is asked where Maxi will look for his treat. The empirical consensus that rapidly hardened from this and similar paradigms—such as the “Sally-Anne” task operationalized by Simon Baron-Cohen, Alan M. Leslie, and Uta Frith in 1985—postulated that an irrevocable cognitive shift takes place around four years of age. Prior to this boundary, three-year-olds universally predicted that Maxi or Sally would search in the location where the object actually rested in physical reality, apparently blind to the agent’s outdated epistemic state.

Despite the revolutionary impact of these unexpected transfer tasks, a vigorous methodological critique emerged within the cognitive developmental community. Critics argued that the classic Maxi and Sally-Anne paradigms suffered from an over-reliance on complex, narrative-heavy linguistic structures and passive spectator formats. The young child sat as an emotionally detached third-party observer, required to decode convoluted syntax, track temporal transitions through verbal descriptions, and answer abstract questions posed by authoritative adult examiners. These verbal and contextual demands raised an urgent question: Were three-year-old children truly experiencing a fundamental conceptual deficit regarding mental states, or were their mentalizing capacities masked by the severe cognitive, linguistic, and conversational processing loads inherent to these artificial testing protocols?

Recognizing the limitations of passive, storybook-based assessments, developmental researchers began searching for interactive, ecologically valid behavioural paradigms that could tap into early social cognition without burdening immature linguistic processing channels. There was a critical need for an experimental setting in which children were not merely reporting on the hypothetical misfortunes of fictional characters, but were active agents within dynamic social ecologies featuring visceral incentives, competitive pressures, and live behavioral choices. It was precisely into this methodological impasse that Joan Peskin introduced her seminal work, seeking to uncover whether children who routinely failed verbally mediated, spectator-based false-belief tasks possessed latent, operational mentalizing abilities when placed in an emotionally charged, real-time interpersonal conflict.

1.2 Conceptualizing Deception as an Epistemic Milestone

To understand the profound diagnostic significance of Peskin’s experimental design, one must trace the vital theoretical distinction between primitive, reflexive concealment and genuine tactical deception. In both ethology and early child development, superficial concealment can frequently occur without any sophisticated understanding of subjective mental representations. A toddler who reflexively hides an illicit object behind their back upon hearing a caregiver’s footsteps, or a young animal that freezes in the presence of an apex predator, may be executing an evolutionarily conserved, conditioned behavioral script or an automatic perceptual-blocking heuristic: “If they cannot see the item, danger is neutralized.” This form of physical obstruction or avoidance operates purely at the level of perceptual lines of sight and stimulus-response reinforcement contingencies.

Tactical deception, by sharp contrast, represents a monumental epistemic leap. It requires the agent to move beyond merely manipulating physical space or visual access to actively engineering, implanting, or preserving a false mental representation inside the consciousness of another intentional agent. In tactical deception, the deceiver recognizes that human behavior is not dictated directly by the objective physical environment, but rather by the target’s internal representational model of that environment. Therefore, to orchestrate another individual’s motor behavior toward a desired end, one must intervene upstream within their internal cognitive architecture—sowing misinformation, encouraging erroneous inferences, or systematically withholding critical updates about the state of the world.

This operational transition from the passive attribution of mental states to proactive, calculated epistemic interventions holds immense theoretical import. In a standard unexpected transfer task, a child merely observes an epistemic error that has unfolded independently through environmental happenstance (e.g., an object was moved while someone was outside). In deliberate deception, however, the child must personally assume the role of the epistemic architect. They must conceive of a counterfactual state of affairs, compute the divergence between actual reality and the target’s impending mental model, and deploy a communicative act specifically designed to construct a non-veridical belief in the target’s mind. The deceptive act thus serves as a window into the child’s explicit grasp of the subjective, constructivist nature of mental life.

Furthermore, evaluating deception provides an empirical battleground for resolving the classic debate between conceptual deficit models and executive performance limitation models of early childhood cognition. Conceptual deficit theorists argued that children younger than four years simply lack the cognitive representational code necessary to represent representational states as representations (the meta-representational deficit model). Conversely, performance-limitation theorists maintained that very young children possess implicit core mentalizing competencies, but routinely fail standard false-belief tasks due to extraneous performance bottlenecks—such as the inability to suppress prepotent responses, fragile verbal comprehension, or task pragmatics. By creating a task where competitive motivation was intense and the behavioral choices were grounded in concrete physical actions (such as pointing), Peskin sought to establish whether the barriers preventing successful deception in young preschoolers were conceptual or executive in nature.

1.3 The Seminal 1992 Publication and Its Core Inquiries

In her landmark 1992 paper, Joan Peskin addressed these foundational questions with an empirical design of elegant simplicity and rigorous diagnostic precision. The overarching objective of the study was to systematically chart the chronological emergence of deceptive intent, tracking how young children maneuver when their desires are placed in direct opposition to an antagonistic partner. Rather than presenting hypothetical narratives, Peskin placed children in an active, repeated game where the preservation of tangible, highly attractive rewards depended entirely on their capacity to regulate the flow of information transmitted to a competitive rival.

Peskin’s theoretical inquiry centered on three core questions:

• First, at what precise developmental juncture do young children spontaneously transition from veridical, honest communication to strategic, misleading communicative acts when competing for desired resources?
• Second, can children who do not spontaneously invent deceptive strategies nevertheless execute them when provided with explicit instructional scaffolding, or does an underlying conceptual barrier render tactical misdirection impossible regardless of external coaching?
• Third, how do children resolve the profound cognitive conflict between the physical pull of an object of intense desire and the strategic imperative to divert an opponent’s attention away from that exact object?

To systematically untangle these questions, Peskin engineered a critical methodological innovation: the deployment of a competitive peer-surrogate puppet agent, immortalized across the developmental literature as the “Mean Monkey.” By utilizing a hand puppet manipulated by an experimenter rather than an adult human confederate, Peskin achieved several crucial experimental controls simultaneously. The puppet could be imbued with an overt, unapologetically selfish personality without triggering the social deference, compliance pressures, or intimidation young children routinely experience when interacting directly with adult authority figures. Furthermore, the puppet’s behavior could be governed by unyielding, predictable, algorithmic rules: it would consistently, explicitly, and remorselessly ask the child which prize they preferred, and promptly take that prize for itself, leaving the child with whatever remained.

This design created an intense micro-ecology of competitive conflict. Peskin’s participant cohorts, spanning precisely categorized age groups from late three-year-olds to five-year-olds, were placed in a high-stakes, iterative interactive environment. By tracking their behavioral selections, pointing gestures, verbal justifications, response latencies, and micro-affective reactions across successive trials, Peskin isolated the precise mechanisms that dictate the transition from cognitive transparency to tactical dissimulation, forever redefining our understanding of early preschool cognitive competence.

2. Theoretical Frameworks Underpinning the Mean Monkey Paradigm

2.1 Theory of Mind and Representational Capacities

The developmental trajectory that underpins the Mean Monkey paradigm is inextricably bound to the transition from perceptual direct-realism to representational mental modeling. During infancy and very early toddlerhood, a child operates largely within a direct perception-action framework. Objects are acted upon directly based on perceptual salience; external reality is treated as an unmediated absolute. While older infants clearly comprehend visual perspective taking at an elementary level—recognizing, for example, that an obstacle physically occludes another person’s line of sight (Level 1 perspective-taking, as delineated by John Flavell)—this ability is distinctly separate from understanding *epistemic* states. Knowing is fundamentally different from seeing; it requires the cognitive agent to realize that visual inputs are translated internally into subjective cognitive structures called beliefs.

To engage in intentional, strategic deception within Peskin’s experimental paradigm, a child must possess the meta-representational infrastructure required to decouple an internal belief from objective reality, a cognitive architecture deeply explored by Alan M. Leslie (1987). The child must form an internal primary representation of the physical state of affairs: “Sticker A is the shiny, desirable dinosaur; Sticker B is an unappealing, plain geometric circle.” Concurrently, to construct a deceptive ruse, the child must generate a secondary, counterfactual representation: “I will indicate that Sticker B is my target.” Crucially, the child must then project this secondary representation into the mental landscape of the opponent, computing the meta-proposition: “The monkey will believe that I desire Sticker B, and because his goal is to deprive me of my desires, his internal false belief will causally drive his motor system to seize Sticker B, thereby preserving Sticker A for myself.”

This representational cascade demonstrates why deceptive actions are such powerful diagnostics of explicit mental state grasp. A child cannot rely on simple reality-tracking heuristics to solve this puzzle. If the child’s cognitive apparatus is restricted to reflecting the direct state of the world, their communication will reflexively mirror their genuine internal desire. To actively mislead, the child must treat the puppet’s mind not as a passive mirror of objective truths, but as an interpretive, representational engine that constructs fallible beliefs based entirely on the perceptual and informational inputs it receives. The child must manipulate the input to intentionally poison the downstream output.

2.2 The Functional Architecture of Tactical Concealment

Within the theoretical framework of tactical communication, a foundational distinction exists between the act of *withholding veridical information* and the act of *disseminating fraudulent information*. These two strategies impose profoundly different computational burdens on the developing central executive. Passive concealment, or information withholding, involves maintaining silence, suppressing verbalization, or physically hiding a target. Semiotically, it is an act of omission. The child recognizes that without informational access, the rival agent remains in an epistemic state of ignorance or uncertainty. While withholding does require inhibitory control to suppress the natural communicative drive, it does not mandate the complex construction of an alternate, counterfactual narrative.

Active tactical deception, conversely, is an act of commission. It relies on misdirection, fabricated declarations, or deliberately misleading gestures. The deceiver does not simply aim to leave the opponent in an uncommitted state of ignorance; rather, they seek to actively instill a specific, actionable, yet veridically false belief. In Peskin’s Mean Monkey setup, passive withholding was structurally curtailed by the social and procedural architecture of the game: the monkey actively interrogated the child, demanding an overt choice (“Tell me which one you want, and I’ll take it!”). The child was thus confronted with an explicit communicative challenge that heavily penalized truth-telling, rendering mere silence awkward or non-viable, and requiring an active fabrication to optimize their physical payoff.

The computational burden of this active fabrication is exceptionally high for a preschool child. The child must calculate the opponent’s prior expectations, anticipate how the opponent will interpret a communicative gesture (such as an indexical point), and evaluate whether the opponent regards them as an honest communicator or a rival. In evolutionary game theory, competitive payoffs serve as the primary fuel accelerating the adoption of tactical reasoning. The immediate prospect of reward forfeiture (losing a prized sticker) creates intense emotional and cognitive pressure. However, to translate this motivational pressure into a successful outcome, the child’s functional architecture must simultaneously manage competitive desire, predict the opponent’s counter-strategy, and deploy a calibrated semiotic signal that inverts their true preference.

2.3 Evolutionary and Comparative Developmental Perspectives

The dynamics captured in Peskin’s paradigm directly mirror broad evolutionary frameworks regarding the emergence of primate intelligence. Foremost among these is the Machiavellian Intelligence Hypothesis, conceptualized by Richard Byrne and Andrew Whiten in their foundational 1988 volume. Byrne and Whiten posited that the extraordinary expansion of the primate neocortex was driven not primarily by the demands of navigating physical topographies, tool manufacture, or foraging logistics, but by the relentless computational arms race of social competition within complex group hierarchies. In these social landscapes, individuals who could deploy tactical deception—misleading competitors regarding the location of hidden caches of high-value food, concealing sexual consortships from alpha males, or feigning disinterest in contested resources—garnered distinct reproductive and survival advantages.

Comparative cognitive research has long demonstrated that non-human primates, particularly chimpanzees (Pan troglodytes), exhibit sophisticated forms of competitive tactical deception while performing notoriously poorly on cooperative, linguistically mediated communicative tasks. Research by Michael Tomasello and his colleagues at the Max Planck Institute for Evolutionary Anthropology has emphasized that chimpanzee social cognition is fundamentally optimized for competitive contexts. A subordinate chimpanzee, for example, instinctively understands that a dominant rival cannot see a food item obscured by an opaque barrier, and will systematically choose to retrieve only the hidden food, thereby demonstrating an operational grasp of another’s visual perspective within competitive paradigms.

In human ontogeny, however, cooperative communication emerges exceptionally early through mechanisms of shared intentionality, joint attention, and mutual helpfulness. Human infants as young as twelve months point cooperatively to inform adults of missing items or to share interest in novel phenomena. Because the human cognitive baseline is so heavily biased toward cooperative, veridical information transmission, the requirement to deploy deceptive communicative acts introduces a unique cognitive dissonance. Peskin’s paradigm bridged primate tactical deception studies and human developmental protocols by stripping away the normative, cooperative scaffolding of typical adult-child interactions and placing human children in an explicitly adversarial, competitive arena reminiscent of primate dominance-competition paradigms, illuminating the exact age at which human ontogeny integrates tactical dissimulation into its social repertoire.

3. Methodological Architecture of the Mean Monkey Experiment

3.1 Experimental Design and Participant Demographics

Joan Peskin designed her experimental protocol to rigorously parse developmental competencies by cross-sectionally stratifying cohorts across critical, highly sensitive preschool age bands. The primary investigation isolated three distinct chronological brackets:

• Young three-year-olds (ranging from approximately 3 years, 0 months to 3 years, 6 months),
• Older three-year-olds to younger four-year-olds (3 years, 7 months to 4 years, 5 months), and
• Older four-year-olds to five-year-olds (4 years, 6 months to 5 years, 5 months).

This fine-grained stratification was vital because standard developmental literature had identified the four-year mark as the broad watershed for passing verbal false-belief tasks; granular sampling was required to catch the precise transitional mechanics occurring immediately prior to and following this threshold.

To eliminate confounding variables, rigorous exclusionary and inclusionary controls were established. Peskin controlled for baseline receptive language skills and general verbal intelligence, ensuring that task failures could not be trivially attributed to a participant’s inability to process syntactic structures or comprehend task instructions. Before entering the competitive paradigm, children underwent a structured pre-screening familiarization phase with the human experimenter. This protocol was designed to normalize the testing environment, alleviate stranger anxiety, build rapport, and ensure that the children felt comfortable actively communicating, pointing, and expressing preferences within the room.

The experimental trials utilized systematic counterbalancing procedures. The spatial positioning of physical stimuli (left versus right presentation) was randomized and counterbalanced across trials to eliminate motor perseveration or lateral spatial biases. The identity and appearance of the puppet were maintained with strict consistency across participants. Confederate behaviors were tightly scripted, standardizing vocal intonation, response timing, and physical movements to prevent unintentional social referencing cues or subtle adult leakage that might artificially tip off the child to the strategic nature of the game.

3.2 Apparatus, Materials, and the Persona of the Mean Monkey

The physical apparatus of the experiment was deliberately engineered to be simple, transparent, and intensely engaging for preschool-aged children. The central incentive materials comprised an array of carefully pre-tested, visually striking, highly desirable stickers. These included vibrant, multi-colored depictions of animals, popular animated characters, and sparkling designs that carried immediate, powerful affective currency for young children. Concurrently, an inventory of intentionally unappealing, drab stimuli was prepared, consisting of plain, monochromatic shapes or mundane geometric cutouts that possessed minimal hedonic value.

The antagonist of the experiment was an expressive hand puppet christened the “Mean Monkey.” The physical construction of the puppet was selected to embody a slightly mischievous, assertive persona without being genuinely frightening or traumatic to a three-year-old child. Crucially, the puppet was endowed with an unwavering, explicit behavioral algorithm, enacted by the experimenter with meticulous consistency. Upon appearing, the Mean Monkey was introduced through an explicit operational rule: “This is the Mean Monkey. Whenever you tell him or show him which sticker you really want, he takes that sticker for himself, because he is mean and wants all the best things. You only get to keep the sticker that the Mean Monkey leaves behind.”

The physical layout of the testing space was arranged to maximize visual transparency and eliminate physical barriers. The child sat directly across a small table from the experimenter, who operated the puppet on one hand while managing task stimuli with the other. Two stickers—one exceptionally desirable, one thoroughly mundane—were placed flat on the table, equidistant from the child and clearly visible to both the child and the puppet. There were no occluding screens, hidden compartments, or deceptive visual illusions. The entire game was played in plain sight, ensuring that any strategic success or failure could be mapped directly to the child’s informational manipulation, rather than their misunderstanding of physical mechanics.

3.3 Trial Sequencing and Procedural Protocol

The experimental sequence unfolded through a structured, multi-phase architecture designed to systematically observe baseline preferences, familiarization, iterative learning, and the response to strategic scaffolding:

Phase 1: Baseline Preference Establishment. Before the Mean Monkey puppet made its appearance, the experimenter presented the child with two distinct stickers (e.g., a sparkling dinosaur versus an unadorned gray square) and asked: “Which of these two stickers do you like the best? Which one do you really, really want to take home?” This step confirmed the child’s explicit, unequivocal hierarchy of desire, establishing an incontrovertible target item for the trial.

Phase 2: The Antagonist Familiarization Rounds. The Mean Monkey puppet was introduced onto the table. The puppet declared its hostile, competitive intent in clear, child-directed speech: “I love stickers! Tell me which one you want, and I’m going to take it right now!” In these initial familiarization trials, the child was observed in their spontaneous, unprompted communicative interaction. The child indicated their preferred sticker (invariably the sparkling dinosaur on round one). The puppet immediately intervened, gloating playfully yet assertively: “Ha ha! You want that one? Then that one is MINE!” The puppet dramatically seized the preferred sticker and deposited it into its own private treasure chest, leaving the child with the drab, unappealing alternative. This procedure was repeated to ensure the child fully internalized the puppet’s immutable behavioral rule: whatever the child pointed to or named, the puppet definitively confiscated.

Phase 3: Repeated Test Trials and Iterative Learning. The child was then subjected to a battery of repeated test trials (typically spanning 4 to 6 consecutive iterations) utilizing new, visually distinct pairs of stickers on each trial. Each round began identically: the child established their true internal preference with the experimenter prior to the monkey’s turn, the monkey re-entered the scene, and the monkey issued its predictable, interrogative demand: “Which one do you want me NOT to take? Which one is your favorite?” These repeated iterations were essential to evaluate whether children could independently invent a deceptive strategy over time through the iterative experience of loss, or whether they would display perseverative behavioral rigidity.

Phase 4: Intervention Prompts and Strategic Scaffolding. For children who repeatedly and consistently failed to safeguard their preferred sticker across the baseline and repeated test trials (a pattern overwhelmingly characteristic of younger cohorts), the protocol integrated standardized, systematically tiered intervention prompts. These prompts were calibrated to measure whether strategic hints could unlock latent, dormant deceptive competence. The scaffolding unfolded along a spectrum of increasing explicitness:

• Level 1 (Indirect Affective Prompt): “Remember, the monkey is mean and always takes the one you say you like. How can you get the one you really want?”
• Level 2 (Counterfactual Behavioral Prompt): “What would happen if you told him you liked the other sticker instead?”
• Level 3 (Explicit Strategic Coaching): “You can trick the monkey! If you point to the ugly sticker, the monkey will take the ugly one, and you will get to keep the beautiful one!”

By recording the exact prompt level required to elicit a deceptive response—and tracking whether that deceptive response generalized to subsequent unprompted trials—Peskin established a nuanced metric separating an absolute conceptual deficit from a fragile performance limitation susceptible to targeted scaffolding.

4. Age-Related Developmental Trajectories in Performance

4.1 The Behavioral Profile of Three-Year-Old Participants

The behavioral patterns exhibited by the three-year-old participants in Joan Peskin’s paradigm provided some of the most striking and counterintuitive footage in modern developmental science. Overwhelmingly, three-year-old children demonstrated a profound, persistent, and seemingly unbreakable pattern of counterproductive honesty. Trial after trial, despite watching the Mean Monkey greedily confiscate their favorite sticker, these young participants would once again look directly at the monkey, point their finger straight at the newly presented, highly coveted sticker, and honestly declare: “I want that one!”

This phenomenon, termed in cognitive psychology as knowledge leakage or veridical transparency, manifested as an involuntary, reflexive disclosure of authentic subjective desire. The children appeared cognitively powerless to disconnect their internal motivational state from their external communicative output. When the monkey predictably swept in, seized the desired sticker, and gloated, the children displayed genuine, acute affective distress. They frowned, pouted, slumped their shoulders, and occasionally voiced vehement protests: “Hey! You took my sticker again!” Yet, shockingly, when the very next trial commenced seconds later with a brand-new set of stickers, the child would repeat the exact same self-defeating behavior with mechanical consistency.

Crucially, this failure was not due to an inability to understand the puppet’s behavioral rule. When the experimenter questioned the three-year-olds offline—asking, “When you tell the monkey which sticker you want, what does he always do?”—the children readily and accurately replied, “He takes it.” They fully grasped the empirical association on an observational level: Pointing to Item X leads to Item X being taken. What they were entirely incapable of doing was weaponizing that association into a reverse communicative strategy. Even more revealing was their profound resistance to indirect verbal scaffolding. When the experimenter gently asked, “Could you tell him something different so you can keep the good one?”, three-year-olds frequently stared with incomprehension or flatly asserted: “No, because I want the dinosaur!” For these children, pointing to an item was inextricably linked to the metaphysical truth of wanting it; executing a deceptive point was not merely difficult—it was conceptually incoherent.

4.2 The Transitional Shift in Four-Year-Old Cohorts

In the four-year-old cohort, Peskin observed a dynamic, volatile, and deeply revealing transitional phase. Unlike the rigid, perseverative honesty of their younger peers, four-year-old children occupied a cognitive middle ground, displaying fluctuating behavioral patterns that captured the very moment of developmental emergence. Within this age bracket, children frequently exhibited sudden, spontaneous conceptual breakthroughs after experiencing one or two initial losses to the puppet.

During the opening rounds, a four-year-old might reflexively point to their true preference, suffer the familiar confiscation, and register the immediate sting of loss. However, on the subsequent round, a visible internal conflict would manifest. Researchers observed marked changes in response latencies: where three-year-olds pointed impulsively within hundreds of milliseconds, four-year-olds hesitated. They would look at the stickers, look up at the Mean Monkey, pause, perhaps bite their lip or hover their hand mid-air, and then, with a mix of trepidation and emerging cunning, point decisively to the *drab, unwanted* sticker. When the Mean Monkey promptly snatched the unappealing distractor, the four-year-olds’ faces illuminated with triumphant realization. They had cracked the code.

Yet, this newly minted competence was characterized by notable fragility. Four-year-olds demonstrated pronounced behavioral instability, often fluctuating between successful deceptive strategies and sudden regressions back to truthful reporting across consecutive trials. A child might brilliantly trick the monkey on Trial 2, only to impulsively revert to pointing at their true desire on Trial 3, particularly if the sticker presented on Trial 3 was exceptionally dazzling or emotionally salient. This back-and-forth illustrates that while the conceptual meta-representation (“I can make him believe I want the bad one”) had been unlocked, the executive inhibitory machinery required to consistently suppress the prepotent pull of the target remained vulnerable to cognitive overload and emotional interference.

4.3 Mastery and Strategic Nuance in Five-Year-Old Children

By age five, the behavioral landscape shifted from effortful, intermittent struggle to absolute, effortless mastery. Five-year-old children engaged with the Mean Monkey paradigm not as a baffling emotional trauma, but as an entertaining strategic game. The counterproductive honesty that dominated the three-year-old demographic was virtually non-existent. From the very first or second trial, five-year-olds seamlessly and spontaneously deployed deceptive signaling without requiring a shred of experimenter intervention or scaffolding.

The behavioral signatures of five-year-olds were defined by anticipatory amusement and supreme cognitive readiness. Even before the Mean Monkey had completed his standardized opening threat, five-year-olds could frequently be observed suppressing smiles, exchanging knowing glances with the human examiner, and deliberately, smoothly gesturing toward the inferior item. They relished the puppet’s predictable error. When the puppet fell into the trap, seizing the worthless sticker and boasting of its triumph, the five-year-olds would erupt into authentic laughter, swiftly scooping up the prized sticker they had cleverly shielded from view.

Furthermore, five-year-olds demonstrated an advanced, fully articulated meta-cognitive awareness of their own actions. When asked by the experimenter to justify their choices (“Why did you tell the monkey you wanted the plain circle?”), five-year-olds provided clear, counterfactual, representational explanations: “Because he’s a mean stealer! I tricked him so he would think I loved the circle, and then he took it, and that meant the shiny star was left for me!” Moreover, if the puppet introduced minor secondary challenges—such as hesitating and asking, “Are you really sure you want this one?”—five-year-olds did not waver or confess; they double-downed on their fabricated claims, maintaining their deceptive facade with unwavering pragmatic coherence.

5. Cognitive Mechanisms: Executive Function and Inhibitory Control

5.1 Inhibition of Prepotent Responses

The dramatic age-related divergence observed in the Mean Monkey experiment cannot be understood solely through the lens of Theory of Mind concepts; it is intimately linked to the maturation of Executive Functions (EF), most critically the capacity for inhibitory control. In early childhood, the human sensorimotor system is heavily biased toward approach behaviors directed at affectively rewarding environmental stimuli. When a three-year-old sees an exceptionally attractive object (a sparkling, colorful sticker), that visual stimulus exerts a formidable, prepotent “magnetic pull.” The automatic, default motor program of the human hand is to reach or point toward the precise locus of desire.

To successfully execute the deceptive maneuver in Peskin’s task, the child must perform a series of sophisticated cognitive operations against this baseline motor bias:

• First, they must completely abort, suppress, and inhibit the prepotent impulse to point toward the authentic object of desire.
• Second, they must overcome the deeply ingrained social training that commands children to speak the truth when interrogated by an adult-like entity.
• Third, they must voluntarily program and execute an unnatural, counter-intuitive motor command: pointing directly toward an object they thoroughly despise.

Empirical investigations corroborating Peskin’s work have repeatedly found powerful, statistically robust correlations between a child’s performance on the Mean Monkey task and their scores on classic neuropsychological batteries of inhibitory control. Children who excel at Aleksandr Luria’s hand game (which requires participants to imitate an examiner by tapping a fist when a finger is shown, and tapping a finger when a fist is shown) or the classic Day-Night Stroop-like task (requiring children to say “day” upon viewing a black card featuring stars, and “night” upon viewing a bright sun) systematically outperform their peers on the Mean Monkey paradigm. However, theoretical consensus emphasizes that while robust inhibitory control is an *indispensable structural prerequisite* for deceptive execution, it cannot act as a complete substitute for Theory of Mind. Without the conceptual insight that an external agent’s beliefs can be manipulated via misleading inputs, the most impeccably inhibited motor system has no strategic blueprint to execute.

5.2 Working Memory Load in Real-Time Counterfactual Manipulation

Beyond motor inhibition, the tactical landscape of the Mean Monkey experiment imposes a substantial load on the preschooler’s central executive and working memory architecture. The task requires the real-time, simultaneous maintenance and active manipulation of multiple disparate cognitive files. The child’s central executive must run what can be conceptualized as a “dual-track” cognitive program.

On Track 1, the child must continuously hold their authentic, subjective internal preference securely in working memory (“I desire Sticker A”). If this internal representation fades or becomes corrupted by task noise, the entire strategic objective dissolves. Concurrently, on Track 2, the child must generate and maintain a completely distinct, fabricated counterfactual reality (“I will communicate that I desire Sticker B”). But the computational burden does not terminate there; the child must cross-reference Track 1 and Track 2 with an internal simulation of the Mean Monkey’s behavioral heuristic: IF Child indicates X -> THEN Monkey confiscates X -> THEREFORE, Child receives Y.

This dynamic calculation requires the continuous updating of cognitive workspaces while managing the intense affective arousal triggered by the threat of losing an attractive prize. In studies evaluating the decay of deceptive performance across extended trial blocks, developmental psychologists have observed that younger children frequently experience rapid cognitive fatigue. A child may successfully marshal the working memory bandwidth to deceive the monkey on one or two trials, but as the session lengthens, the central executive becomes depleted. This depletion often precipitates sudden catastrophic regressions to baseline truth-telling, illustrating the immense energetic cost of sustaining counterfactual simulations in the nascent preschool brain.

5.3 Cognitive Flexibility and Attentional Set-Shifting

The third pillar of executive function heavily implicated in Joan Peskin’s paradigm is cognitive flexibility, classically operationalized as the capacity to switch attentional sets, shift cognitive rules, and abandon perseverative behavioral scripts. In developmental cognitive science, this capacity is traditionally indexed via Philip Zelazo’s Dimensional Change Card Sort (DCCS) task. In the DCCS, children are taught to sort cards according to one dimension (e.g., color: red cards go here, blue cards go there), and are then abruptly instructed to shift to a completely new sorting dimension (e.g., shape: stars go here, trucks go there). Three-year-old children notoriously fail this shift; they can recite the new rule perfectly, yet they perseveratively continue to sort the cards by the old, prepotent dimension.

The cognitive parallels between DCCS performance curves and Mean Monkey trajectory curves are profound. In the Mean Monkey task, the child is essentially confronted with a rapid-fire rule-switching demand. When interacting with the benevolent, neutral human experimenter, the operative communicative rule is: Truthful Self-Disclosure (“Show me the one you like”). But the moment the Mean Monkey enters the interaction space, the child must rapidly shift their entire communicative set to an inverted rule: Counterfactual Misdirection. They must switch from an approach-truth schema to an avoidance-deception schema, and then switch *back* again once the puppet steps aside.

Younger preschool children become cognitive prisoners of perseverative sets. Having spent their entire developmental lives immersed in communicative environments where adults continuously reinforce honest labeling and truthful self-reporting, three-year-olds find themselves cognitively locked into the “honest-pointing” script. They cannot flexibly disengage from the established association between indexical pointing and true intention. Neurobiologically, this cognitive inflexibility reflects the protracted, ongoing maturation of the prefrontal cortex—specifically the dorsolateral prefrontal cortex and its reciprocal connections with the basal ganglia and anterior cingulate cortex—which forms the physiological substrate necessary for dynamic, real-time set-shifting.

6. Representational Understanding: First-Order vs. Second-Order Beliefs

6.1 Epistemic State Attribution within Competitive Interactions

A fundamental theoretical debate ignited by Joan Peskin’s research concerns the precise cognitive mechanism driving successful deceptive performance. Does a child who successfully points to the unwanted sticker truly understand that they are altering the *epistemic state* (the internal belief) of the monkey, or are they merely operating on an external, non-mentalistic behavioral level? That is, has the child grasped that the puppet has a mind, or have they simply discovered a clever behavioral hack: “Pointing to the ugly sticker makes the monkey take the ugly sticker”?

To differentiate true epistemic state attribution from simple stimulus-response behaviorism, developmental theorists emphasize the critical causal chain linking perceptual input, internal mental representation, and physical motor execution:

1. Perceptual Input: The monkey sees the child point to Item B.
2. Epistemic Representation: The monkey infers, “The child desires Item B,” and forms the internal belief, “Item B is the good prize.”
3. Intentional Execution: Driven by its antagonistic desire to deprive the child, the monkey acts on its false belief by seizing Item B.

Peskin’s empirical paradigm was uniquely equipped to verify whether children genuinely appreciated the causal efficacy of their own communicative misdirection. If a child were relying purely on physical behavioral obstruction, they would prefer physical sabotage (such as hiding the sticker under a hand or sitting on it). However, when children in Peskin’s task engaged in pointing to the distractor, they carefully monitored the monkey’s gaze, ensuring the puppet *saw* the gesture. They recognized that the deception was entirely dependent upon information being received and processed by the puppet’s perceptual faculties. They understood that an unobserved lie has no functional power, confirming that their interventions were aimed squarely at the puppet’s epistemic apparatus.

6.2 First-Order Belief Attribution in the Mean Monkey Task

In standard Theory of Mind terminology, the cognitive operation executed by successful four- and five-year-olds in Peskin’s paradigm constitutes classic First-Order Belief Attribution. First-order mentalizing concerns the direct modeling of an agent’s internal thoughts regarding the external world. In the Mean Monkey task, the specific propositional attitude formulated by the child can be formally modeled as:

“I know that [the sticker on the left is the desirable dinosaur], but I want to make [the monkey believe that [the sticker on the right is the one I desire]].”

Here, the child is operating on two distinct tiers of representation. The child’s own belief directly mirrors reality, while the attributed belief of the puppet is deliberately counterfactual. The child evaluates the monkey not merely as an inanimate obstacle, but as an *intentional, gullible engine*—an agent whose actions are predictably governed by its mental representations, and whose representations can be systematically modified through communicative inputs.

The decisive evidence supporting genuine first-order belief attribution in Peskin’s work emerged from the post-test clinical interviews. When Peskin and subsequent replicators interrogated children about why they had pointed to the incorrect sticker, children who had achieved the four-to-five-year cognitive transition did not respond with purely physical, mechanical descriptions. They did not merely say, “Because it makes his arm move over there.” Instead, their justifications were saturated with rich, unequivocal epistemic language: “Because I wanted to trick him,” “Because I wanted him to think that was the pretty one,” or “So he wouldn’t know my secret.” These verbal justifications provide incontrovertible proof that the children were actively operating within a representational, mentalistic framework.

6.3 Foundations for Second-Order and Advanced Recursive Mentalizing

While the Mean Monkey experiment provides an exceptional diagnostic of first-order mental state attribution, it also clearly exposes the outer boundaries of early preschool cognition, establishing the empirical baseline from which Second-Order Mentalizing will eventually emerge during middle childhood. Second-order mentalizing involves recursive, embedded belief modeling—often stylized as “thinking about thinking about thinking.” It requires the cognitive agent to construct propositions of nested complexity:

“I believe that [the monkey thinks that [I believe the dinosaur is the better prize]].”

In high-level, mature strategic deception—such as the psychological maneuvering found in high-stakes poker, geopolitical statecraft, or sophisticated literary plots—first-order deception rapidly becomes obsolete because the opponent anticipates the lie. If the Mean Monkey were endowed with second-order recursive Theory of Mind, the game would immediately transform. The monkey would compute: “The child knows I am mean and that I will take whatever they point to; therefore, if the child points to the circle, the child is actually trying to protect the dinosaur; consequently, I will ignore their point and seize the dinosaur instead!”

Faced with an opponent capable of second-order mentalizing, a player must deploy advanced counter-strategies, such as the *double-bluff*—pointing to the *actual* target of desire, banking on the opponent assuming that the gesture is a deceptive ruse. Joan Peskin’s preschool participants never displayed double-bluff mechanics, and for good developmental reason: second-order Theory of Mind does not reliably emerge in human ontogeny until approximately seven to nine years of age, as demonstrated in the classic developmental research of Josef Perner and Heinz Wimmer (1985). The preschool child treats the Mean Monkey as an absolute first-order mark—a predictable, naive processor of surface communication. The paradigm thus perfectly captures the initial, triumphant leap into tactical deception while anchoring the preschooler firmly within the first-order developmental band.

7. The Mechanics of Tactical Deception: Pointing, Misdirection, and Withholding

7.1 Non-Verbal Deception: The Dynamics of Misleading Pointing

One of the most fascinating micro-behavioral dimensions of the Mean Monkey experiment lies in its reliance on the indexical pointing gesture. In early human development, pointing is the quintessential cooperative social tool. Emerging between nine and fourteen months of age, protodeclarative pointing represents the infant’s first formal entry into shared intentionality—a pure, uncorrupted act of directing a social partner’s attention to an interesting event for the sole purpose of sharing an affective state. Pointing is fundamentally built upon trust, mutual engagement, and veridical communicative cooperation.

In Peskin’s paradigm, the child is asked to weaponize this sacred cooperative gesture into an instrument of misdirection. This creates acute sensorimotor and spatial alignment challenges. Physical proximity plays a profound role: the two stickers lie directly before the child’s somatic space. For a three-year-old, the indexical finger functions almost like an extension of desire; pointing away from the beloved object feels like an unnatural, somatic betrayal of their own internal feelings. The spatial trajectory of the arm carries severe deictic weight. To point to the unwanted object requires the child to physically aim their body toward a stimulus that repulses them, generating an acute physical-cognitive conflict.

Eye-tracking analyses and micro-behavioral video codings have revealed extraordinary gaze fixation patterns during this process. When three-year-olds fail the task, their eyes and fingers are locked in perfect, unyielding synchrony: they stare fixatedly at the prized sticker, and their finger faithfully follows their gaze. When four- and five-year-olds execute successful non-verbal deception, a profound decoupling occurs. The older child often glances rapidly between the two items, checks the puppet’s face, fixates their gaze firmly onto the *unwanted* item, and then drives their pointing gesture toward that distractor. Some children even deploy deliberate gaze misdirection—staring wide-eyed at the mundane item to amplify the illusion that they are fascinated by it—demonstrating a comprehensive orchestration of non-verbal communicative channels.

7.2 Verbal Misdirection versus Passive Withholding

The Mean Monkey paradigm also provides critical insights into the asymmetric cognitive difficulty separating passive informational withholding from active, spoken fabrication. In spontaneous, unstructured social communication, passive concealment universally precedes active lying in the developmental timeline. A child learns to remain silent about a broken vase or to hide illicitly obtained candy long before they possess the cognitive sophistication to construct an elaborate, plausible, fabricated alibi. Silence simply requires the inhibition of speech; active fabrication requires the real-time invention of an alternative reality.

In Peskin’s experimental design, this divergence was systematically probed by analyzing how children communicated when given varying degrees of verbal freedom. When children were permitted to either remain silent or actively speak, younger preschoolers exhibited a profound bias toward absolute transparency. They were fundamentally constrained by early social and pragmatic training. From the onset of language acquisition, preschool children are continuously conditioned by parents and educators to answer direct questions truthfully: “Where are your shoes?”, “Did you eat the cookie?” The conversational pragmatics of early childhood operate on Paul Grice’s classic Maxim of Quality: make your contribution one that is true; do not say that which you believe to be false.

To violate this universal conversational maxim when the Mean Monkey looks them in the eye and asks, “Which one do you want?”, the child must perform an extraordinary pragmatic breach. Peskin discovered that even when children could conceptually recognize that the monkey would take their preferred item, uttering an explicit, bold-faced linguistic lie (“I want the gray square”) was significantly more difficult for transitional four-year-olds than non-verbal, gestural misdirection. The spoken lie leaves an indelible acoustic trace; it directly violates linguistic socialization. Consequently, tactical pointing often served as the developmental bridge: children could bring themselves to point misleadingly several months before they could comfortably articulate an explicit, smooth verbal fabrication.

7.3 The Problem of Leakage and Micro-Behaviors

Even when four-year-old children successfully cross the developmentalrubicon and deploy tactical deception, they are notoriously poor at managing non-verbal behavioral leakage. In social psychology, Paul Ekman’s leakage theory posits that when individuals attempt to deceive, their true emotional states and authentic knowledge inevitably leak out through micro-expressions, involuntary motor shifts, vocal inflections, and somatic gestures that escape central executive inhibition.

In the Mean Monkey paradigm, emotional and behavioral leakage is rampant among transitional deceivers. Typical four-year-olds executing a deceptive trial exhibit a constellation of tell-tale micro-behaviors:

• Nervous Somatic Tics and Giggling: Upon pointing to the drab sticker, children frequently burst into nervous, high-pitched giggles, bite their lower lip, or rapidly clap their hands over their mouths. This hand-to-mouth gesture is a classic somatic indicator of having spoken or signaled a falsehood.
• Involuntary Gaze Leakage: Immediately after pointing to the unwanted distractor, the child’s eyes dart involuntarily toward the authentic, desired sticker. They check to make sure their prized possession is still safe, a dead giveaway that any perceptive human opponent would instantly exploit.
• Affective Incoherence: Immature deceivers fail to coordinate their facial expressions with their stated desires. A five-year-old might successfully feign excitement over the drab sticker, putting on a theatrical performance of enthusiasm. A four-year-old, however, points to the drab sticker with an expression of profound disgust or utter indifference, displaying a stark affective mismatch between their communicative signal and their emotional presentation.

These leakage phenomena reflect the split-second latency between cognitive decision-making and motor inhibition. The preschool brain must exert immense effort simply to program the false point; it lacks the surplus executive capacity required to simultaneously police facial micro-expressions, smooth out vocal prosody, and suppress involuntary saccadic eye movements. The deceit is functionally successful in the paradigm only because the Mean Monkey puppet is an unyielding, script-bound confederate that ignores these screaming non-verbal clues.

8. Comparative Analysis: Peskin’s Paradigm vs. Other Developmental Tasks

8.1 Comparison with the Wimmer and Perner Unexpected Transfer Task

To fully appreciate Joan Peskin’s contribution to cognitive science, one must rigorously compare the Mean Monkey paradigm with the dominant gold standard of Theory of Mind assessment: Wimmer and Perner’s classic Unexpected Transfer (Maxi/Sally-Anne) task. While both tasks map the profound cognitive shift occurring between the ages of three and five, they differ radically in their motivational, structural, and executive dimensions, as illustrated in the following comparative analysis:

Structural Dimension	Wimmer & Perner (Maxi / Sally-Anne)	Joan Peskin (Mean Monkey Experiment)
Participant Role	Passive third-party spectator	Active, emotionally invested protagonist
Linguistic Load	High: Narrative-heavy story tracking	Low: Minimal, action-oriented, direct communication
Motivational Stakes	Abstract, hypothetical reward for characters	Concrete, tangible, physical incentive (real stickers)
Epistemic Direction	Passive observation of an existing false belief	Active generation and planting of a false belief
Cognitive Conflict	“Cold” cognitive tracking	“Hot” social-affective executive conflict

The Maxi task is fundamentally an exercise in “cold” social cognition. The child sits as a detached judge evaluating a theatrical narrative. The child loses nothing if Maxi looks in the wrong cupboard; there is no personal emotional stake in the outcome. The task demands high levels of verbal narrative processing: the child must track who was in the room, who left the room, what physical event took place during the absence, and decipher an abstract test question: “Where will Maxi look for his chocolate?”

Peskin’s Mean Monkey paradigm, by sharp contrast, is an embodiment of “hot” social cognition. It unfolds within an interactive space saturated with personal stakes, emotional tension, and immediate behavioral feedback. The child is not evaluating a detached story; they are fighting to protect an authentic possession that will physically belong to them at the conclusion of the session. Furthermore, the epistemic dynamic is inverted: rather than merely *detecting* an epistemic error that occurred accidentally, the child must *create* the error strategically. Interestingly, research directly comparing the two paradigms has revealed that while performance curves across large populations broadly align chronologically, certain subsets of children pass competitive deception paradigms slightly earlier than narrative false-belief tasks, suggesting that hot, motivated interaction can occasionally liberate latent mentalizing insights before they can be cleanly articulated in abstract storytelling frameworks.

8.2 Comparison with the Sodian et al. Deception Box Paradigm

In 1991, one year prior to Peskin’s landmark publication, Beate Sodian, Catherine Taylor, Paul L. Harris, and Josef Perner published an influential study in Child Development examining early deception through an experimental protocol commonly known as the “Deception Box” paradigm. Sodian and her colleagues sought to distinguish between two distinct forms of competitive intervention: physical sabotage versus communicative deception.

In the Sodian et al. framework, children were introduced to two characters: a cooperative friend (who gave treats) and a competitive thief (who stole treats). To safeguard their treats, children were provided with two different options. In the sabotage condition, the child could manipulate a physical barrier—placing a solid lock and key on the treat box to physically block the thief’s access. In the deception condition, the treat box remained unlocked, but the child was instructed to place a misleading visual cue (a marker indicating the box was empty when it was full, or vice versa) to trick the thief. Sodian’s findings demonstrated that three-year-olds easily grasped physical sabotage—they gladly locked the box to physically bar the competitor. However, the exact same three-year-olds failed completely when required to deploy communicative deception via symbolic markers.

Joan Peskin’s Mean Monkey paradigm built upon and elegantly refined Sodian’s architecture in several critical ways. While Sodian utilized static boxes, symbolic markers, and contrasting friendly/unfriendly characters, Peskin streamlined the entire dynamic into an immediate, real-time interpersonal game driven by natural deictic communication (pointing and direct speech). Peskin eliminated the confounding necessity of understanding mechanical locks, keys, and symbolic instructional markers, focusing purely on the raw semiotics of interpersonal interaction. In doing so, Peskin established that the preschooler’s difficulty was not with the concept of competition per se (since children readily lock boxes or fight for toys), but specifically with the mentalistic requirement to treat another agent’s mind as an informational system that can be modulated via deliberate misinformation.

8.3 Comparison with Russell et al.’s Windows Task

Another profound comparative touchstone in the cognitive developmental literature is the “Windows Task,” formulated by James Russell, Frances Mauthner, Sonia Sharpe, and Tim Tidswell in 1991. The Windows Task was designed to specifically interrogate the executive, inhibitory demands of competitive pointing without requiring explicit mentalizing language. In this task, two opaque boxes are placed before the child; each box features a cut-out window on the experimenter’s side and the child’s side. Through the window, the child can clearly see that Box A contains a delicious chocolate, while Box B is empty.

An antagonistic adult partner sits across the table. The immutable rule of the game is established: whichever box the child points to, the opponent will open, and the opponent gets to keep whatever is inside! Therefore, to obtain the chocolate for themselves, the child must point to the *empty* box (Box B), leaving the chocolate box (Box A) to be retrieved by the child on the subsequent step. The empirical findings of the Windows Task mirrored Peskin’s results with uncanny precision: three-year-old children exhibited an overwhelming, perseverative tendency to point directly to the window containing the visible chocolate, systematically forfeiting the prize to their rival on trial after trial.

However, an intense methodological debate erupted between researchers regarding what these tasks were truly measuring. In the Windows Task, the opponent is an adult human sitting silently, and the boxes are inanimate physical objects with transparent windows. Many cognitive theorists argued that the Windows Task was not measuring Theory of Mind at all, but was simply a pure test of motor-inhibitory control and associative learning contingencies (stimulus-reward reversal learning). Peskin’s Mean Monkey paradigm fundamentally diverged from this sterile mechanical setup by imbuing the opponent with vivid, explicit, intentional social agency. The Mean Monkey was not a silent box-opener; he was an active conversational partner who interrogated the child about their internal preferences. By embedding the task within rich, deictic social pragmatics, Peskin ensured that the paradigm measured social-epistemic inference rather than mere non-social contingency learning.

9. Ecological Validity, Motivation, and Experimental Variables

9.1 The Efficacy of Puppet-Based Methodologies in Early Childhood

The enduring success of Joan Peskin’s experimental design is profoundly rooted in its brilliant exploitation of early childhood developmental psychology through the use of puppet-based methodologies. In adult cognitive neuroscience, introducing a hand puppet would be regarded as an eccentric, potentially confounding intervention. In the psychological reality of the preschool child, however, puppets represent an ideal methodological bridge between real-world social ecology and controlled laboratory standardization.

Developmental research has extensively established that preschool-aged children readily engage in anthropomorphic attribution. They naturally endow puppets with thoughts, feelings, independent agency, intentionality, and moral accountability. Furthermore, interacting with a puppet dramatically reduces social intimidation. When an adult human examiner sits before a three-year-old child and begins issuing commands, the child is immediately subject to intense social compliance pressures, authority-figure deference, and performance anxiety. These adult-child dynamics frequently cause young participants to freeze, defer, or adopt passive response strategies. A puppet, conversely, operates at the child’s own perceived social and symbolic level. It is a peer-surrogate.

Crucially, the puppet allows for the ethical and affective suspension of normal social reality. If an adult human experimenter were to repeatedly, aggressively seize a three-year-old’s favorite stickers while gloating in their face, the experiment would rapidly run into severe ethical boundaries regarding child distress, and the child’s resulting tears would terminate the session. But when the *Mean Monkey* does it, the encounter is framed within the universal preschool grammar of pretend play. The child recognizes the puppet as an antagonist within a playful game world. This pretend-play frame allows the experimenter to maintain high levels of competitive tension and repeated tangible losses without inflicting psychological trauma or triggering social avoidance, keeping the young participant actively and enthusiastically engaged across demanding, repetitive trial blocks.

9.2 Incentive Salience and Reward Structuring

A critical, often underestimated variable governing performance in the Mean Monkey experiment is incentive salience. The entire cognitive architecture of tactical deception is fueled by the motivational drive to protect an item of authentic value. If the experimental rewards lack sufficient hedonic currency, the paradigm collapses: a child who does not genuinely care about the outcome has no reason to incur the formidable mental and executive costs required to construct a counterfactual ruse.

However, developmental science reveals that motivational arousal functions as a complex, double-edged sword, precisely mirroring the classic Yerkes-Dodson Law:

• Optimal Arousal: Highly attractive stickers (glittering holographic designs, beloved cartoon icons) generate the intense competitive motivation necessary to drive strategic, tactical thinking, prompting the child to search for ways to outsmart their opponent.
• Hyper-Arousal and Cognitive Collapse: If the incentive salience is calibrated *too* high, the prepotent affective pull of the reward can completely overwhelm the nascent inhibitory control networks of the prefrontal cortex. Under hyper-salient conditions, even a transitional four-year-old who possesses the conceptual capacity to deceive may suffer executive collapse, their hand impulsively shooting out to grasp or point toward the dazzling prize before their higher-order mentalizing strategies can intervene.

To guard against these extremes, Peskin carefully calibrated her incentive structures. The baseline preference trials ensured that there was a marked, unequivocal disparity in subjective value between the target item and the distractor item, eliminating ambiguous indifference. Furthermore, the physical array of stickers was continuously refreshed across trials to prevent satiation effects—a phenomenon wherein a child becomes indifferent after collecting multiple identical stickers, which would artificially deflate competitive motivation on later trials.

9.3 Socioeconomic and Linguistic Variables in Task Performance

While the transition from counterproductive honesty to tactical deception is a universal developmental milestone, the precise chronological timing of this transition is significantly moderated by environmental, socioeconomic, and linguistic variables. Extensive cross-sectional research following Peskin’s paradigm has demonstrated that a child’s family environment plays a critical role in scaffolding early mentalizing capabilities.

One of the most robust predictors of early success on the Mean Monkey task is the frequency and quality of internal state language utilized within the home. Children raised by parents who frequently use mentalistic vocabulary in natural conversation—words such as *think*, *believe*, *know*, *remember*, *pretend*, and *trick*—cross the four-year false-belief and deception thresholds significantly earlier than children whose linguistic environments are dominated by purely behavioral or imperative language (“Sit down,” “Eat your food”). Exposure to mental state discourse provides the child with the symbolic, linguistic scaffolding required to represent unobservable mental representations as discrete, manipulable entities.

Furthermore, the structural composition of the family exerts a profound influence. The presence of older siblings has been repeatedly identified as a powerful catalyst accelerating performance on competitive deception tasks. A preschool child growing up with older siblings is immersed in a continuous, daily social crucible of negotiation, competitive resource allocation, teasing, and tactical maneuvering. Sibling conflict provides an authentic, high-stakes training ground where the functional consequences of knowledge leakage and the tangible payoffs of tactical misdirection are discovered organically. Consequently, children with older siblings routinely outperform only-children on the Mean Monkey paradigm, demonstrating that competitive social ecology actively refines the ontogenetic timeline of the human social brain.

10. Neurodevelopmental Foundations and Atypical Trajectories

10.1 Autism Spectrum Conditions and the Deception Deficit

The application of Joan Peskin’s Mean Monkey paradigm to atypical clinical populations has provided some of the most theoretically profound insights into the neurodevelopmental architecture of human social cognition. Chief among these investigations has been the testing of children diagnosed with Autism Spectrum Disorder (ASD). Decades of research initiated by Simon Baron-Cohen, Alan Leslie, and Uta Frith demonstrated that individuals with autism experience profound, domain-specific difficulties with Theory of Mind—a condition famously conceptualized as mindblindness.

When high-functioning children with autism (who possess chronological ages and verbal IQs far exceeding the typical five-year-old baseline, often ranging from 8 to 14 years old) are introduced to the Mean Monkey experiment, the results are startling. Despite demonstrating exceptional mechanical intelligence, intact visual-spatial perception, and advanced vocabulary, children with autism display a persistent, debilitating inability to deploy tactical deception. Trial after trial, they point honestly to the sticker they genuinely desire, and watch in bafflement as the Mean Monkey repeatedly steals it.

Crucially, this clinical failure cannot be attributed to an inability to understand the physical or mechanical rules of the game. A child with autism can explicitly tell the experimenter: “The monkey takes the sticker I point to.” They understand the physical contingency perfectly. What they fundamentally lack is the intuitive mentalistic bridge: the understanding that the pointing gesture functions to instill a false mental state in the monkey’s mind, which then directs the monkey’s physical behavior. They treat communication as an absolute, literal, veridical pipeline of information. The Mean Monkey paradigm thus isolates the core social-communicative deficit of autism with surgical precision, proving that tactical deception is not a trivial byproduct of general intelligence, but relies on specialized, domain-specific social-cognitive mechanisms that are selectively disrupted in spectrum conditions.

10.2 Attention-Deficit/Hyperactivity Disorder and Inhibitory Vulnerabilities

Testing children with Attention-Deficit/Hyperactivity Disorder (ADHD) on the Mean Monkey paradigm reveals a behavioral profile that stands in stark, fascinating contrast to the performance of children with autism. This contrast provides a vivid double-dissociation between the conceptual components of Theory of Mind and the executive machinery of behavioral inhibition.

Children with ADHD possess a fully intact, developmentally appropriate Theory of Mind. When tested on standard false-belief tasks or presented with hypothetical narrative vignettes, they readily identify what characters think, predict behavioral outcomes based on internal beliefs, and fully articulate the conceptual logic of a trick. When placed in the Mean Monkey game, an older preschool or early school-aged child with ADHD knows *exactly* what strategy should be deployed. They can excitedly tell the experimenter before the trial begins: “I’m going to trick him! I’m going to point to the ugly one so I get the cool car!”

Yet, when the Mean Monkey pops onto the table, leans forward, and demands an immediate choice, the severe motor-inhibitory vulnerabilities characteristic of ADHD cause the behavioral execution to fracture catastrophically. The child’s hand impulsively shoots out and points straight at the desirable car. The child often gasps in immediate horror the moment their finger lands on the prize, exclaiming: “Oh no! Why did I do that?! I didn’t mean to point there!” Here, there is zero conceptual failure. The child understands deception; they understand mental representations; their working memory holds the goal. But their prefrontal dopaminergic inhibitory gating fails to suppress the prepotent motor impulse in real time. This clinical profile underscores the vital necessity of differentiating between a conceptual mindblindness deficit (as seen in ASD) and an executive performance breakdown (as seen in ADHD), demonstrating the exceptional diagnostic utility of Peskin’s experimental framework.

10.3 Neural Substrates Implicated in the Mean Monkey Task

Modern advances in developmental cognitive neuroscience and pediatric functional neuroimaging have illuminated the intricate neural circuitry that supports the triumphant developmental transition captured by the Mean Monkey experiment. The execution of real-time tactical deception recruits a highly distributed, integrated network spanning both the “Mentalizing Network” and the “Central Executive Network”:

• The Temporoparietal Junction (TPJ): The right and left TPJ, situated at the intersection of the temporal and parietal lobes, serve as the core computational hub for mental state attribution. The TPJ is specifically activated when an individual tracks dynamic, shifting epistemic states, distinguishing between one’s own internal knowledge and the counterfactual beliefs of another social agent. In the Mean Monkey task, the TPJ computes the monkey’s gullible mental state.
• The Medial Prefrontal Cortex (mPFC): The mPFC is heavily implicated in social evaluation, mentalizing about intentionality, and moral self-referential processing. It enables the child to reflect upon the competitive nature of the interaction and conceptualize the monkey as an intentional rival.
• The Dorsolateral Prefrontal Cortex (dlPFC): The dlPFC provides the raw computational power required for executive working memory maintenance and the active suppression of prepotent responses. It is the dlPFC that holds the dual tracks of authentic preference versus fabricated misdirection and prevents the motor cortex from executing an impulsive point toward the true target.
• The Anterior Cingulate Cortex (ACC): The dorsal ACC functions as the brain’s internal conflict detection and monitoring engine. In Peskin’s paradigm, the ACC fires intensely as the child experiences the profound cognitive conflict between the powerful desire to claim the prized sticker and the strategic demand to point to the unwanted distractor, signaling the need for heightened executive control.

Pediatric neuroimaging indicates that between the ages of three and five, these disparate cortical regions undergo massive synaptic pruning, white-matter myelination, and functional connectivity integration. The maturation of structural tracts—most notably the superior longitudinal fasciculus connecting the frontal and parietal hubs—physiologically enables the rapid cross-talk between the TPJ (understanding the monkey’s mind) and the prefrontal cortex (inhibiting the truthful point). The behavioral victory observed in Joan Peskin’s five-year-old participants is, at its root, the behavioral manifestation of a fully integrated social-executive neurodevelopmental axis.

11. Replications, Cross-Cultural Findings, and Methodological Refinements

11.1 Direct Replications and Systematic Variations of Peskin’s Design

In the decades following its initial 1992 publication, Joan Peskin’s Mean Monkey experiment has achieved classic status within developmental science, prompting hundreds of direct replications, conceptual variations, and methodological extensions across global laboratories. The fundamental empirical finding—the robust, dramatic shift from rigid counterproductive honesty at age three to fluid strategic misdirection by age five—has demonstrated extraordinary empirical replicability, standing as one of the most stable behavioral phenomena in cognitive developmental literature.

Subsequent researchers have systematically varied key parameters of the original design to isolate subtle operational variables. One notable variation evaluated the effect of altering the puppet’s disposition. When the antagonist is replaced with a “Confused Monkey” or a “Helpful Monkey” (who genuinely wants to give the child what they desire, but gets mixed up), researchers observed changes in communicative behavior. Children never deceived the helpful puppet, confirming that young preschoolers do not deploy false pointing randomly; their deceptive communicative acts are selectively and adaptively deployed exclusively within competitive contexts characterized by opposing goals.

In modern contemporary laboratories, the physical hand puppet has frequently been updated with digital equivalents: interactive computer avatars, touchscreen animations, and autonomous humanoid robots (such as SoftBank Robotics’ NAO or Pepper). Interestingly, these technological adaptations have reaffirmed the enduring wisdom of Peskin’s original design. Children interacting with two-dimensional flat-screen avatars often show slight delays in deceptive strategic emergence compared to those interacting with a physical, three-dimensional puppet agent seated right across the table. The visceral, immediate physicality of a co-present puppet maximizes social engagement and ecological validity, ensuring that the child’s social-cognitive faculties are mobilized to their absolute peak capacity.

11.2 Cross-Cultural Testing and Universal Developmental Trajectories

A central imperative of modern cognitive science is establishing whether observed developmental milestones represent culturally specific artifacts of Western, Educated, Industrialized, Rich, and Democratic (WEIRD) societies, or whether they reflect universal ontogenetic trajectories of the human species. The Mean Monkey paradigm and its close conceptual analogs have been extensively deployed in diverse cross-cultural settings across Asia, Africa, Europe, and Indigenous communities globally.

The comparative cross-cultural data reveal a remarkable, universal biological baseline: across virtually all studied human cultures, the fundamental transition from transparent honesty to functional tactical deception reliably occurs between the ages of three and five. The cognitive mechanisms enabling counterfactual mentalizing and prepotent motor inhibition appear to mature along an evolutionary conserved biological schedule. A three-year-old child in rural Cameroon, an urban preschooler in Tokyo, and a toddler in Toronto all exhibit an initial, intuitive bias toward veridical, honest communication when asked to indicate their preferences.

However, the cultural landscape introduces profound, nuanced variations in the *pragmatic deployment* and societal evaluation of deception. In collectivist cultures—such as contemporary urban China or Japan—socialization practices heavily prioritize social harmony, group cohesion, emotional restraint, and the avoidance of interpersonal conflict. In these settings, children develop an exceptionally early, highly sophisticated capacity for **prosocial lying** or “white lies”—feigning joy upon receiving an unappealing gift to spare the feelings of an elder, or telling polite untruths to maintain social decorum. Conversely, in highly individualistic cultures where competitive independence, personal resource defense, and assertive autonomy are socialized early, children often display earlier, more assertive mastery of **tactical, competitive deception** against an antagonistic rival. Furthermore, cross-linguistic evaluations have highlighted the critical importance of instructional phrasing: verbs meaning “to trick” carry vastly different moral and pragmatic connotations across global languages, requiring researchers to carefully calibrate task semantics to avoid unintentionally suppressing children’s natural strategic instincts through moral intimidation.

11.3 Eye-Tracking and Implicit Measures of Deceptive Competence

The dawn of twenty-first-century eye-tracking technologies, high-speed pupil-dilation recording, and micro-latency kinematic analyses has revolutionized our understanding of Joan Peskin’s paradigm, uncovering a hidden layer of **implicit mentalizing competence** that long escaped naked behavioral observation. For decades, developmental science assumed that because three-year-old children pointed honestly to the desirable sticker, they were completely oblivious to the impending strategic disaster. Modern implicit measurements tell a dramatically different, far more complex story.

When contemporary researchers equip three-year-old participants with wearable, non-invasive eye-tracking systems during the Mean Monkey task, an astonishing dissociation emerges between gaze kinematics and explicit motor execution:

• Anticipatory Eye Movements: Seconds before their index finger shoots out to point toward the desirable sticker, the three-year-old’s eyes routinely dart toward the Mean Monkey’s hand, and then instantly toward the prized sticker. The child’s ocular tracking system displays clear, robust *anticipatory looking*—they implicitly predict that the monkey is going to seize the item they are about to touch!
• Pupillometric Arousal: Pupil dilation metrics indicate a massive surge in sympathetic autonomic nervous system arousal precisely at the moment of pointing. The three-year-old child is not blithely unaware; their physiological system registers high cognitive conflict and impending dread.
• Micro-Latency Dynamics: When a three-year-old points to the desirable sticker in the presence of the Mean Monkey, the latency of the reaching movement is significantly slower and displays greater kinematic trajectory deviation than when they point to the sticker in a non-competitive baseline condition.

These cutting-edge implicit data demonstrate that three-year-old children possess an emergent, implicit awareness of the opponent’s intentions and the behavioral contingency long before they can translate that awareness into an explicit, executed deceptive strategy. The cognitive bottleneck is not an absolute, total ignorance of the social dynamics, but an inability to coordinate implicit knowledge, working memory manipulation, and motor-inhibitory control into a unified, outward behavioral victory. Peskin’s paradigm thus captures a dynamic, evolving cognitive system mid-flight, where implicit understanding gradually forces its way through executive barriers to emerge into daylight as triumphant explicit strategy.

12. Theoretical Legacy and Contemporary Educational Implications

12.1 Impact on Foundational Developmental Cognitive Theory

Joan Peskin’s 1992 Mean Monkey experiment holds an unshakeable, foundational legacy in the annals of cognitive developmental theory. Its most decisive conceptual contribution was fundamentally breaking the absolute monopoly of passive, narrative-heavy, spectator-based false-belief tasks. By demonstrating that social cognition could be rigorously assessed through an interactive, ecologically authentic, competitive game, Peskin shifted the entire paradigm of developmental psychology from the study of detached theoretical observation to the study of active, situated social engagement.

Furthermore, Peskin’s work provided critical empirical fuel for resolving one of the most intense theoretical disputes in developmental science: the debate over the developmental priority of desire versus belief. Prominent constructivist theorists, such as Henry Wellman and Karen Bartsch, posited that young children progress through an evolutionary sequence termed “Desire Psychology” (at age two and early three, understanding that people act according to subjective desires) before graduating to “Desire-Belief Psychology” (around age four, understanding that actions are mediated by beliefs, which may be false). The Mean Monkey experiment captured this theoretical shift with supreme clarity. The tragic failure of the three-year-old is that they are entirely trapped within Desire Psychology: they believe that external actions must directly reflect internal desires (“I want it, so I point to it”). The five-year-old’s victory reflects the full integration of Belief Psychology: they realize that to satisfy their *desire*, they must strategically manipulate the other agent’s *belief*.

Beyond its specialized academic influence, the Mean Monkey study fundamentally revolutionized how society understands childhood lying. In classic pedagogical and traditional moral frameworks, lying was historically pathologized as a moral failing, a behavioral disturbance, or a sign of emerging delinquency requiring immediate punitive correction. Peskin’s research inverted this paradigm, proving that the emergence of tactical deception between the ages of three and five is, in reality, a vital, triumphant milestone of healthy cognitive maturation. Far from indicating moral defect, early childhood lying provides irrefutable empirical proof that a child’s Theory of Mind, prefrontal executive networks, working memory capacity, and social perspective-taking architecture are firing on all cylinders. Today, the Mean Monkey experiment is celebrated across the globe as an indispensable, classic teaching paradigm in undergraduate and graduate cognitive science curricula.

12.2 Educational and Clinical Assessment Applications

The practical, real-world translations of Joan Peskin’s paradigm have generated profound benefits across educational policy, clinical developmental screening, and pediatric diagnostic frameworks. In modern developmental and clinical psychology, modified and simplified variants of the Mean Monkey game are increasingly utilized as diagnostic behavioral assays for early social-communicative vulnerabilities.

Because the Mean Monkey task strips away the heavy linguistic and syntactic demands of standard verbal false-belief stories, it offers a clean, non-linguistically biased diagnostic tool for evaluating social cognition in children presenting with:

• Developmental language disorders,
• Childhood apraxia of speech,
• Selective mutism, or
• Socioeconomic and linguistic-minority backgrounds.

By observing a child’s capacity to deploy non-verbal tactical pointing within an interactive game, clinicians can quickly distinguish between a pure linguistic processing delay and a deeper, domain-specific socio-cognitive impairment, allowing for highly targeted early intervention strategies.

In educational and early childhood pedagogical domains, Peskin’s paradigm has inspired revolutionary curricula designed to cultivate perspective-taking and social problem-solving through structured, playful peer games. Early childhood educators are increasingly trained to move away from purely punitive reactions when preschoolers begin experimenting with untruths. Instead, educators use puppet-mediated interactive play to coach children through perspective-taking exercises. By introducing benevolent, confusing, or competitive puppet characters into classroom activities, teachers provide young children with a safe, emotionally regulated playground where they can explore how informational access shapes internal mental states, cultivating empathy, perspective-taking, and social-cognitive flexibility through joyful, scaffolded play.

12.3 Open Questions and Future Directions in Deception Research

As developmental cognitive science navigates the twenty-first century, the theoretical trail blazed by Joan Peskin continues to expand into uncharted empirical territories. One of the most urgent and active frontiers of contemporary research focuses on the developmental bridge connecting early **tactical, self-serving deception** (as assessed in the Mean Monkey game) with late-emerging **prosocial, altruistic, and moral deception**.

While the Mean Monkey task examines how a child manipulates an informational channel to protect a personal physical asset against an antagonist, how does that same child learn to deploy falsehoods to protect *others*? How does the human mind navigate the hyper-complex cognitive calculus of telling a “white lie” to prevent hurting a friend’s feelings, keeping a prosocial secret to protect a peer from harm, or deceiving a hostile authority figure to safeguard a sibling? Current developmental investigations are designing sophisticated, multi-agent game-theoretic paradigms to track how the raw first-order deceptive machinery unlocked in the Mean Monkey experiment is subsequently modulated by moral reasoning, empathy, and cultural norms across middle childhood and adolescence.

Another rapidly accelerating frontier concerns the profound impact of the contemporary digital ecosystem on early mentalizing trajectories. In an era where preschool children are immersed in daily interactions with screen-mediated agents, algorithmic digital characters, voice-activated AI assistants (such as Siri and Alexa), and responsive robotic toys, how is the ontogenetic timeline of social cognition being reshaped? Do children attribute independent epistemic states, beliefs, and gullibility to artificial intelligence engines in the same manner they do to anthropomorphic hand puppets? Can children learn to deceive a digital algorithm, and does competitive interaction with non-human computational agents accelerate or delay the emergence of human Theory of Mind?

Finally, the integration of real-time developmental neuroimaging—such as functional Near-Infrared Spectroscopy (fNIRS), which allows high-resolution monitoring of cortical hemodynamic activity in awake, actively moving preschool children—with ecologically authentic behavioral games like the Mean Monkey experiment promises to finally dissolve the remaining mysteries of the mind. By mapping the precise split-second neural activations within the temporoparietal junction, anterior cingulate, and prefrontal cortex as a child looks at a rival, smiles, and intentionally points to the wrong sticker, cognitive neuroscience stands on the precipice of directly visualizing the physical birth of counterfactual thought.

In the final synthesis, Joan Peskin’s 1992 Mean Monkey experiment remains an enduring masterpiece of developmental methodology. With a hand puppet, two stickers, and a simple table, Peskin unlocked one of the most profound, beautiful secrets of the human species: the moment the emerging mind escapes the prison of literal reality, discovers the subjective interiority of other human beings, and steps boldly, playfully, and strategically into the complex, marvelous theatre of the social world.

References

• Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a “theory of mind”? Cognition, 21(1), 37–46. https://doi.org/10.1016/0010-0277(85)90022-8
• Bartsch, K., & Wellman, H. M. (1995). Children talk about the mind. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195091892.001.0001
• Byrne, R. W., & Whiten, A. (Eds.). (1988). Machiavellian intelligence: Social expertise and the evolution of intellect in monkeys, apes, and humans. Oxford University Press.
• Carlson, S. M., & Moses, L. J. (2001). Individual differences in inhibitory control and children’s theory of mind. Child Development, 72(4), 1032–1053. https://doi.org/10.1111/1467-8624.00333
• Ekman, P., & Friesen, W. V. (1969). Nonverbal leakage and clues to deception. Psychiatry, 32(1), 88–106. https://doi.org/10.1080/00332747.1969.11023575
• Flavell, J. H. (1986). The development of children’s knowledge about the appearance-reality distinction. American Psychologist, 41(4), 418–425. https://doi.org/10.1037/0003-066X.41.4.418
• Leslie, A. M. (1987). Pretense and representation: The origins of “theory of mind.” Psychological Review, 94(4), 412–426. https://doi.org/10.1037/0033-295X.94.4.412
• Perner, J., & Wimmer, H. (1985). “John thinks that Mary thinks that…”: Attribution of second-order beliefs by 5- to 10-year-old children. Journal of Experimental Child Psychology, 39(3), 437–471. https://doi.org/10.1016/0022-0965(85)90051-7
• Peskin, J. (1992). Ruse and representations: On children’s ability to conceal information. Developmental Psychology, 28(1), 84–89. https://doi.org/10.1037/0012-1649.28.1.84
• Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a theory of mind? Behavioral and Brain Sciences, 1(4), 515–526. https://doi.org/10.1017/S0140525X00076512
• Russell, J., Mauthner, N., Sharpe, S., & Tidswell, T. (1991). The ‘windows task’ as a measure of strategic deception in preschoolers and autistic subjects. British Journal of Developmental Psychology, 9(2), 331–349. https://doi.org/10.1111/j.2044-835X.1991.tb00881.x
• Saxe, R., & Kanwisher, N. (2003). People thinking about thinking people: The fMRI investigations of theory of mind. NeuroImage, 19(4), 1835–1842. https://doi.org/10.1016/S1053-8119(03)00230-1
• Sodian, B., Taylor, C., Harris, P. L., & Perner, J. (1991). Early deception on the functional and representational level: To take or to teach? Child Development, 62(3), 468–483. https://doi.org/10.2307/1131124
• Tomasello, M., Call, J., & Hare, B. (2003). Chimpanzees understand psychological states—the question is which ones and to what extent. Trends in Cognitive Sciences, 7(4), 153–156. https://doi.org/10.1016/S1364-6613(03)00035-4
• Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72(3), 655–684. https://doi.org/10.1111/1467-8624.00304
• Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception. Cognition, 13(1), 103–128. https://doi.org/10.1016/0010-0277(83)90004-5
• Zelazo, P. D., Frye, D., & Rapus, T. (1996). An age-related dissociation between knowing rules and using them in a sorting task. Cognitive Development, 11(1), 37–63. https://doi.org/10.1016/S0885-2014(96)90027-1