Josef Perner – 1948 Present

Biography

The landscape of developmental cognitive science underwent a profound ontological and methodological transformation in the final quarter of the twentieth century. For decades, the dominant paradigm had been Jean Piaget’s genetic epistemology, which conceptualized childhood intellect as a gradual progression through domain-general operational stages, moving from sensorimotor reflexes to abstract, formal operations. While Piaget illuminated the child’s burgeoning mastery over the physical world of conservation, quantity, and spatial relations, his framework left the child’s understanding of the social world—specifically, the inner architecture of human minds—largely subsumed under the pejorative umbrella of cognitive “egocentrism.” In this intellectual climate, young children were viewed as fundamentally trapped within their own perceptual horizons, unable to distinguish their subjective vantage point from objective reality, yet the precise cognitive machinery governing this transition remained elusive and theoretically underspecified.

The decisive breakthrough that dismantled this orthodoxy arrived in the early 1980s, spearheaded by the Austrian cognitive psychologist Josef Perner. Born in 1948 in Radstadt, Austria, Perner embarked on an intellectual journey that fused the rigorous conceptual rigor of European continental philosophy with the empirical exactitude of Anglo-American cognitive psychology. Collaborating with Heinz Wimmer at the University of Salzburg, Perner engineered an experimental paradigm that redefined the scientific study of childhood: the unexpected-transfer false-belief task. By operationalizing what it means for a child to comprehend that another individual can hold a mental representation of the world that directly contradicts physical reality, Wimmer and Perner provided empirical science with an objective litmus test for the presence of a “Theory of Mind”—a phrase originally coined by David Premack and Guy Woodruff in their 1978 primatological inquiries, but which Perner transformed into a cornerstone of developmental psychology.

Across more than four decades of relentless empirical investigation and philosophical synthesis, Perner expanded this initial breakthrough into a comprehensive cognitive architecture. His landmark 1991 treatise, Understanding the Representational Mind, moved far beyond empirical data collection to establish a foundational taxonomy of cognitive representation: tracing the phylogenetic and ontogenetic progression from direct, primary representations to decoupled secondary representations, and ultimately to metarepresentations capable of encoding the medium-content distinction of intentional mental states. Perner’s scholarship has systematically demonstrated that the watershed transition observed around the fourth year of life is not a localized, domain-specific heuristic, but a profound representational reorganization that simultaneously unlocks counterfactual reasoning, episodic memory, autonoetic consciousness, executive control, and the philosophical appreciation of referential opacity. This comprehensive treatise explores the life, experimental paradigms, theoretical syntheses, and enduring intellectual legacy of Josef Perner—a scholar whose work fundamentally decrypted how the human child learns to navigate the invisible, representational universe of the mind.

1. Biographical Trajectory and Academic Foundations (1948–Early Career)

1.1 Formative Years and Education in Austria and the United Kingdom

Josef Perner was born on May 1, 1948, in the historic alpine town of Radstadt, situated in the federal state of Salzburg, Austria. Growing up in the immediate aftermath of the Second World War, Perner’s early education took place in an Austrian society undergoing vigorous socio-cultural reconstruction. During this period, the Austrian academic landscape was heavily anchored in classical traditions, emphasizing foundational philosophy, formal logic, and rigorous scientific methodologies. As a young student, Perner exhibited a distinct aptitude for analytical inquiry, displaying an intellectual temperament that resisted superficial explanations in favor of deep structural mechanisms.

Entering the Paris Lodron University of Salzburg in 1968, Perner pursued an interdisciplinary course of undergraduate studies spanning Psychology, Mathematics, and Philosophy, completing his degree in 1972. This triadic curriculum was instrumental in shaping his intellectual identity. While psychology provided the empirical domain of inquiry, mathematics instilled an uncompromising demand for formal modeling and logical precision, and philosophy grounded him in the classical problems of epistemology, phenomenology, and intentionality. Unlike many contemporary developmentalists who approached child behavior from purely descriptive or clinical vantage points, Perner interrogated childhood cognitive transitions through an explicitly epistemological lens: What does it mean for an epistemic agent to know, to represent, or to misrepresent reality?

Seeking to expand his empirical and theoretical training beyond the continental milieu, Perner moved to North America to undertake doctoral studies at the University of Toronto. There, he worked under the supervision of the renowned mathematical psychologist and systems theorist Anatol Rapoport, completing his Ph.D. in 1978. Rapoport’s mentorship exposed Perner to game theory, decision-making dynamics, mathematical modeling, and rigorous experimental design. This rigorous post-graduate training, followed by formative academic exposure in the United Kingdom, allowed Perner to achieve a rare synthesis in cognitive science: merging the philosophical heritage of Franz Brentano and Gottlob Frege with the analytic philosophy of mind and the experimental paradigms of modern cognitive psychology.

1.2 Academic Appointments and Institutional Affiliations

Following the completion of his doctorate in Toronto, Perner transitioned to the United Kingdom in 1979 to join the faculty of the Laboratory of Experimental Psychology at the University of Sussex. Sussex during the late 1970s and 1980s was an exceptionally fertile intellectual epicenter for cognitive science, artificial intelligence, and psycholinguistics. Operating within this highly collaborative and computationally driven environment, Perner rose from Lecturer to Reader and ultimately Professor of Experimental Psychology, establishing an internationally recognized laboratory dedicated to unpacking the cognitive mechanisms underlying children’s causal, counterfactual, and social reasoning.

During his fifteen-year tenure at Sussex (1979–1994), Perner built vital research collaborations across Britain, Western Europe, and North America. He formed close intellectual ties with leading cognitive theorists in the emerging field of social cognitive development, engaging in profound debates with scholars such as Alan M. Leslie, Simon Baron-Cohen, Uta Frith, Alison Gopnik, and Henry M. Wellman. Perner’s laboratory rapidly became a proving ground for international researchers seeking to understand how mental state concepts are acquired and mobilized during early childhood.

In 1995, Perner returned to his alma mater, accepting the Chair of Cognitive Psychology at the Paris Lodron University of Salzburg. His appointment marked the repatriation of one of Austria’s most influential living psychological scientists. At Salzburg, Perner spearheaded the modernization of the Department of Psychology, transforming it into an internationally celebrated hub for experimental cognitive science. He was central to the establishment and intellectual direction of the Centre for Cognitive Neuroscience (CCNS) in Salzburg, integrating classical behavioral and developmental methodologies with cutting-edge functional neuroimaging (fMRI) and electrophysiological techniques. Perner served as an active professor at Salzburg until his formal retirement in 2016, at which point he was appointed Professor Emeritus, continuing his theoretical writing, empirical collaborations, and philosophical debates.

1.3 Intellectual Context of Mid-to-Late 20th-Century Developmental Psychology

To fully appreciate the conceptual revolution catalyzed by Perner, one must understand the theoretical paralysis that characterized developmental psychology in the late 1970s. The orthodox paradigm of Jean Piaget, which had dominated the field for half a century, was experiencing widespread empirical and theoretical critiques. Piaget had framed young children (specifically those in the preoperational stage, roughly ages two to six) as radically “egocentric”—incapable of decentering from their own immediate perceptual schemas to appreciate that another individual could view a three-dimensional mountain landscape differently. However, Piagetian tasks were increasingly recognized as possessing high cognitive and linguistic overhead, often conflating a child’s operational limitations with failures of motor coordination, spatial memory, or task comprehension.

Concurrently, the cognitive revolution had ushered in information-processing models that conceptualized the human mind as a computational device operating over symbolic representations. In linguistics, Noam Chomsky had demonstrated that domain-general learning theories could not account for the rapid, poverty-of-the-stimulus acquisition of syntactic structures. Extending this logic, philosopher Jerry Fodor published The Modularity of Mind in 1983, arguing that human cognition consists of encapsulated, innately specified, domain-specific modules designed to process specific environmental inputs—such as facial configurations, linguistic speech streams, and basic physical trajectories—with high computational efficiency and minimal central cognitive interference.

Amid these tectonic debates on modularity and domain-specificity, a catalytic spark arrived from comparative primatology. In 1978, David Premack and Guy Woodruff published a landmark paper in The Behavioral and Brain Sciences titled “Does the chimpanzee have a theory of mind?” Premack and Woodruff sought to determine whether a chimpanzee (Sarah) understood that human actors possessed internal mental states—such as goals, intentions, and knowledge—that guided their problem-solving behavior. In the extensive open peer commentary that followed, three prominent philosophers of mind—Jonathan Bennett, Daniel Dennett, and Gilbert Harman—independently identified a fatal methodological flaw in Premack and Woodruff’s design. If an animal or child merely predicts an agent’s behavior when that agent’s belief corresponds to reality, one cannot determine whether the observer is tracking the agent’s internal mental state or simply tracking the objective state of the physical world itself. The definitive, unimpeachable criterion for the possession of a Theory of Mind was the capacity to understand and predict behavior driven by a false belief—a mental representation explicitly decoupled from, and in direct contradiction to, empirical reality. This philosophical gauntlet set the stage for Josef Perner’s historic intervention.

2. The Genesis of Theory of Mind Research: The 1983 Paradigm Shift

2.1 Collaboration with Heinz Wimmer and Experimental Conceptualization

Recognizing the profound implications of the philosophical commentary surrounding Premack and Woodruff’s work, Josef Perner formed a critical research partnership with his Salzburg colleague, Heinz Wimmer. Wimmer and Perner realized that the theoretical impasse in comparative psychology could be resolved within human developmental laboratories. While philosophers had established the theoretical imperative of the false-belief criterion, no empirical protocol existed that could reliably, cleanly, and unambiguously assess this capacity in young children without introducing catastrophic linguistic or attentional artifacts.

The core challenge confronting Wimmer and Perner was fundamentally experimental: how could an experimenter present a complex social narrative involving nested intentional states (Subject A knows that Subject B believes X, while the child knows that Y is the case) to a preschool-aged child in a manner that was intuitively comprehensible, completely transparent in its physical causal sequence, and free from confounding linguistic ambiguity? If the verbal instructions or memory demands were too burdensome, children would fail not because they lacked the conceptual understanding of mental representation, but because their executive resources had collapsed under the narrative weight of the prompt.

Wimmer and Perner systematically designed an empirical methodology built around visual enactments using physical props, miniature figurines, and high-contrast, color-coded hiding locations. The protocol was engineered to clearly isolate physical reality from the subjective epistemic access of the agent. By requiring the child to track an agent who becomes physically absent at a pivotal moment, the experiment created an epistemic divergence: the child became privy to an objective change in the physical world, whereas the absent agent remained causally and perceptually isolated from that change. To eliminate linguistic confusions, Wimmer and Perner instituted strict linguistic controls, standardizing comprehension checks and memory probes to verify that children remembered both the initial conditions and the subsequent physical transformation before answering the critical test question.

2.2 The Seminal 1983 Publication: ‘Beliefs About Beliefs’

The resulting empirical breakthrough was published in the journal Cognition in 1983, under the title “Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception”. This paper is universally acknowledged as the foundational document of empirical Theory of Mind research. In this study, Wimmer and Perner introduced the world to the “unexpected-transfer paradigm,” operationalized through the now-legendary narrative of Maxi and the Chocolate.

The experimental architecture proceeded with elegant clarity:
Children were introduced to a scenario enacted with small dolls and a mock kitchen setting. The protagonist, a boy named Maxi, assists his mother in unpacking groceries and places a bar of chocolate into a distinctive green cupboard. Maxi then leaves the kitchen to play in the garden. While Maxi is entirely absent and unable to witness the kitchen events, his mother uses some of the chocolate for baking and subsequently returns the remaining chocolate not to the green cupboard, but to a blue cupboard. Maxi then returns to the kitchen, feeling hungry and wanting his chocolate. The critical test question is posed to the participating child: “Where will Maxi look for his chocolate?”

The empirical findings revealed a stark, developmental dividing line that sent shockwaves through the cognitive sciences:
Children aged three years systematically failed the task. Overwhelmingly, three-year-olds pointed directly to the blue cupboard—the actual, current location of the chocolate in objective reality. Even though memory control probes confirmed that these three-year-olds remembered that Maxi had originally placed the chocolate in the green cupboard, and that he had not seen his mother move it, they were utterly incapable of attributing to Maxi a mental representation distinct from the physical truth they themselves witnessed. Conversely, between the ages of four and five, a decisive conceptual transition emerged: children reliably identified the green cupboard, demonstrating that they could predict Maxi’s behavior based on his subjective, outdated, and false belief rather than the objective state of the world.

The theoretical implications drawn by Wimmer and Perner were revolutionary. The four-year-old child had ceased to be merely a direct tracker of physical states; they had become a mentalizer. They had developed a metarepresentational capacity—the ability to hold two conflicting models of reality in mind simultaneously: the primary representation of where the chocolate truly is, and the secondary/metarepresentation of where Maxi *believes* it to be.

2.3 Immediate Reception and Methodological Repercussions

The publication of Wimmer and Perner’s 1983 paper acted as an immediate catalyst, sparking an explosion of empirical and theoretical inquiry across developmental psychology, cognitive anthropology, linguistics, evolutionary biology, and the philosophy of mind. The unexpected-transfer false-belief task was rapidly adopted globally as the gold standard, objective litmus test for social cognitive maturity. It offered researchers an operational tool of extraordinary elegance: a binary, clear-cut behavioral response that appeared to demarcate a qualitative transformation in human cognitive architecture.

The immediate reception, however, was accompanied by intense theoretical friction. Radical constructivists and traditional Piagetians argued that Wimmer and Perner had merely charted a transition in verbal reasoning or general executive decentering. Early nativists, such as Alan M. Leslie, contended that the false-belief task did not capture the *acquisition* of a mentalizing concept from scratch, but rather reflected the maturation of an innate “Theory of Mind Mechanism” (ToMM) whose behavioral expression was merely unmasked as inhibitory processing matured. Sociolinguists and cultural psychologists questioned whether the four-year milestone was an artifact of Western conversational norms, syntax, or formal schooling.

Crucially, Wimmer and Perner’s paradigm fundamentally relocated developmental psychology from the periphery of cognitive science directly to its epistemological core. Developmental psychology was no longer viewed simply as the study of how children grow; it had become an indispensable empirical laboratory for addressing classical philosophical dilemmas regarding the nature of intentionality, belief-desire psychology, propositional attitudes, and the cognitive architecture required to represent misrepresentation.

3. ‘Understanding the Representational Mind’ (1991): Foundational Theoretical Framework

3.1 The Tripartite Architecture of Representation

While the 1983 paper established the empirical gold standard for Theory of Mind, Josef Perner recognized that the psychological sciences lacked a comprehensive, logically rigorous theory explaining *what* a representation actually is and how its structural properties evolve across ontogeny. To provide this foundational architecture, Perner published his magnum opus in 1991: Understanding the Representational Mind (MIT Press / Bradford Books). In this landmark monograph, Perner articulated a tripartite taxonomy of representational development that remains one of the most sophisticated theoretical frameworks in cognitive psychology.

Perner delineated three distinct levels of cognitive architecture:

• Level 1: Primary Representations (Early Infancy, ~0 to 18 months). At this initial stage, the infant’s cognitive system relies on direct representations that serve as real-time, single-updating models of the immediate physical world. A primary representation maintains an isomorphic mapping to current reality; it functions to track the environment and guide direct motor actions. Crucially, primary representations lack detachment. If the environment changes, the primary representation is overwritten. The infant cannot represent alternative states of affairs simultaneously; they live in a cognitive universe defined entirely by a single model of the “here and now.”
• Level 2: Secondary Representations (Toddlerhood, ~18 months to 4 years). Emerging alongside symbolic play and early language, secondary representations are “decoupled” or hypothetical models of situations. Children at this level can construct multiple models simultaneously. They can represent past events, imagine future possibilities, entertain counterfactual scenarios, and engage in sophisticated pretend play (e.g., treating a banana as a telephone receiver) without confusing the pretend scenario with the real state of affairs. However, Perner emphasizes that children at this level are situation theorists. They represent multiple alternative situations, but they do *not* represent the structural relationship between a representation and the reality it represents. They understand that a situation can be different from reality, but they cannot comprehend how a *mental state* can serve as an inaccurate vehicle of that reality.
• Level 3: Metarepresentations (Representational Theory of Mind, ~4 years onward). The definitive watershed occurs when the child moves from being a situation theorist to a mental state representationalist. At this level, the child does not merely construct an alternative model of a situation; they explicitly represent the representational relation itself. They understand that minds contain representations (beliefs, thoughts, percepts) that refer to the world, and that the mental representation (the vehicle) can misalign with the state of affairs it purports to depict. This is the metarepresentational threshold required to pass the false-belief task.

3.2 The Medium-Content Distinction in Mental Life

At the very heart of Perner’s 1991 theoretical framework is the mastery of the medium-content distinction (often formulated as the distinction between the vehicle and the propositional content of a representation). For an organism possessing only primary or secondary representations, an internal model is simply an transparent window onto a state of affairs; the child looks *through* the representation directly at the situation being represented. To achieve a mature representational theory of mind, the child must make the cognitive vehicle itself an object of thought.

Perner illustrated this cognitive leap by analyzing the structural nature of misrepresentation. To understand that someone holds a false belief, a child cannot simply conclude that the person is focused on an alternative reality. The child must grasp three interrelated ontological facts:

• There is an objective reality (e.g., the chocolate is in the blue cupboard).
• There is an internal representational medium or vehicle (e.g., Maxi’s mental belief state).
• The propositional content instantiated within that medium (“The chocolate is in the green cupboard”) asserts a condition that fails to correspond with objective reality.

To contextualize this transition, Perner incorporated Gottlob Frege’s classical philosophical distinction between Sinn (sense or mode of presentation) and Bedeutung (reference or semantic value). Prior to age four, children operate purely at the level of reference; they evaluate statements and thoughts solely based on whether they correctly correspond to the referent in the world. Around age four, children develop an explicit appreciation of sense—they realize that a single referent can be mentally represented under multiple, mutually incompatible modes of presentation. This breakthrough enables the child to resolve referential opacity and understand that Maxi’s action is governed not by the referent (the true location of the object) but by the specific sense through which Maxi internally represents it.

3.3 Philosophical Synthesis: Reconciling Brentano, Pylyshyn, and Dennett

What distinguished Perner from his contemporaries was the profound philosophical depth underlying his cognitive modeling. In *Understanding the Representational Mind*, Perner achieved a brilliant synthesis of three major philosophical traditions: the intentionality theory of Franz Brentano, the computational cognitive science of Zenon Pylyshyn, and the intentional systems theory of Daniel Dennett.

From Franz Brentano, Perner adopted the foundational concept of *intentionality*—the defining hallmark of mental phenomena as being directed toward an object (“aboutness”), characterized by what Brentano termed “intentional inexistence.” Perner recognized that the developmental challenge facing the young child is precisely the problem of learning to comprehend intentionality: how can an internal thought be *about* a state of affairs that does not physically exist (such as a false belief or an unrealized goal)?

To ground Brentano’s phenomenological insights within modern computational cognitive science, Perner engaged deeply with Zenon Pylyshyn’s computational architectures. Pylyshyn had argued that cognitive systems operate over syntactically structured mental codes. Perner translated this into developmental terms, demonstrating that the child’s cognitive transition at age four represents an upgrade in their computational architecture—specifically, the emergence of a cognitive parsing mechanism capable of tagging representational structures as representational, thereby preventing “representational leakage” between hypothetical models and primary reality.

Finally, Perner operationalized Daniel Dennett’s philosophy of the “intentional stance.” Dennett argued that predicting the behavior of complex systems is best accomplished by treating them as rational agents whose actions are guided by intentional states (beliefs and desires). Perner demonstrated that the human child does not spontaneously adopt the full intentional stance at birth. Instead, the child begins with what Perner later termed a “teleological stance”—explaining action through objective, real-world constraints—before upgrading between ages three and four to Dennett’s full intentional stance, wherein internal, subjective mental representations are attributed to explain why an agent behaves in ways that are objectively irrational or maladaptive.

4. The Mechanics of the False-Belief Task: Variations, Criticisms, and Replications

4.1 Paradigm Variants: Maxi, Sally-Anne, and the Deceptive Box

Following Wimmer and Perner’s initial 1983 demonstration, the unexpected-transfer paradigm underwent rapid experimental refinement to maximize diagnostic clarity, lower unnecessary cognitive overhead, and test cross-contextual stability. The most iconic modification arrived in 1985, when Simon Baron-Cohen, Alan M. Leslie, and Uta Frith adapted Perner’s Maxi design into the streamlined Sally-Anne task to investigate cognitive profiles in children with autism. In this variant, Sally places a marble into her basket and leaves the room; Anne takes the marble and places it into her box; Sally returns, and the child is asked the canonical belief question: “Where will Sally look for her marble?” The Sally-Anne design simplified the narrative pacing, utilizing two female doll protagonists and highly distinctive containers (a basket with a flap versus a closed box), making the scenario accessible to clinical populations while preserving Perner’s essential experimental logic.

In parallel, Josef Perner, along with Susan Leekam and Heinz Wimmer, recognized that the unexpected-transfer paradigm tested only one specific dimension of false belief: the misattribution of an external spatial location. To evaluate whether the representational deficit was truly domain-general across all epistemic contexts, they developed the unexpected contents task (frequently referred to as the “Smarties task” or deceptive box task; Perner, Leekam, & Wimmer, 1987). In this paradigm, a child is presented with a familiar, branded candy box (such as a Smarties box) and asked what it contains. The child naturally replies, “Sweets” or “Smarties.” The experimenter then opens the box, revealing to the child’s surprise that it actually contains something completely unexpected, such as pencils. The experimenter closes the box and poses two critical questions:

• Other-Belief Question: “When your friend Nicky comes in, what will *she* think is inside the box before we open it?”
• Self-Belief (Epistemic Hindsight) Question: “When you first saw this box, before we opened it, what did *you* think was inside?”

The results mirrored the unexpected-transfer findings with stunning fidelity. Three-year-old children demonstrated a double failure: they claimed that Nicky would know the box contained pencils, and they shockingly claimed that *they themselves* had always known it contained pencils, completely erasing their own immediately preceding false belief. By age four to five, children succeeded on both fronts. The deceptive box paradigm proved that the four-year transition was not a quirk of spatial tracking, but a generalized cognitive breakthrough in understanding that *any* mental vehicle—whether belonging to oneself or another—can instantiate an erroneous propositional content.

4.2 Linguistic and Performance Artifacts: The Competence-Performance Debate

As the false-belief task became ubiquitous, an intense methodological counter-reaction emerged. Critics such as Michael Chandler, Michael Siegal, and Paul Bloom questioned whether three-year-olds genuinely lacked the *competence* to understand mental representation, or whether their widespread failure was merely a *performance artifact* driven by the linguistic complexity and pragmatic traps inherent to experimental testing. Siegal argued that when an adult asks a three-year-old, “Where will Maxi look for his chocolate?”, the child interprets this through conversational implicature as a request for where the chocolate actually is, or where Maxi *ought* to look if he wishes to succeed in finding it.

Josef Perner met this competence-performance challenge with rigorous empirical counter-experiments. He and his colleagues systematically altered the syntactic, narrative, and pragmatic framing of the test questions. In exhaustive control studies, Perner contrasted standard prompts with highly explicit linguistic queries, such as: “Where will Maxi look *first* for his chocolate?” or “Which place will Maxi go to to find his chocolate?” Perner demonstrated that while linguistic scaffolding, prominent visual markers, and pragmatic clarity could slightly reduce peripheral errors, they did *not* eliminate the profound developmental gulf. Three-year-olds continued to systematically default to the objective physical reality, whereas four-year-olds maintained robust metarepresentational accuracy.

Moreover, Perner instituted stringent control questions within every paradigm:
A reality question (“Where is the chocolate right now?”) and a memory question (“Where did Maxi put the chocolate in the beginning?”). Perner proved that three-year-old task failure occurred even when children exhibited flawless performance on both reality and memory probes. This confirmed that the deficit was neither an amnesic collapse regarding the initial state nor a misunderstanding of current reality, but an authentic, structural inability to bridge the gap between subjective misrepresentation and physical fact.

4.3 Implicit versus Explicit False-Belief Understanding

The debate surrounding Theory of Mind entered a dramatic new phase in the late 1990s and 2000s with the development of non-verbal, implicit testing paradigms. A landmark study by Perner himself, conducted with Wendy Clements in 1994 (Clements & Perner, 1994), revealed a fascinating developmental dissociation. Clements and Perner tracked children’s spontaneous anticipatory looking while they watched an unexpected-transfer scenario enacted on a stage. Astonishingly, children as young as two years and eleven months spontaneously looked toward the original, correct location (the green cupboard) *before* the protagonist emerged, anticipating his arrival at the site of his false belief. Yet, when asked the standard explicit verbal question (“Where will he look?”), these exact same children pointed directly to the true physical location (the blue cupboard).

This dissociation was pushed to radical extremes in 2005 by Kristine Onishi and Renée Baillargeon, who utilized a violation-of-expectation (VOE) looking-time paradigm to claim that infants as young as 15 months old possess an implicit understanding of false beliefs. In their study, infants looked significantly longer when an actor searched in a location that contradicted her previous perceptual access, which Onishi and Baillargeon interpreted as proof of an innate, fully functional Theory of Mind module operating in early infancy.

Josef Perner mounted a rigorous, theoretically profound resistance against these nativist interpretations. Perner argued that equating looking-time discrepancies in infants with a genuine conceptual Theory of Mind represents a fundamental category error. In a series of influential theoretical critiques, Perner distinguished between sub-personal, behavioral tracking and personal-level intentional understanding. He argued that infants rely on low-level perceptual heuristics, novelty preferences, or what he termed “teleological behavior-tracking rules”—such as associating agents with specific objects and spatial pathways based on statistical regularities. Perner maintained that looking-time tasks reflect perceptual expectations within a single reality model, whereas the explicit false-belief task uniquely requires the cognitive system to represent the medium-content distinction of a propositional attitude. For Perner, confusing perceptual anticipation with metarepresentation misses the very essence of what makes human mentalizing unique.

5. Representational Development: From Direct Reference to Metarepresentation

5.1 Pretense, Imagination, and Dual Representation

One of the most fiercely contested battlegrounds in cognitive development during the late 1980s and 1990s centered on the cognitive architecture of toddler pretend play. In 1987, Alan M. Leslie published a highly influential paper in Psychological Review arguing that the emergence of pretend play at 18 months of age is the primary evolutionary manifestation of an innate Theory of Mind Mechanism (ToMM). Leslie posited that when a toddler pretends that a dry sponge is a cake, or that a plastic banana is a telephone, their cognitive system deploys a specialized computational device—a “decoupler”—that copies the primary representation (“This is a sponge”), strips it of its normal truth relations, and outputs a metarepresentation (“I pretend this sponge is a cake”) to protect the child’s primary knowledge base from semantic corruption.

Josef Perner launched a devastating theoretical and empirical critique of Leslie’s decoupler model. In *Understanding the Representational Mind*, Perner demonstrated that Leslie had conflated a secondary representation of an alternative situation with a metarepresentation of a mental state. Perner showed that pretense does not require the child to represent a mental representation as such; rather, the pretending toddler simply acts within an imagined, hypothetical scenario. When a two-year-old pretends a banana is a telephone, they are not contemplating the epistemic mechanics of their own mental vehicle; they are mentally manipulating an alternative *situation* in which an object possesses the functional affordances of a telephone.

Perner’s model elegantly accounted for why two-year-olds excel at vibrant pretend play yet fail miserably on false-belief tasks for two more years. Pretense operates at Level 2 (secondary representation of hypothetical states of affairs), whereas false belief operates at Level 3 (metarepresentation of the relationship between an internal vehicle and the world). Pretend play serves as a vital developmental bridge: by allowing the toddler to practice manipulating decoupled models of reality without confusing them with current perceptual inputs, secondary representation builds the operational workspace that will ultimately be subordinated to metarepresentational analysis at age four.

5.2 The Mastery of Misrepresentation across Non-Mental Domains

A central tenet of Perner’s theoretical architecture is that the cognitive watershed occurring around age four is not a domain-specific social heuristic, but a domain-general representational transformation. If Perner’s model is correct, the young child’s difficulty with false belief does not stem from a localized inability to read social cues or decipher human emotions; rather, it reflects a structural limitation in processing *misrepresentation* as an abstract semiotic and epistemological relation. Consequently, children should exhibit identical developmental trajectories when resolving misrepresentation in non-mental, physical artifacts.

To test this prediction, developmental researchers introduced the false-photo task, originally pioneered by Deborah Zaitchik in 1990 and extensively investigated by Perner and Susan Leekam. In the false-photo paradigm, an experimenter takes an instant Polaroid photograph of a physical scene—for example, a teddy bear sitting on a red chair. The photograph is placed face down on a table to develop. While the photograph is developing, the experimenter moves the teddy bear from the red chair to a blue sofa. The child is then asked: “In the picture, where is the bear sitting?”

The structural parallels between the false-belief task and the false-photo task are mathematically exact:
Both paradigms present an outdated, physical or mental representation (the photograph / the agent’s belief) that accurately depicted reality at Time 1, but which misrepresents reality at Time 2 due to an unrecorded physical transformation. In a landmark series of comparative studies, Perner demonstrated that three-year-old children fail the false-photo task at virtually the identical developmental rate that they fail the false-belief task. When asked about the photo, three-year-olds point to the blue sofa, asserting that the photograph depicts the object’s current physical location in reality. Only around age four do children successfully grasp that the photograph represents a historical, outdated state of affairs distinct from current reality. This profound synchrony provided empirical verification for Perner’s claim: the four-year milestone marks the emergence of a domain-general capacity to parse the medium-content distinction across mental and physical media alike.

5.3 Perspective-Taking and Dual-Aspect Objects

To map the precise mechanics of how children transition from tracking physical objects to coordinating representational perspectives, Perner built upon and fundamentally expanded John Flavell’s classical taxonomy of visual perspective-taking. Flavell had distinguished between two levels of spatial cognition:

• Level 1 Perspective-Taking (~2.5 to 3 years): The ability to understand that another person may not see an object that the child sees (or vice versa), governed by lines of sight and occlusion. At Level 1, the question is binary: Is the object visible or invisible to the observer?
• Level 2 Perspective-Taking (~4 to 5 years): The ability to understand that an identical, mutually visible object presents a fundamentally different spatial appearance or aspect when viewed from different physical vantage points. At Level 2, the question is qualitative: How does the mutually visible object appear to the other observer?

Josef Perner recognized that Level 2 perspective-taking represents an exact structural analog to false-belief understanding. To demonstrate this, Perner conducted a series of elegant experiments examining children’s comprehension of dual-aspect objects. In one famous paradigm, children were shown a drawing of a turtle resting flat on a table between the child and the experimenter. From the child’s vantage point, the turtle was oriented right-side up; from the experimenter’s opposite vantage point across the table, the exact same turtle appeared upside down.

Perner discovered that three-year-olds, despite knowing that the experimenter was looking at the drawing, stubbornly maintained that the experimenter saw the turtle right-side up. The three-year-old system could track that the experimenter saw the *referent* (Level 1), but could not coordinate two distinct representational *senses* or aspects of that identical referent simultaneously (Level 2). Perner showed that success on Level 2 perspective-taking tasks correlates robustly with passing both the false-belief task and the false-photo task. The cognitive synchronization required to comprehend alternative perspectives on an identical referent is functionally identical to the cognitive machinery that resolves false belief: both demand the metarepresentational capacity to isolate the vehicle of representation from its worldly object.

6. The Intersection of Executive Function and Theory of Mind

6.1 Empirical Correlations: Inhibitory Control and Mental State Attribution

Throughout the 1990s, one of the most robust, highly replicated findings in developmental psychology was the intimate empirical correlation between the emergence of executive functions (EF)—specifically inhibitory control, working memory, and cognitive flexibility—and the acquisition of a Theory of Mind. Children who display advanced performance on executive batteries consistently pass false-belief tasks earlier than their peers, a relationship that holds true across cultures, socioeconomic strata, and linguistic backgrounds.

A primary vehicle for exploring this synchrony was the Dimensional Change Card Sort (DCCS) task, developed by Philip David Zelazo. In the DCCS, children are presented with cards displaying colored shapes (e.g., red rabbits and blue boats) and asked to sort them into target trays. In the first phase (“color game”), children sort the cards by color (reds here, blues there). In the second phase (“shape game”), the sorting rule is switched: children must now sort the cards by shape (rabbits here, boats there). Strikingly, three-year-old children exhibit perseverative sorting: they state the new rule aloud with absolute accuracy, yet when handed a card, they continue to sort it according to the old, previously rewarded rule. Exactly around age four, children overcome this perseveration, smoothly switching rules.

Josef Perner, collaborating with Douglas Frye, Philip Zelazo, and later Marion Lang, conducted extensive empirical investigations charting the developmental synchrony between the DCCS and false-belief mastery (Perner & Lang, 1999). Their studies revealed that passing the DCCS and passing the unexpected-transfer task occur within the exact same developmental window. Furthermore, longitudinal analyses confirmed that this relationship is not a general maturational artifact driven by raw chronological age or general verbal IQ; rather, it reflects a deeply coupled, functional co-evolution between executive control networks and mentalizing abilities.

6.2 Theoretical Accounts of the Executive-Mentalizing Nexus

While the empirical correlation between executive function and Theory of Mind was uncontested, a fierce theoretical debate erupted over the *causal direction* and *mechanistic nature* of this link. Two dominant, competing hypotheses emerged in the cognitive literature:

• The Executive Demand Hypothesis (Expression Account): Championed by Alan Leslie and colleagues, this view asserted that Theory of Mind competence is fully formed in early infancy as an innate module. However, the standard false-belief task imposes heavy inhibitory demands: because the true location of the object (e.g., the blue cupboard) is highly salient to the child, the child must actively *inhibit* their prepotent impulse to state current reality in order to express their underlying knowledge of the agent’s false belief. In this view, executive maturation merely serves as an “expression tool,” removing an inhibitory bottleneck to reveal a pre-existing social concept.
• The Cognitive Complexity Hypothesis (Emergence Account): Advanced forcefully by Josef Perner, Douglas Frye, and Philip Zelazo, this model rejected the notion that executive function is merely an external performance filter. Instead, Perner argued that executive function and false-belief understanding share an identical underlying logical and hierarchical structure.

Perner conceptualized this shared structure as the mastery of hierarchical conditional rules (an embedded “if-if-then” cognitive calculus). To pass the DCCS, a child must formulate a higher-order setting rule: “If we play the color game, then if red, put here; but if we play the shape game, then if red, put there.” Exactly mirroring this logical calculus, the false-belief task requires an embedded epistemic conditional: “Under reality, the object is at location B; but under Maxi’s perspective, the object is at location A.” Perner demonstrated that what matures around age four is not merely the muscular ability to suppress an action, but the conceptual capacity to build hierarchical representational systems capable of coordinating mutually exclusive conditional rules under a unifying higher-order frame.

6.3 Working Memory and Cognitive Flexibility in Belief Updating

Beyond inhibitory suppression, Perner’s empirical work highlighted the vital role of working memory buffers in managing the computational overhead of mentalizing. In a standard false-belief narrative, the cognitive architecture must execute a series of high-demand operations simultaneously: it must maintain the primary representation of current reality, encode the historical representation of the agent’s perceptual access, compute the divergence between the two, and generate a behavioral prediction based on the counter-factual model without allowing the primary model to overwrite the working memory trace.

Perner investigated how cognitive switching mechanisms operate when children are forced to continually update diverging epistemic states. When an agent acquires partial or misleading information, the child must evaluate not only *that* the agent is mistaken, but *the specific degree* to which the agent’s internal model aligns with reality. Perner’s research revealed that prior to age four, the working memory buffer is structurally incapable of maintaining two active, mutually contradictory truth-evaluations of the same proposition. When cognitive load is artificially increased, even older children can experience momentary executive regressions, defaulting to reality-congruent responses.

These theoretical insights have carried profound clinical implications. Perner’s work provided a unified framework for understanding the neurodevelopmental profiles of individuals with Attention-Deficit/Hyperactivity Disorder (ADHD), Frontal Lobe Traumatic Brain Injuries, and Autism Spectrum Conditions. By untangling whether a patient’s social-cognitive difficulties stem from a primary conceptual metarepresentational deficit or a secondary collapse in working-memory switching and inhibitory gating, clinicians and neuropsychologists have been able to design targeted cognitive remediation protocols that address the specific structural level of cognitive breakdown.

7. Counterfactual Reasoning and Causal Understanding in Childhood

7.1 The Structural Parallels Between ‘What Is’ and ‘What Might Have Been’

A defining hallmark of Josef Perner’s intellectual career is his refusal to confine Theory of Mind to a social-cognitive silo. In the late 1990s and 2000s, Perner turned his analytical attention to an adjacent domain of human rationality: counterfactual reasoning—the capacity to mentally alter a past event and deduce the causal consequences that would have unfolded under the hypothetical antecedent (reasoning about “what might have been”).

Perner formulated a bold theoretical hypothesis: counterfactual conditional reasoning and false-belief attribution are structurally identical. To assert, “If it had not rained, the picnic would have succeeded,” requires the mind to construct a hypothetical model that directly contradicts the known primary reality (it *did* rain, and the picnic was canceled), hold both models in consciousness simultaneously, and systematically map the causal relationships between them without confusing the hypothetical world with the actual world. This is the exact same computational architecture required to assert, “Maxi thinks the chocolate is in the green cupboard, but it is actually in the blue cupboard.”

To test this hypothesis, Perner and his collaborators designed ingenious experimental paradigms evaluating physical and epistemic counterfactuals in children aged three to six (e.g., Riggs, Peterson, Robinson, & Mitchell, 1998; Perner, Sprung, & Steinkogler). In a typical scenario, a firefighter doll named Peter is working in the garden, leaving his boots outside. A fire alarm sounds, and Peter rushes to the fire house, running straight inside wearing his socks. The child is asked two critical questions:

• Epistemic (False-Belief) Question: “Peter’s friend John wants to fetch Peter’s boots. Where does John think the boots are?”
• Physical Counterfactual Question: “If there had been no fire, where would Peter’s boots be?”

The empirical results confirmed Perner’s structural identity hypothesis with extraordinary statistical consistency. Children who failed the false-belief task systematically failed the physical counterfactual question, answering that Peter’s boots would be in the fire house (basing their answer on current reality). Passing the counterfactual conditional task occurred at the exact same developmental juncture (around age four) as passing the unexpected-transfer task. Perner demonstrated that the capacity to reason counterfactually about physical mechanics is functionally indistinguishable from the capacity to reason about erroneous mental states: both require the operational machinery of secondary and metarepresentations to manage decoupled possibilities.

7.2 Causal Attribution and the Calibration of Action Explanations

The deep structural symmetry between counterfactual reasoning and mentalizing naturally extends into the realm of causal attribution. How does a human agent explain *why* an event occurred? In classical philosophy of science, causal explanation requires the counterfactual test: event A caused event B if and only if, under the condition that A had not occurred, B would not have occurred (David Lewis’s counterfactual analysis of causation).

Perner investigated how young children calibrate causal chains when explaining human actions, particularly when those actions culminate in unexpected or maladaptive outcomes. If an actor walks into a mud puddle, a three-year-old typically provides a physical or teleological explanation: “He is in the mud because his shoes are dirty,” or “He went there to splash.” They rationalize the outcome by linking the physical state directly to a physical result or an objective goal. Conversely, a five-year-old provides an epistemic causal explanation: “He stepped into the mud because he *thought* the water was shallow,” or “He didn’t see how deep it was.”

Perner’s analysis of causal teleology proved that younger children do not perceive human action as driven by internal, mentalistic combustion; rather, they explain behavior through objective rationality. For a three-year-old, agents simply move along optimal paths toward desirable worldly targets. It is only when the child masters the counterfactual machinery of metarepresentation that they can introduce subjective epistemic causation into their explanatory repertoire: understanding that an actor’s behavior is caused not by the world as it actually is, but by the world as it was *represented* in the actor’s mind prior to the action.

7.3 Future-Oriented Thinking and Temporal Prospection

The cognitive mechanics that allow a child to reconstruct the past counterfactually are inextricably linked to the mechanisms that allow them to simulate the future. In collaboration with international colleagues, Perner explored the developmental emergence of temporal prospection—the capacity for episodic future thinking and mental time travel.

Perner demonstrated that episodic future thinking is not simply general planning or associative anticipation. An animal can associate a bell with food delivery, but this associative link does not constitute episodic simulation. True prospection requires an agent to construct a secondary representation of a hypothetical future self, experiencing a specific emotional and epistemic state within a decoupled scenario. In empirical experiments testing children’s capacity to select a tool today that will be needed in a novel, physically separated environment tomorrow, Perner and his team showed that genuine episodic prospection does not stabilize until the fourth year of life.

This work unified three seemingly disparate cognitive domains into a singular, breathtaking developmental timeline:
Between the ages of three and five, the human child simultaneously acquires epistemic Theory of Mind (representing false beliefs), counterfactual modal reasoning (representing what might have been), and episodic prospection (representing what could be in the future). Josef Perner demonstrated that these are not three isolated milestones, but three distinct phenomenological expressions of a solitary, foundational cognitive revolution: the emergence of the representational mind.

8. Episodic Memory, Mental Time Travel, and the Experiential Self

8.1 Re-Evaluating Tulving: Perner’s Epistemic View of Episodic Memory

Josef Perner’s theoretical contributions extended deeply into the cognitive neuroscience of memory. In 1972 and 1983, the legendary cognitive psychologist Endel Tulving had revolutionized memory research by drawing a fundamental distinction between semantic memory (the encyclopedic database of generalized world knowledge, such as knowing that Paris is the capital of France) and episodic memory (the capacity to mentally travel back in time and re-experience a personally lived event, characterized by autonoetic consciousness—the subjective feeling that “I personally experienced this”).

While Perner deeply admired Tulving’s taxonomy, he identified a critical conceptual gap: Tulving’s definition of episodic memory was heavily descriptive, emphasizing temporal tags and phenomenological feelings of “pastness,” but failing to explain the underlying epistemic architecture that makes autonoetic re-experiencing cognitively possible. In a series of groundbreaking papers published with Ted Ruffman and Johannes Roessler (e.g., Perner & Ruffman, 1995), Perner formulated an epistemic view of episodic memory.

Perner argued that episodic memory is not merely remembering an event with a time-stamp; it is fundamentally a metarepresentation of the acquisition event. To genuinely episodically remember seeing a red ball hidden under a box yesterday, it is not enough to possess the semantic proposition, “The ball is under the box.” The subject must represent the epistemic causal chain: “I know the ball is under the box because I remember *myself seeing* it being put there.” Episodic memory, according to Perner, is an act of epistemic source monitoring—it requires an explicit representation of the past self as a perceptual experiencing agent.

8.2 Source Monitoring and Metacognition of Past Experience

To prove this epistemic theory of episodic memory empirically, Perner and his colleagues devised rigorous experimental paradigms testing children’s capacity for source monitoring—the metacognitive ability to identify *how* one came to acquire a specific piece of factual knowledge. In these paradigms, preschool children were introduced to various hidden objects through three distinct informational modalities:

• Visual Perception: The child looked inside a drawer and saw an object (e.g., a toy car).
• Verbal Testimony: The experimenter told the child what was inside a box (e.g., “There is a blue ribbon inside”).
• Causal Inference: The child inferred what was inside an opaque container based on a physical clue (e.g., hearing a bell rattle).

Following the presentation, the children were asked two diagnostic questions: a direct knowledge question (“What is inside the container?”) and the critical source-monitoring question (“How do you know that? Did you see it, did I tell you, or did you figure it out?”).

The findings directly corroborated Perner’s model:
Three-year-old children successfully recalled the factual content (they knew the car was in the drawer), but they were utterly incapable of identifying the source of their knowledge. They frequently asserted that they had “seen” the object that had only been described to them, or claimed that they had “always known” what was hidden inside. Exactly around age four—in tight developmental lockstep with their performance on the unexpected-transfer false-belief task—children suddenly acquired the capacity for accurate source monitoring. Perner demonstrated that passing false-belief tasks and passing source-monitoring batteries are deeply correlated because both depend on the identical metarepresentational mechanism: the ability to explicitly represent the causal and epistemic relation between an experiential medium (seeing, hearing, inferring) and the resulting propositional knowledge.

8.3 Autonoetic Consciousness and the Temporally Extended Self

Perner’s epistemic framework provided a revolutionary cognitive explanation for one of the oldest mysteries in psychological science: infantile or childhood amnesia—the universal human inability to recall episodic memories from the first three to four years of life. Classical psychoanalytic theory had attributed this amnesia to the active repression of traumatic infantile desires, while early neurological accounts blamed the structural immaturity of the hippocampus and prefrontal cortex.

Josef Perner provided a cognitive-representational solution: infantile amnesia does not occur because early memories are violently repressed or entirely lost to biological rot; rather, genuine episodic memories cannot be encoded prior to age four because the child does not yet possess the metarepresentational architecture of the experiential self. Prior to age four, the toddler encodes vast amounts of functional, semantic, and procedural knowledge about the world, but they lack the autonoetic cognitive framework required to tag those experiences as “events that happened to ME as a perceiving epistemic agent.”

Around age four, the emergence of metarepresentation enables the construction of what Perner termed the temporally extended self. The child suddenly conceptualizes themselves as an enduring conscious entity that stretches backward into a personal past and forward into an anticipated future. By understanding that their *current* knowledge is causally derived from their *past* perceptual experiences, the child establishes the autobiographical memory system. Perner thus demonstrated that the cessation of childhood amnesia and the birth of Theory of Mind are two sides of the exact same cognitive coin: both mark the moment the human child discovers their own mind as an active, representing medium within the stream of time.

9. Teleology, Teleological Action Explanation, and Belief-Desire Psychology

9.1 Teleology versus Mentalism: The Developmental Shift

In the later stages of his career, beginning around 2005 and continuing through the 2010s, Josef Perner undertook a profound theoretical recalibration that challenged the prevailing orthodoxy of the entire Theory of Mind field. For decades, the dominant consensus—advanced by scholars such as Alison Gopnik, Henry Wellman, and early Perner himself—was that young children operate as intuitive “belief-desire psychologists” from toddlerhood, gradually fine-tuning their internal mentalistic axioms until they successfully incorporate false beliefs at age four.

In a series of landmark philosophical and empirical papers, many co-authored with the philosopher Johannes Roessler (e.g., Perner & Roessler, 2010, 2012), Perner broke decisively with this consensus. He introduced a radical distinction between mentalism (explaining action through internal, subjective mental representations) and teleology (explaining action through objective, real-world goals and facts). Drawing on György Gergely and Gergely Csibra’s work on the “teleological stance” in infancy, Perner argued that two- and three-year-old children are not primitive belief-desire psychologists at all; they are pure teleologists.

When a three-year-old observes an actor retrieving an apple from a tree, the child does not calculate an internal mental chain: “The actor possesses a subjective desire for fruit, combined with a subjective belief that the apple is edible, which generates an intentional motor command.” Instead, the child operates via objective teleology: the actor’s behavior is guided directly by the objective facts of the situation. The apple is objectively good to eat, the path to the apple is physically clear, and therefore the actor moves toward it because it is the *rational, objective thing to do in that environment*. Teleology bypasses internal mental states entirely, evaluating action through an external calculus of objective goodness-of-fit and physical affordances.

9.2 Teleology-in-Perspective: The Crucial Transitional Phase

If toddlers are pure teleologists who explain action through objective real-world facts, how do they navigate situations where different individuals pursue different goals or possess conflicting access to reality? To explain this developmental progression, Perner introduced a groundbreaking intermediate theoretical stage: Teleology-in-Perspective.

In the teleology-in-perspective phase (typically observed in three-year-olds), the child does not yet possess a representational Theory of Mind, but they have evolved beyond pure, universal teleology. They recognize that what is rational for an actor depends on the agent-relative physical affordances and perceptual exposure of that specific actor. For example, a three-year-old can easily understand that Daddy wants to eat the broccoli while Johnny wants to eat the cookie; they do not need to attribute a complex representational propositional attitude to explain this, but merely track individual subjective preferences as basic behavioral orientations toward objects in the world.

Similarly, Perner demonstrated that three-year-olds can calculate actions based on whether an agent has had physical contact with an object. However, this is still not mentalism. In teleology-in-perspective, the child views the agent as acting within a restricted *objective situation*, not acting upon an internal *mental misrepresentation*. The catastrophic failure at age three occurs precisely when the child encounters a situation where the agent’s behavior cannot be deduced from any objective fact in the physical world—namely, the false-belief task. In the unexpected-transfer scenario, there is no physical fact that can justify Maxi searching in the green cupboard; the chocolate is objectively in the blue cupboard. To explain Maxi’s irrational movement toward the empty green cupboard, teleology completely collapses. The child is forced to abandon teleology and embrace full-blown epistemic representationalism: discovering that agents act not on the basis of facts, but on the basis of internal representations that may be completely false.

9.3 Integrating Reason Explanations with Epistemic Reality

This profound theoretical shift allowed Perner to resolve a longstanding philosophical ambiguity regarding the nature of human action explanations. In contemporary philosophy of action (originating with G.E.M. Anscombe and Donald Davidson), there is a crucial ontological distinction between acting for a reason (rational, normative justification) and action driven by mental causation (the psychological mechanics of belief-desire pairs).

Perner, working closely with philosophical colleagues at the University of Warwick (including Johannes Roessler and Naomi Eilan), demonstrated that standard developmental psychology had systematically confounded normative teleology with representational belief. When an adult or child explains, “She took an umbrella because it was raining,” they are providing a reason explanation grounded in an objective worldly condition (the rain), not necessarily an introspective claim about an internal propositional attitude. The rain itself is the reason for her action.

Perner argued that human children begin life entirely within this normative, reason-based teleological framework. Children learn to navigate social life by mastering the normative rules that govern how people ought to interact with objective reality. Metarepresentational Theory of Mind is recruited as a late-emerging, specialized cognitive adaptation designed specifically to handle normative breakdowns. When an agent acts bizarrely, mistakenly, or counter-productively—such as looking for chocolate in an empty cupboard or taking an umbrella on a scorching sunny day—reason explanations grounded in worldly facts fail. It is at this precise explanatory failure point that the four-year-old cognitive system deploys the metarepresentational mind, inventing subjective mental errors to salvage human rationality in the face of physical error.

10. Comparative, Cross-Cultural, and Clinical Extensions (Autism and Beyond)

10.1 Theory of Mind in Neurodevelopmental Disorders: The Case of Autism

The empirical paradigms engineered by Josef Perner played an indispensable, historic role in cracking open the cognitive architecture of neurodevelopmental conditions, most notably Autism Spectrum Conditions (ASC). In 1985, Simon Baron-Cohen, Alan M. Leslie, and Uta Frith deployed their modified version of Perner’s unexpected-transfer paradigm (the Sally-Anne task) to test the revolutionary hypothesis that the core social-communicative impairments in autism stem from a specific, domain-isolated cognitive deficit: “mindblindness” (the inability to attribute mental states to self and others).

Collaborating with Frith and Baron-Cohen in subsequent comparative studies (e.g., Perner, Frith, Leslie, & Leekam, 1989), Perner helped establish the clinical specificity of this deficit. In rigorous empirical trials, children with autism were contrasted with neurotypical preschoolers and children with Down syndrome matched for mental age. The results were stark and definitive: while 85% of neurotypical four-year-olds and 86% of Down syndrome children successfully passed the false-belief task, over 80% of children with autism failed, stubbornly pointing to the marble’s true physical location despite possessing normal or superior chronological and non-verbal mental ages.

Crucially, Perner’s false-photo paradigm provided the decisive diagnostic control that confirmed the specificity of the autistic mentalizing deficit. When presented with the Zaitchik/Perner false-photo task, children with autism passed with flying colors: they easily understood that a Polaroid photograph depicted an outdated, physical state of affairs (the bear on the chair). This created a double dissociation of historic significance: children with autism possessed spared, often superior physical-causal and mechanical-representational reasoning, but suffered a profound, domain-specific impairment in computing social-epistemic metarepresentations. Perner’s paradigms thus provided the empirical bedrock for modern clinical cognitive neuroscience’s understanding of social neurodiversity.

10.2 Cross-Cultural Universality of the Representational Milestone

As Theory of Mind research expanded globally, a critical anthropological and linguistic question emerged: Was the four-year transition discovered by Wimmer and Perner in Salzburg merely an artifact of Western European bourgeois child-rearing practices, modern nursery schooling, and Indo-European linguistic structures? Or did it reflect a universal, biological-maturational milestone in the cognitive development of the human species?

To settle this question, Henry M. Wellman, David Cross, and Julanne Watson published a monumental quantitative meta-analysis in 2001 (Wellman, Cross, & Watson, 2001), analyzing 178 separate empirical studies encompassing over 4,000 children tested across diverse global populations, including North America, Western and Eastern Europe, East Asia, Indigenous Australian communities, and non-literate agrarian societies in Africa and South America. The meta-analysis established an unequivocal conclusion: the developmental shift from failing to passing explicit false-belief tasks between the ages of three and five is a culturally universal human phenomenon.

While the fundamental cognitive infrastructure was proved universal, cross-cultural studies (such as the landmark investigations conducted by Callaghan et al. in 2005 across Canada, India, Peru, Samoa, and Thailand) revealed fascinating variations in the rate of acquisition. Children exposed to rich maternal mental state talk (mothers who regularly discuss thoughts, desires, and perspectives) and children who grow up in households with older siblings consistently pass false-belief tasks months ahead of their peers. Josef Perner evaluated these cross-cultural and environmental variations not as contradictions to his model, but as proof of its representational nature: while the biological brain provides the executive workspace, social interaction, conversational friction, and linguistic scaffolding provide the rich epistemic data required for the child to actively construct the metarepresentational mind.

10.3 Comparative Primate Cognition: Do Non-Human Primates Possess a Theory of Mind?

The question that originally catalyzed the modern developmental movement—David Premack and Guy Woodruff’s 1978 query, “Does the chimpanzee have a theory of mind?”—remained a persistent theoretical beacon throughout Perner’s career. Armed with the rigorous operational distinctions he developed in *Understanding the Representational Mind*, Perner continuously evaluated the burgeoning literature in comparative primate cognition, engaging in profound critical dialogues with leading primatologists such as Michael Tomasello, Josep Call, and Brian Hare.

In the early 2000s, Tomasello, Call, and Hare conducted groundbreaking experiments demonstrating that subordinate chimpanzees selectively retrieve hidden food items that a dominant chimpanzee *cannot see* due to physical barriers, proving that chimpanzees can track visual lines of sight and Level 1 visual perspective. Furthermore, chimpanzees intentionally deceive competitors by sneaking food only when the competitor’s back is turned. Primatologists argued that these findings finally proved that non-human primates possess a Theory of Mind.

Josef Perner mounted an uncompromising, conceptually devastating philosophical critique of these claims. Applying the strict operational criteria of his representational framework, Perner demonstrated that the chimpanzee behavioral data can be fully, parsimoniously explained through teleology and behavior-tracking heuristics without attributing a single metarepresentation. A chimpanzee tracking whether a dominant ape has an unobstructed line of sight is not attributing a subjective mental percept; it is calculating an objective environmental relation: “Target is in direct line of sight of dominant = danger.”

Perner emphasized that despite decades of intensive experimental testing, non-human primates have never reliably passed an explicit, non-verbal false-belief task requiring the comprehension of misrepresentation. Chimpanzees excel as sophisticated, tactical teleologists who read behaviors, monitor spatial affordances, and track objective fields of visibility; however, they remain permanently stranded at Level 2 secondary situation modeling. For Perner, the capacity to form metarepresentations of mental vehicles—to comprehend that another creature acts upon an internal, erroneous representation of reality—is the unique, evolutionary Rubicon that separates the human species from the rest of the animal kingdom, serving as the cognitive foundation that made cumulative human cultural transmission, institutional reality, and advanced pedagogy uniquely possible.

11. Philosophical Collaborations and Epistemological Debates

11.1 Theory-Theory versus Simulation Theory: Perner’s Stance

Within the cognitive philosophy of mind, the discovery of Theory of Mind ignited a titanic intellectual war that raged throughout the 1980s and 1990s: the debate between Theory-Theory (TT) and Simulation Theory (ST).

• Theory-Theory: Advanced by scholars such as Alison Gopnik, Henry Wellman, and early cognitive functionalists, TT posited that human folk psychology consists of an internalized, theoretical system of nomological laws and abstract, unobservable constructs (beliefs, desires, intentions) that function identically to a formal scientific theory, undergoing radical conceptual revolutions across childhood development.
• Simulation Theory: Championed by philosophers such as Alvin Goldman, Robert Gordon, and Paul Harris, ST rejected the claim that humans deploy a theoretical conceptual apparatus to understand others. Instead, ST argued that mentalizing is executed through direct experiential simulation: an observer takes their own cognitive and emotional decision-making system “off-line,” feeds in the target agent’s initial parameters, runs a simulation using their own neurological machinery, and projects the resulting output onto the other person.

Josef Perner adopted a uniquely sophisticated, nuanced stance that exposed the fundamental logical flaws inherent to pure Simulation Theory. Perner argued that experiential simulation is structurally incapable of explaining false-belief attribution without already presupposing the very metarepresentational theoretical concepts it claims to replace.

Perner demonstrated that in order to “feed in the parameters” of an agent who holds a false belief, the simulating subject must deliberately insert an *untrue premise* into their off-line cognitive engine: “Simulate: The chocolate is in the green cupboard.” But how can the subject prevent this false premise from contaminating their own primary knowledge base, and how can they understand that this premise represents an *erroneous mental state belonging to someone else*, unless they already possess the theoretical concept of a mental representation? Perner showed that simulation can only function as an auxiliary operational heuristic *subordinate* to a prior, overarching representational theory. Without a conceptual grasp of the medium-content distinction, an off-line simulation simply produces confusion or reality-congruent default errors. Perner thus firmly established the conceptual supremacy of a representational theoretical framework over pure experiential simulation.

11.2 Collaborative Synergies with Epistemologists and Philosophers of Mind

Unlike many empirical psychologists who operate in isolation from contemporary philosophy, Josef Perner maintained deep, productive institutional partnerships with leading European epistemologists and philosophers of mind. A major epicenter of this interdisciplinary synthesis was the legendary Mind and Action Research Group at the University of Warwick, where Perner engaged in sustained, decades-long collaborative projects with philosophers including Johannes Roessler, Naomi Eilan, and Christoph Hoerl.

This interdisciplinary cross-pollination resulted in groundbreaking volumes such as Joint Attention: Communication and Other Minds (Oxford University Press). The Warwick-Salzburg collaboration directly addressed the deep epistemological problems connecting infant joint attention to childhood action explanation. Perner and Roessler painstakingly dismantled the barrier separating analytical philosophy’s conceptual analysis of propositional attitudes from developmental psychology’s experimental protocols.

Together, they explored the philosophy of action: How do young agents learn to navigate the normative commitments of intentionality? What does it mean for a child to understand a promise, a mistake, or a joint commitment? Perner demonstrated that empirical developmental laboratories provide an indispensable testing ground for testing the psychological reality of philosophical models. By showing that children understand reasons as objective worldly facts long before they understand them as subjective mental causes, Perner forced modern philosophy of mind to radically reassess its historical assumptions regarding folk-psychological rationality.

11.3 The Semantics of Identity and Reference: Intensionality vs. Extensionality

One of Josef Perner’s most intellectually dazzling contributions lies at the intersection of developmental psychology, formal logic, and the philosophy of language: the empirical investigation of referential opacity and the semantics of identity statements (the distinction between intensional and extensional contexts).

In classical formal logic, extensional contexts obey the principle of substitutivity of identicals (Leibniz’s Law): if Entity A is identical to Entity B, any true statement about A remains true if A is replaced by B. For example, if “The Morning Star” is physically identical to “The Evening Star” (both being the planet Venus), then if the statement “The Morning Star is a planet” is true, the statement “The Evening Star is a planet” is necessarily true. However, within mentalistic, propositional attitude contexts, this substitutivity violently collapses—the context becomes referentially opaque (intensional). If King Oedipus desires to marry Jocasta, and Jocasta is identical to his mother, it does *not* follow that Oedipus desires to marry his mother; internally, Oedipus represents Jocasta under the *sense* of “Queen of Thebes,” not under the *sense* of “biological mother.”

To determine when human children master the semantics of referential opacity, Perner conducted a series of brilliant experimental studies exploring what he termed the “Oedipus problem” in childhood (e.g., Perner, Rendl, & Garnham, 2007). In these experiments, children were introduced to dual-identity objects or characters—for example, a woman named Mrs. Jones who is simultaneously the town’s primary school teacher. The children were presented with a scenario where a character knows that Mrs. Jones is in the store, but has no idea that Mrs. Jones is the teacher. The child was then asked: “Does the character know that the teacher is in the store?”

Perner discovered that three-year-old children commit systematic extensional errors: because they know that Mrs. Jones *is* the teacher, they assert that the character knows the teacher is in the store, illegitimately substituting co-referential terms. Only around age four to five do children master the intensionality of mental states, realizing that knowledge and belief are opaque contexts governed by the specific representational *mode of presentation* through which an agent conceptualizes a referent. This work provided absolute empirical proof that the four-year milestone is fundamentally an epistemological revolution: the child masters Gottlob Frege’s distinction between sense and reference, discovering that human actions are guided not by the extensional identity of objects in the world, but by the intensional perspectives under which those objects are mentally represented.

12. Lasting Legacy, Critical Assessments, and Future Directions in Developmental Cognitive Science

12.1 The Replication Crisis and Methodological Re-Evaluation in Social Cognition

In the 2010s and 2020s, psychological science was engulfed by a massive, discipline-wide replication crisis. Open-science reproducibility initiatives, large-scale multi-lab consortia (such as ManyBabies), and rigorous pre-registered replication protocols began systematically stress-testing foundational findings across cognitive and developmental psychology. Amid this sweeping methodological reckoning, the field of social cognition faced a dramatic divergence in empirical stability.

On one side of this divide stood the non-verbal, implicit Theory of Mind paradigms in infants that had dominated headlines in the 2000s—most notably, the violation-of-expectation (VOE) and anticipatory looking tasks popularized by Onishi and Baillargeon (2005) and Southgate et al. (2007). In a devastating series of large-scale international replication attempts (e.g., Poulin-Dubois et al., 2018; ManyBabies consortia), researchers repeatedly failed to replicate the infant implicit false-belief looking-time effects. Multi-lab collaborations revealed that the infant data were plagued by low statistical power, high sample attrition, baseline gaze biases, and severe susceptibility to experimenter expectancy effects, throwing the entire theoretical enterprise of “infant nativist Theory of Mind” into profound crisis.

On the opposite side of the crisis, standing as an unshakeable monument of empirical stability, was Heinz Wimmer and Josef Perner’s classical 1983 explicit false-belief paradigm. Across hundreds of independent replications spanning five decades, dozens of cultures, and diverse linguistic families, the four-year developmental milestone charted by Wimmer and Perner has emerged as one of the most robust, highly reproducible, and rock-solid behavioral findings in the entire history of the behavioral sciences. Contemporary refinements utilizing high-speed computational eye-tracking, pre-registered computational modeling, and pupillometry have consistently vindicated Perner’s original core assertion: while low-level sensory-motor orientation and statistical behavior-tracking emerge early, the explicit, personal-level capacity to represent misrepresentation undergoes a genuine, qualitative, structural reorganization around the fourth year of life.

12.2 Neuroimaging Correlates and the Neural Substrate of Metarepresentation

As cognitive psychology embraced the functional neuroimaging revolution, the conceptual architectures articulated by Josef Perner provided cognitive neuroscientists with the precise functional blueprints required to map the “Social Brain Network.” Neuroimaging research (pioneered by Rebecca Saxe, Uta Frith, and Chris Frith) revealed that the execution of false-belief attribution consistently activates a highly specialized, interconnected neural circuit encompassing:

• The Temporoparietal Junction (TPJ), specifically the right TPJ (rTPJ), recognized as the core computational hub for transiently parsing perspective divergences and processing the medium-content distinction.
• The Medial Prefrontal Cortex (mPFC), heavily implicated in the longitudinal maintenance of social-normative schemas, decoupled hypothetical scenarios, and agent-specific intentional preferences.
• The Precuneus and Posterior Cingulate Cortex (PCC), serving as the central spatial-temporal engine mediating episodic memory retrieval and autonoetic mental time travel.

At the University of Salzburg’s Centre for Cognitive Neuroscience (CCNS), Josef Perner directly spearheaded landmark fMRI investigations mapping how these neural networks mature and interact across ontogeny. A major breakthrough of Salzburg’s neuroimaging program was demonstrating the profound neural overlap between episodic memory, counterfactual reasoning, and Theory of Mind. In rigorous functional neuroimaging trials, Perner and his neuroscience collaborators demonstrated that when adult and adolescent subjects process false beliefs, evaluate counterfactual conditionals (“What would have happened if…”), or retrieve autobiographical episodic memories of past learning events, they activate a shared, core neuroarchitectural network centered on the bilateral TPJ and mPFC.

Developmental neuroimaging studies further verified Perner’s theoretical model by charting the functional specialization of the rTPJ. In children younger than four, the rTPJ exhibits diffuse, unspecialized functional connectivity, responding identically to general social stimuli, biological motion, and physical transformations. Exactly around the fourth to fifth year of life—synchronous with the child’s behavioral triumph on the unexpected-transfer task—the rTPJ undergoes a dramatic functional reorganization, tuning its firing patterns specifically to the computational resolution of *representational misalignments*. Josef Perner’s cognitive model thus received direct, physiological confirmation: the birth of the representational mind is a biological reality written into the developing wiring of the human cerebral cortex.

12.3 Josef Perner’s Enduring Impact on Contemporary Cognitive Science

Josef Perner’s academic career stands as a monument to theoretical brilliance, methodological rigor, and interdisciplinary fearlessness. His election to the prestigious British Academy as a Corresponding Fellow in 2001, his induction into the Academia Europaea in 2008, and his election to the German National Academy of Sciences Leopoldina in 2011 reflect his standing as one of the most towering European psychological scientists of the modern era. The conferral of the William Thierry Preyer Award for Excellence in Developmental Psychology in 2011 and the prestigious Bielefeld Science Prize in 2012 formally recognized a lifetime of scholarship that permanently transformed human self-understanding.

Perner’s institutional legacy is permanently anchored at the Paris Lodron University of Salzburg. Through his leadership, the Salzburg Department of Psychology and the Centre for Cognitive Neuroscience evolved into a world-class institutional powerhouse, generating generations of leading international cognitive scientists who continue to push the frontiers of developmental psychology, social neuroscience, and cognitive philosophy. His academic lineage spans dozens of doctoral students and postdoctoral fellows who now occupy influential professorial chairs across Europe, North America, and Australasia.

Today, Perner’s theoretical framework is experiencing a dramatic, cutting-edge renaissance within the realms of Artificial Intelligence (AI), Multi-Agent Machine Learning, and Social Robotics. As computational engineers strive to move beyond large language models’ brute-force statistical pattern-matching toward building genuinely autonomous, socially intelligent artificial agents, they are running headfirst into the exact theoretical barriers that Josef Perner mapped decades ago. Computer scientists recognize that an artificial agent cannot achieve true common-sense social intelligence simply by tracking statistical associations or operating via low-level teleology. For an AI agent to safely collaborate with human beings, navigate ethical accountability, and understand human error, it must be endowed with a functional metarepresentational architecture: the capacity to maintain a medium-content distinction, resolve referential opacity, engage in counterfactual simulation, and explicitly represent the false beliefs of its human partners. In this vital computational frontier, Josef Perner’s *Understanding the Representational Mind* serves not merely as a historical record of child development, but as an indispensable architectural blueprint for the future of cognitive technology.

Conclusion

The scientific odyssey of Josef Perner represents one of the most triumphant chapters in the history of cognitive science. Prior to his groundbreaking interventions, the inner mental life of the young child was obscured by theoretical dogmatism—either dismissively categorized under Piagetian egocentrism or surrendered to unspecifiable biological nativism. Beginning with his and Heinz Wimmer’s historic 1983 unexpected-transfer false-belief paradigm, Perner cracked open the black box of childhood social cognition, transforming what was once an intractable philosophical debate into an exact, rigorously replicable empirical science.

Across more than forty years of relentless intellectual inquiry, Perner demonstrated that the child’s mastery of false belief at age four is not an isolated social heuristic, but the behavioral tipping point of a profound, domain-general representational revolution. Through his tripartite architecture of primary, secondary, and metarepresentation, Perner provided a unified cognitive taxonomy that simultaneously illuminated how the developing human mind unlocks the mechanics of pretend play, masterfully parses dual-aspect perspectives, navigates the hierarchical conditionals of executive control, constructs counterfactual worlds, and builds the autonoetic scaffolding of episodic memory and autobiographical consciousness. His later theoretical work on teleology and reason explanations radically dismantled simplistic mentalistic dogmas, demonstrating that human rationality begins not with complex mental state attributions, but with objective teleological engagement with worldly facts, reserving the representational mind as a specialized adaptation to resolve moments of human error.

Josef Perner permanently bridged the historical chasm separating European continental philosophy from Anglo-American empirical cognitive science. By operationalizing the insights of Franz Brentano, Gottlob Frege, and Daniel Dennett within the developmental nursery, he proved that the laboratory study of the young child is indispensable to solving the classical mysteries of intentionality, meaning, and human consciousness. In unveiling the structural machinery through which the human child discovers that minds can misrepresent reality, Josef Perner did not merely illuminate childhood development; he fundamentally decrypted the very cognitive architecture that defines what it means to be human.

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
• Callaghan, T., Rochat, P., Lillard, A., Kastarlak, P., Peskin, H., Corbit, J., Sharma, S., & Singh, S. (2005). Synchrony in the onset of mental-state reasoning: Evidence from five cultures. Psychological Science, 16(5), 378–384. https://doi.org/10.1111/j.0956-7976.2005.01544.x
• Clements, W. A., & Perner, J. (1994). Implicit understanding of belief. Cognitive Development, 9(4), 377–395. https://doi.org/10.1016/0885-2014(94)90012-4
• Eilan, N., Hoerl, C., McCormack, T., & Roessler, J. (Eds.). (2005). Joint attention: Communication and other minds: Issues in philosophy and psychology. Oxford University Press.
• Fodor, J. A. (1983). The modularity of mind: An essay on faculty psychology. MIT Press.
• 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
• Onishi, K. H., & Baillargeon, R. (2005). Do 15-month-old infants understand false beliefs? Science, 308(5719), 255–258. https://doi.org/10.1126/science.1107621
• Perner, J. (1991). Understanding the representational mind. MIT Press.
• Perner, J., Frith, U., Leslie, A. M., & Leekam, S. R. (1989). Exploration of the autistic child’s theory of mind: Knowledge, belief, and communication. Child Development, 60(3), 688–700. https://doi.org/10.2307/1130734
• Perner, J., & Lang, B. (1999). Development of theory of mind and executive control. Trends in Cognitive Sciences, 3(9), 337–344. https://doi.org/10.1016/S1364-6613(99)01362-5
• Perner, J., Leekam, S. R., & Wimmer, H. (1987). Three-year-olds’ difficulty with false belief: The case for a conceptual deficit. British Journal of Developmental Psychology, 5(2), 125–137. https://doi.org/10.1111/j.2044-835X.1987.tb01048.x
• Perner, J., Rendl, B., & Garnham, A. (2007). Objects of desire, thought, and reality: Problems of anchoring discourse referents in development. Mind & Language, 22(5), 475–513. https://doi.org/10.1111/j.1468-0017.2007.00318.x
• Perner, J., & Roessler, J. (2010). Teleology and causal reasoning in children’s theory of mind. Philosophical Psychology, 23(4), 519–538. https://doi.org/10.1080/09515089.2010.514569
• Perner, J., & Roessler, J. (2012). From infants’ to children’s appreciating false beliefs: The development of teleology-in-perspective. In J. Roessler, H. Lerman, & N. Eilan (Eds.), Perception, causation, and objectivity (pp. 249–276). Oxford University Press.
• Perner, J., & Ruffman, T. (1995). Episodic memory and autonoetic consciousness: Developmental evidence and a theory of childhood amnesia. Journal of Experimental Child Psychology, 59(3), 516–548. https://doi.org/10.1006/jecp.1995.1024
• Poulin-Dubois, T., Rakoczy, H., Burnside, K., Crivello, C., Dörrenberg, S., Edwards, K., Krist, H., Kulke, L., Liszkowski, U., Low, J., Perner, J., Powell, L. J., Priewasser, B., Ruffman, T., & Yott, J. (2018). Do infants understand false beliefs? We need to begin again. Monographs of the Society for Research in Child Development, 83(4), 1–173.
• 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
• Riggs, K. J., Peterson, D. M., Robinson, E. J., & Mitchell, P. (1998). Are errors in false belief tasks symptomatic of a broader difficulty with counterfactuality? Cognitive Development, 13(1), 73–90. https://doi.org/10.1016/S0885-2014(98)90021-1
• Tulving, E. (1983). Elements of episodic memory. Oxford University Press.
• 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
• Zaitchik, D. (1990). When representations conflict with reality: The preschooler’s problem with false beliefs and “false” photographs. Cognition, 35(1), 41–68. https://doi.org/10.1016/0010-0277(90)90036-J
• Zelazo, P. D., Frye, D., & Rapus, T. (1996). An age-related dissociation between knowing rules and using them in a card sort task. Cognitive Development, 11(1), 37–63. https://doi.org/10.1016/S0885-2014(96)90027-1