Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Embodied cognition and the neural basis of syntax: Reinterpreting Kant, Chomsky, and Humboldt Cognição corporificada e as bases neurais da sintaxe: Reinterpretando Kant, Chomsky e Humboldt Cognición encarnada y las bases neurales de la sintaxis: Reinterpretando a Kant, Chomsky y Humboldt Received: 09/06/2026 | Accepted: 09/09/2026 | Published: 09/11/2026 Giovanna Azevedo Rodrigues1 ORCID: https://orcid.org/0009-0003-9020-9368 ASMIGO Hospital, Brazil E-mail: giovannagirodrigues@hotmail.com Fernando Martins Castanheira Júnior ORCID: https://orcid.org/0009-0008-7612-9413 ASMIGO Hospital, Brazil E-mail: castanheirasms@gmail.com Marcelo Caixeta ORCID: https://orcid.org/0000-0003-2357-6558 Research Unit – UPC / HC/ UFGO- Clinical Research Unit, Brazil E-mail: cfm11041963@ufg.br Abstract This paper aims to argue that, under typical developmental conditions, syntactic productivity is grounded in embodied cognition, emerging primarily through the vocalization of human praxis within a causally structured world. The core claim is that the mapping from action structure to clausal structure follows a specific pattern: causal graph motifs in motor planning map onto predicate-argument templates in perisylvian networks via learned alignment. This mapping is operationalized by specifying how motor planning circuits scaffold combinatorial encoding in LIFG and LSTG. Nonaction semantics, including complementation, quantification, and abstract concepts, emerge through metaphorical extension and conventionalization of action-based templates. While alternative pathways exist, such as statistical learning from text, embodied praxis represents the typical ontogenetic pathway. The paper generates falsifiable developmental predictions, addresses cross-linguistic typology and clinical dissociations, and outlines implications for cognitive psychiatry. This perspective grounds intelligibility in the synchronization between body, action, memory, and objective causality. Keywords: Cognitive Science; Linguistics; Psychomotor Performance; Language Development; Mirror Neurons; Sign Language; Causality. Resumo Este artigo tem como objetivo defender que, em condições típicas de desenvolvimento, a produtividade sintática está fundamentada na cognição corporificada, emergindo primariamente por meio da vocalização da práxis humana em um mundo causalmente estruturado. A afirmação central é que o mapeamento da estrutura da ação para a estrutura oracional segue um padrão específico: motivos de grafos causais no planejamento motor mapeiam-se em modelos de predicadoargumento nas redes perissilvianas por meio de alinhamento aprendido. Esse mapeamento é operacionalizado especificando-se como os circuitos de planejamento motor sustentam a codificação combinatória no giro frontal inferior esquerdo (GFIE) e no giro temporal superior esquerdo (GTSE). A semântica não relacionada à ação, incluindo complementação, quantificação e conceitos abstratos, emerge por meio da extensão metafórica e da convencionalização de modelos baseados em ação. Embora existam vias alternativas, como a aprendizagem estatística a partir de texto, a práxis corporificada representa a via ontogenética típica. O artigo gera previsões desenvolvimentais falseáveis, aborda a tipologia interlinguística e as dissociações clínicas, e delineia implicações para a psiquiatria cognitiva. Essa perspectiva fundamenta a inteligibilidade na sincronização entre corpo, ação, memória e causalidade objetiva. Palavras-chave: Ciência Cognitiva; Linguística; Desempenho Psicomotor; Desenvolvimento da Linguagem; Neurônios Espelho; Linguagem de Sinais; Causalidade. 1 Corresponding author: giovannagirodrigues@hotmail.com. 1 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Resumen Este artículo tiene como objetivo sostener que, en condiciones típicas de desarrollo, la productividad sintáctica se fundamenta en la cognición encarnada, emergiendo principalmente a través de la vocalización de la praxis humana en un mundo causalmente estructurado. La afirmación central es que el mapeo de la estructura de la acción hacia la estructura oracional sigue un patrón específico: los motivos de grafos causales en la planificación motora se mapean sobre plantillas de predicado-argumento en las redes perisilvianas mediante un alineamiento aprendido. Este mapeo se operacionaliza especificando cómo los circuitos de planificación motora sustentan la codificación combinatoria en el giro frontal inferior izquierdo (GFII) y el giro temporal superior izquierdo (GTSI). La semántica no relacionada con la acción, incluyendo la complementación, la cuantificación y los conceptos abstractos, emerge mediante la extensión metafórica y la convencionalización de plantillas basadas en la acción. Aunque existen vías alternativas, como el aprendizaje estadístico a partir de texto, la praxis encarnada representa la vía ontogenética típica. El artículo genera predicciones evolutivas falsables, aborda la tipología interlingüística y las disociaciones clínicas, y esboza implicaciones para la psiquiatría cognitiva. Esta perspectiva fundamenta la inteligibilidad en la sincronización entre cuerpo, acción, memoria y causalidad objetiva. Palabras clave: Ciencia Cognitiva; Lingüística; Desempeño Psicomotor; Desarrollo del Lenguaje; Neuronas Espejo; Lenguaje de Señas; Causalidad. 1. Introduction In his influential work Kant and the Foundations of Analytic Philosophy, Robert Hanna identifies a convergence between Humboldt, Chomsky, and Kant. All three recognize that human minds are fundamentally productive rather than merely receptive. From finite means, the mind generates infinite outputs. This productive capacity represents a decisive break with behaviorism and empiricism. This convergence, however, carries a significant risk. By identifying a common thread, one risks obscuring crucial distinctions between phenomena that operate at fundamentally different ontological and neurophysiological levels. The productivity of generative grammar, the productivity of Kantian categories, and the productivity of spatiotemporal forms each possess distinct origins and functions. Conflating them under a single rubric of internal productivity obscures rather than illuminates their nature. This paper aims to argue that, under typical developmental conditions, syntactic productivity is grounded in embodied cognition, emerging primarily through the vocalization of human praxis within a causally structured world. Rather than proposing that all syntactic competence derives exclusively from motor praxis, we treat this as the predominant, though not exclusive, ontogenetic pathway, since alternative developmental routes to syntactic competence may exist. From this perspective, the syntax of language mirrors the syntax of action because both instantiate a common underlying organizational structure. Accordingly, cognition is conceptualized as being grounded in the dynamic integration of the body, action, memory, and objective causality. 2. Methodology A qualitative study (Risemberg et al., 2026a; Pereira et al., 2018) was conducted, taking the form of descriptive and reflective research (Risemberg et al., 2026b). This is a theoretical and integrative article; no empirical data were collected. The present framework was developed through a non-systematic, targeted review and synthesis of primary and secondary sources spanning four bodies of literature: (1) classical and contemporary philosophy of mind, particularly Kant's transcendental categories and Robert Hanna's comparative analysis of Kant, Humboldt, and Chomsky; (2) generative and usage-based linguistic theory, including debates between nativist and constructivist accounts of syntactic acquisition; (3) developmental and cognitive neuroscience literature on motor planning, mirror neuron systems, hierarchical predictive coding, and the neurobiology of sign language; and (4) embodied cognition research, including conceptual metaphor theory, consensus perspectives on situated and embodied language acquisition (Reggin et al., 2023), and critiques of strong embodied cognition. 2 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Sources were selected for theoretical relevance rather than through a systematic or exhaustive search protocol, consistent with the conventions of hypothesis-generating and theoretical-synthesis articles in cognitive science and philosophy of mind. Priority was given to foundational texts defining the theoretical positions under discussion (Chomsky, Humboldt, Hanna, Lakoff and Johnson) and to recent empirical and review literature (2021-2025) capable of constraining or informing the proposed model. The theoretical construction proceeded in three steps. First, the productivity arguments advanced by Hanna, Chomsky, and Humboldt were examined to identify points of convergence and the risk of conflating ontologically distinct forms of productivity. Second, a schematic correspondence was proposed between the hierarchical organization of goal-directed motor action and the combinatorial structure of clauses, informed by neuroscientific evidence on motor planning, perisylvian language networks, and predictive coding. Third, this correspondence was used to derive a set of falsifiable developmental, neurostimulation, and lesion-based predictions, formulated with heuristic effect-size benchmarks to facilitate future empirical testing. Clinical dissociations (aphasia-apraxia), cross-linguistic typology, and critiques of embodied cognition were then examined as boundary conditions against which the model's explanatory scope and limitations were assessed. 3. Results and Discussion 3.1 Productivity in Humboldt and Chomsky A central insight shared by Humboldt and Chomsky is that children routinely generate linguistic expressions they have never previously encountered. Explaining this generative capacity remains a fundamental challenge for theories of language acquisition (Hauser et al., 2002). Usage-based and constructivist frameworks, most prominently advanced by Tomasello and Goldberg, provide a compelling alternative to strong nativist accounts. Rather than positing an innate, domain-specific grammatical architecture, these approaches propose that children gradually construct grammatical competence through intentionreading, statistical learning (Saffran, Aslin, & Newport, 1996), and pattern abstraction, progressing from concrete, item-based constructions to increasingly productive linguistic representations. A simple illustration comes from verb morphology. After exposure to a limited set of past-tense forms, such as I left, I ran, and I felt, children readily produce novel inflected forms that they have never explicitly encountered. This productive capacity reflects the ability to extract abstract regularities from finite linguistic input and generalize them to new lexical items and contexts. Within the present framework, this capacity need not be attributed to an autonomous grammatical mechanism. Instead, it may reflect the brain's ability to identify, stabilize, and flexibly recombine structured regularities embedded in both linguistic experience and goal-directed human action through domain-general processes of statistical learning and distributional inference. Large language models present an important counterpoint. LLMs acquire sophisticated syntax from text-only training, without the developmental scaffolding provided by praxis. This does not falsify the claim that embodied praxis is the typical ontogenetic pathway. Rather, it suggests that syntax, once abstracted and conventionalized in a language community, can be learned through purely statistical means. LLMs represent an alternative, non-typical pathway to syntactic competence. 3.2 Operationalizing the Action-to-Syntax Mapping The central proposal of this framework is that language acquisition relies on a systematic mapping between the hierarchical organization of goal-directed action and the combinatorial structure of clauses. Specifically, recurrent causal motifs in motor planning are hypothesized to provide the scaffolding from which predicate-argument structures emerge during development (Glenberg & Kaschak, 2002). At a schematic level, this relationship can be represented as follows: Motor planning: Agent → Action → Patient → Instrument → Goal → Result Clausal structure: Subject → Predicate → Object → Adjunct → Complement → Modifier 3 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 This correspondence should not be interpreted as a one-to-one translation between motor and linguistic representations. Rather, it reflects a developmentally acquired alignment between action schemas and grammatical templates. Through repeated experience, hierarchical representations generated by motor planning systems become progressively associated with the combinatorial structures encoded by the language network. Within this framework, the left inferior frontal gyrus contributes primarily to the representation of hierarchical structure, whereas the left superior temporal gyrus supports the semantic content integrated within these structures. The proposed neural mechanism can be formalized in terms of hierarchical predictive representations (Friston, 2010; Pezzulo, Parr, & Friston, 2022). Premotor and supplementary motor cortices generate prospective action schemas organized according to nested causal relations, in which distal goals are recursively decomposed into subgoals and ultimately into elementary motor acts, each associated with predicted sensory consequences (Kim, Chua, & Hu, 2025; Tsay et al., 2024). Linguistic composition exhibits a comparable hierarchical organization, with clauses decomposing into phrases, phrases into words, and words into morphemes. Across development, repeated co-occurrence of action and language enables the brain to learn stable correspondences between these two hierarchical domains through statistical learning. These mappings are proposed to emerge from the coordinated activity of motor and perisylvian networks, with cerebellar learning mechanisms contributing to the calibration and refinement of the underlying distributed representations (Pulvermüller, 2005; Rizzolatti & Craighero, 2004; Carriere, Tomasello, & Pulvermüller, 2025). This framework yields a set of empirically testable predictions. First, transient disruption of motor planning circuits, for example through transcranial magnetic stimulation targeting the supplementary motor area, should selectively impair the processing of syntactically complex sentences describing novel goal-directed actions, while exerting comparatively weaker effects on the comprehension of abstract, non-action-related sentences. Second, developmental variation in fine motor control should predict subsequent variation in syntactic complexity, even after controlling for vocabulary size and general cognitive ability. Third, focal lesions involving motor planning networks should be associated with selective impairments in hierarchical syntactic operations, such as recursive embedding and the processing of nested dependencies, while relatively preserving other aspects of linguistic competence, including lexical-semantic knowledge and basic sentence comprehension. 3.3 Addressing Non-Action Semantics and Abstract Concepts A central challenge for any praxiological account of language concerns the emergence of abstract semantics. If syntax is grounded in the hierarchical organization of action, how do concepts such as belief, obligation, possibility, number, and time develop beyond the domain of overt motor behavior? The present framework proposes that abstract semantic systems emerge through the progressive metaphorical extension and conventionalization of action-based representational structures. As argued by Lakoff and Johnson, abstract concepts are systematically structured by embodied experience through conceptual metaphor. Domains such as time, causation, knowledge, and mental states are routinely understood in terms of more concrete sensorimotor experiences, allowing action-derived representational schemas to be generalized well beyond their original domain (Garello, Ferroni, Gallese, Ardizzi, & Cuccio, 2024). Within this framework, complementation and subordination arise from the recursive organization of action planning itself. Goal-directed behavior is inherently hierarchical: distal goals are achieved through nested sequences of subgoals, each of which may require its own internally structured action plan. Linguistic embedding reflects this organizational principle by allowing clauses to be recursively incorporated within other clauses. During development, children repeatedly experience hierarchical action organization before mastering analogous syntactic structures, enabling language to externalize preexisting patterns of nested causal organization. 4 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Quantification may similarly originate from the iterative structure of action. Goal-directed behaviors such as reaching, grasping, and object manipulation consist of repeated sensorimotor cycles through which children acquire stable representations of numerosity and cardinality. Number words subsequently provide symbolic labels for distinctions that are first established through recurrent interactions with discrete objects and events. Function words can likewise be interpreted as linguistic expressions of relational structures initially experienced through embodied interaction. Prepositions encode spatial relations that emerge from navigation and object manipulation; articles grammaticalize distinctions in reference, familiarity, and identifiability established during repeated interactions with objects; and auxiliary verbs encode aspectual, temporal, and modal properties of action, including distinctions between completed and ongoing events, as well as between realized, possible, and necessary actions. Importantly, this framework does not claim that all semantic knowledge is reducible to motor representations. Rather, it proposes that motor planning provides the initial compositional architecture upon which language is constructed. Once this hierarchical combinatorial system has emerged, semantic representations can progressively expand beyond the motor domain through conceptual metaphor, abstraction, cultural transmission, and linguistic conventionalization, ultimately supporting the full range of human conceptual thought. 3.4 Addressing Cross-Linguistic Typology A comprehensive theory of syntax must account for the remarkable diversity of grammatical systems observed across languages. At first glance, a praxiological account based on action structure might appear difficult to reconcile with typological phenomena such as ergative alignment, extensive applicative systems, noun incorporation, polysynthesis, or relatively free word order. The present framework accommodates this diversity by treating grammatical constructions as language-specific conventionalizations that selectively profile different aspects of a common underlying action schema. Cross-linguistic variation therefore reflects differences in how languages encode and highlight components of goal-directed events rather than differences in the fundamental organization of praxis itself. From this perspective, ergative-absolutive systems foreground the affected participant and the causal organization of events, whereas nominative-accusative systems privilege the initiating agent. Likewise, noun incorporation and polysynthetic morphology represent highly integrated linguistic encodings of actions and their participants, paralleling the unified representation of objects, goals, and motor sequences characteristic of skilled action planning. Variability in constituent order can be understood as alternative strategies for organizing information structure and discourse, while preserving the same underlying relational architecture among event participants. Signed languages provide particularly compelling evidence for the proposed action-language relationship. Although they rely on the visual-manual rather than the auditory-vocal modality, natural signed languages exhibit the full range of hierarchical and recursive syntactic organization found in spoken languages. Neuroimaging and lesion studies further demonstrate that sign language processing recruits largely the same left-lateralized perisylvian network that supports spoken language, despite the use of different sensory and motor effectors (Emmorey, 2021). These findings suggest that syntactic organization is not intrinsically linked to speech production, but instead depends on domain-general mechanisms for the hierarchical organization of motor representations. Within the present framework, signed languages therefore constitute strong evidence that syntax emerges from the structured organization of praxis rather than from vocalization itself. 3.5 Three Qualitatively Distinct Levels of Productivity Although Hanna's framework interprets productive cognition as manifestations of a single underlying principle, the present account argues that greater explanatory precision is achieved by distinguishing three qualitatively different levels of organization. 5 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Level 1: Generative Grammar. This level concerns the linguistic expression of thought through speech or sign. Its representational foundation lies in the externalization of hierarchically organized praxiological structures, enabling the productive generation of an effectively unbounded number of utterances from a finite set of lexical and combinatorial resources. Its neurobiological implementation is hypothesized to depend primarily on distributed interactions among perisylvian language regions, motor planning systems, sensory cortices appropriate to the linguistic modality, mirror neuron networks, and phonological or visuospatial working memory. Level 2: Kantian Categories. This level concerns the formal organization of cognition itself. Rather than expressing action, it involves the mind's capacity to represent, relate, and evaluate its own cognitive operations. Functions such as unification, differentiation, causal attribution, and inferential integration belong to this level. Its neural basis is expected to involve prefrontal executive systems, metacognitive networks, and large-scale association networks that integrate information across multiple cognitive domains. Level 3: Spatiotemporal Forms of Experience. This level corresponds to the embodied structure of primary sensibility. Its foundation lies in proprioceptive, vestibular, interoceptive, and sensorimotor experience, together with the temporal continuity provided by memory. These systems organize the body's initial engagement with the physical world, supplying the experiential framework within which both action and cognition develop. Candidate neural substrates include somatosensory cortices, cerebellar circuits, hippocampal memory systems, and distributed sensorimotor networks. These three levels should not be regarded as different expressions of a single computational mechanism. Rather, they represent distinct but interacting forms of organization. Generative grammar externalizes the hierarchical structure of praxis into language; the Kantian categories provide the formal principles through which cognition organizes and evaluates experience; and the spatiotemporal forms constitute the embodied conditions that make both action and cognition possible. Together, they define complementary dimensions of human cognition without collapsing into a single explanatory level. 3.6 Clinical Dissociations and Motor Independence A critical challenge for any praxiological account of language is to explain the well-established clinical dissociation between aphasia and apraxia (Whiteside, Dyson, Cowell, & Varley, 2015). Neuropsychological evidence has documented double dissociations between disorders of language and disorders of motor planning, indicating that although these capacities share important developmental and evolutionary relationships, they are supported by partially distinct neural systems in the mature brain. The present framework accommodates these findings through a developmental scaffolding hypothesis. During language acquisition, motor planning systems are proposed to provide the representational substrate from which hierarchical linguistic structures emerge. With increasing experience, however, these representations become progressively consolidated within specialized perisylvian language networks, particularly those involving the left inferior frontal and left superior temporal cortices. As a result, damage to these regions can produce aphasia while relatively sparing domain-general motor planning, whereas lesions affecting premotor and motor programming systems may impair praxis without substantially disrupting the abstract syntactic representations encoded within the language network. Evidence from neurostimulation and lesion studies further supports this distinction. Disruption of motor cortical regions often produces measurable effects on the processing of action-related semantics, yet exerts comparatively limited influence on core combinatorial operations or hierarchical syntax. Such findings are consistent with the view that mature linguistic representations retain functional links to motor systems while achieving a substantial degree of computational autonomy. 6 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 This interpretation also addresses prominent critiques of strong embodied cognition, particularly those advanced by Mahon and Caramazza. Their work has demonstrated that motor activation during language processing is neither universal nor sufficient to explain linguistic competence, and that experimental task demands can substantially influence estimates of motor involvement. The developmental scaffolding hypothesis is fully compatible with these observations. It does not propose that motor representations remain constitutive of all linguistic processing throughout life. Rather, it argues that motor planning plays a privileged role during the emergence of syntactic competence, after which increasingly specialized language networks support linguistic processing with progressively less dependence on the systems that initially scaffolded their development. This developmental transition provides a principled explanation for why motor involvement is often most evident during language acquisition, novel learning, and the processing of unfamiliar action-related material, while being less consistently required for fluent, automatized language use in adulthood. 3.7 Developmental Predictions and Measurement Strategies The vocalized praxis framework generates a series of empirically testable predictions that can be evaluated using developmental, neurostimulation, and longitudinal methodologies. Prediction 1: Fine Motor Milestones and Syntactic Emergence. Developmental trajectories of fine motor control should systematically predict the emergence of increasingly complex syntactic constructions. Specifically, mastery of multi-step object manipulation and sequential motor planning is expected to precede, and significantly correlate with, the acquisition of hierarchical syntactic structures, including relative clauses, complement clauses, and ditransitive constructions, even after controlling for overall cognitive development. This hypothesis can be tested by relating standardized measures of fine motor ability, such as the Bayley Scales of Infant and Toddler Development, to indices of syntactic complexity, including mean length of utterance, clausal subordination, and syntactic diversity. A moderate-to-large association is predicted (e.g., r > 0.40). Prediction 2: TMS Disruption of Motor Planning. Transient disruption of motor planning networks, for example through transcranial magnetic stimulation targeting the supplementary motor area or dorsal premotor cortex, should selectively impair the processing of syntactically complex sentences describing novel goal-directed actions. In contrast, comprehension of canonical sentence structures and abstract, non-action-related sentences should be comparatively preserved. This prediction follows from the hypothesis that action-based representations remain particularly relevant when linguistic processing requires the construction of unfamiliar hierarchical event structures. The expected effect would be moderate for action-related syntactic processing (approximately d > 0.60) and minimal for abstract sentence comprehension (d < 0.20). Prediction 3: Motor Delays and Language Outcomes. Individual differences in early motor development should prospectively predict later syntactic competence, even after accounting for vocabulary size, nonverbal intelligence, socioeconomic status, and language exposure. Longitudinal cohort studies should therefore reveal that early motor planning abilities explain unique variance in subsequent syntactic development beyond established cognitive and environmental predictors. A moderate prospective association is anticipated (e.g., r > 0.35). Importantly, these quantitative effect sizes should be regarded as heuristic estimates rather than definitive predictions. Their primary purpose is to provide benchmarks that facilitate prospective hypothesis testing and power calculations. The central claim of the present framework is therefore not the precise magnitude of these associations, but the existence of selective and reproducible relationships between the development of hierarchical motor planning and the emergence of hierarchical syntactic competence. 7 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 4. Implications for Cognitive Psychiatry and Intervention If syntactic competence is developmentally scaffolded by praxis, language disorders should not be conceptualized solely as impairments of isolated linguistic mechanisms. Rather, their assessment should also consider the integrity of the motor, praxiological, and sensorimotor systems that may contribute to the emergence of linguistic structure. Under this framework, children with primary deficits in motor planning or skilled action may be at increased risk for associated impairments in syntactic development, even when broader cognitive abilities are relatively preserved. This perspective also has important implications for intervention. If hierarchical action provides developmental support for the acquisition of syntax, therapies that systematically engage goal-directed action, object manipulation, and participation in causally structured activities may enhance language development when integrated with conventional speech and language interventions. The present framework therefore predicts that randomized controlled trials comparing combined praxiologicallinguistic interventions with language-focused therapy alone could provide a direct empirical test of the developmental scaffolding hypothesis. The framework further suggests that motor-related measures may offer valuable developmental biomarkers. Quantitative analyses of speech motor control and vocal production, together with objective assessments of fine motor performance and motor planning, may provide complementary indicators of the maturation of embodied language systems. A lifespan perspective is equally important, as the influence of praxiological scaffolding is expected to vary across developmental stages, with the strongest effects occurring during periods of heightened neuroplasticity and early language acquisition. More broadly, this framework offers cognitive psychiatry an integrative theoretical model that bridges developmental neuroscience, neuropsychiatry, embodied cognition, and situated linguistics. Rather than viewing language as the product of an encapsulated cognitive module, it conceptualizes linguistic competence as the outcome of progressively specialized neural systems that emerge through the coordinated development of body, action, memory, and the causal structure of experience. 5. Limitations and Open Questions Several limitations warrant acknowledgment. First, the account does not fully explain how abstract, non-action semantics emerge from praxis. While metaphorical extension is proposed as a mechanism, this requires further development and empirical support. Second, the predictions are specific to typical development; alternative pathways, such as statistical learning from text as in LLMs, clearly exist. Third, the neural mechanisms linking motor planning to linguistic composition require more detailed specification and empirical validation. Fourth, important aspects of language remain unaddressed, including pragmatics, discourse structure, and language change. These represent important areas for future work. 6. Conclusion Hanna's synthesis of Kant, Humboldt, and Chomsky correctly emphasizes the active, constructive nature of human cognition, yet it also risks obscuring important distinctions among the forms of productivity they each sought to explain. The present framework argues that, in typical human development, syntactic productivity emerges through the motor expression of hierarchically organized praxis grounded in embodied interaction with a causally structured world. On this view, language does not originate from an autonomous grammatical faculty alone, but from the progressive mapping of action representations onto linguistic structures during development. This reinterpretation situates human intelligibility within the coordinated dynamics of body, action, memory, and objective causality, thereby integrating insights from classical philosophy with contemporary developmental neuroscience, embodied cognition, and cognitive psychiatry. 8 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 From this perspective, syntax can be understood as the linguistic organization of praxis, whether realized through speech or sign. Language is therefore not merely a symbolic system, but the structured expression of an embodied agent acting within, and learning from, a causally organized world. Figures and Tables Figure 1 below presents the model of the relationships between productivity, creativity, and intelligence: Figure 1. Conceptual model linking productivity, creativity, and intelligence through embodied causal cognition. Source: Research data (2026). Figure 1 presents the theoretical framework underlying the proposed model of intelligent action. Objective causal structures emerging from the physical and social world interact with perceived causal relationships through perception, inference, causal attribution, prediction, and control. These processes are integrated with embodied experience, temporal memory gradients, and praxiological action (Agent-Act-Object-Purpose), generating internal action schemas that progressively evolve into speech syntax and structured language. The model proposes that intelligence emerges from the continuous synchronization between bodily experience, causal interaction with the environment, memory organization, and linguistic productivity, rather than from isolated cognitive computation. Figure 2 below presents the three macro-adaptive strategies that underlie intelligent action: 9 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Figure 2. Three macro-adaptive strategies underlying intelligent action. Source: Research data (2026). The figure summarizes three complementary adaptive systems that collectively support intelligent behavior. The reproductive adaptive strategy is based on innate biological programs and associative learning mechanisms that maximize survival and environmental adaptation. The rational adaptive strategy introduces conscious instrumental reasoning, executive control, working memory, and goal-directed planning. The creative adaptive strategy enables the generation of novel possibilities through divergent thinking, cognitive flexibility, insight, and the recombination of existing knowledge. Rather than functioning independently, these three adaptive systems dynamically interact to produce integrated intelligent action capable of solving complex real-world problems. Figure 3 below presents the internal hierarchical organization through which intelligent action progresses from sensorimotor experience to structured language. 10 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Figure 3. Internal hierarchical organization of intelligent action from sensorimotor experience to structured language. Source: Research data (2026). This figure illustrates the proposed developmental sequence through which intelligent action progressively acquires symbolic complexity. Reflex analogies strengthen interaction patterns that are stabilized through temporal gradient memory and organized into internal action schemas. Mirror neuron mechanisms contribute to motor simulation and social understanding, supporting the emergence of speech syntax based on praxiological representations (Agent-Verb-Object structure). Finally, structured language arises as a stable symbolic system exhibiting generative productivity from finite cognitive structures. The model integrates concepts from developmental psychology, cognitive neuroscience, embodied cognition, and generative linguistics into a unified explanatory framework. Figure 4 below presents the integrated theoretical synthesis comparing Hanna's productivity hypothesis with the proposed multi-level embodied cognition model: 11 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Figure 4. Integrated theoretical synthesis comparing Hanna's productivity hypothesis with the proposed multi-level embodied cognition model. Source: Research data (2026). This figure contrasts Hanna's concept of homogeneous internal productivity with the present theoretical proposal, which distinguishes three qualitatively different but functionally interconnected levels of cognitive organization: (1) Generative Grammar, responsible for linguistic productivity grounded in embodied praxis; (2) Kantian Categories, representing higher-order conceptual organization and reflective understanding; and (3) Space-Time, corresponding to embodied sensorimotor experience and neurophysiological integration. These levels are coordinated through continuous body-action-memory synchronization interacting with the objective causal structure of the world. The central contribution of the model is the reinterpretation of cognition as an embodied, dynamic, and hierarchically organized process rather than a single homogeneous computational system. References Carriere, M., Tomasello, R., & Pulvermüller, F. (2025). What is the role of the sensori-motor system in semantics? Evidence from a brain-constrained neural network model of action verbs. Language, Cognition and Neuroscience, 40(1), 1-25. https://doi.org/10.1080/23273798.2025.2596802 Chomsky, N. (1965). Aspects of the theory of syntax. MIT Press. Chomsky, N. (1980). Rules and representations. Columbia University Press. Emmorey, K. (2021). New perspectives https://doi.org/10.3389/fcomm.2021.748430 on the neurobiology of sign languages. Frontiers in Communication, 6, 748430. Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138. https://doi.org/10.1038/nrn2787 Garello, S., Ferroni, F., Gallese, V., Ardizzi, M., & Cuccio, V. (2024). The role of embodied cognition in action language comprehension in L1 and L2. Scientific Reports, 14, 12781. https://doi.org/10.1038/s41598-024-61891-w Glenberg, A. M., & Kaschak, M. P. (2002). Grounding language in action. Psychonomic Bulletin & Review, 9(3), 558-565. https://doi.org/10.3758/BF03196313 12 Research, Society and Development, v. 15, n. 9, e3215951684, 2026 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v15i9.51684 Goldberg, A. E. (2006). Constructions at work: The nature of generalization in language. Oxford University Press. Hanna, R. (2001). Kant and the foundations of analytic philosophy. Oxford University Press. Hauser, M. D., Chomsky, N., & Fitch, W. T. (2002). The faculty of language: What is it, who has it, and how did it evolve? Science, 298(5598), 1569-1579. https://doi.org/10.1126/science.298.5598.1569 Kim, H. E., Chua, R., & Hu, D. (2025). Causal inference, prediction and state estimation in sensorimotor learning. Proceedings of the Royal Society B, 292(2052), 20251320. https://doi.org/10.1098/rspb.2025.1320 Lakoff, G., & Johnson, M. (1980). Metaphors we live by. University of Chicago Press. Mahon, B. Z., & Caramazza, A. (2008). A critical look at the embodied cognition hypothesis and a new proposal for grounding conceptual content. Journal of Physiology-Paris, 102(1-3), 59-70. https://doi.org/10.1016/j.jphysparis.2008.03.004 Pereira, A. S. et al. (2018). Metodologia da pesquisa científica. Editora da UFSM. https://repositorio.ufsm.br/bitstream/handle/1/15824/Lic_Computacao_Metodologia-Pesquisa-Cientifica.pdf?sequence=1. Pezzulo, G., Parr, T., & Friston, K. (2022). The evolution of brain architectures for predictive coding and active inference. Philosophical Transactions of the Royal Society B, 377(1844), 20200531. https://doi.org/10.1098/rstb.2020.0531 Pulvermüller, F. (2005). Brain mechanisms linking language and action. Nature Reviews Neuroscience, 6(7), 576-582. https://doi.org/10.1038/nrn1706 Reggin, L. D., Gómez Franco, L. E., Horchak, O. V., Labrecque, D., Lana, N., Rio, L., & Vigliocco, G. (2023). Consensus paper: Situated and embodied language acquisition. Journal of Cognition, 6(1), 63. https://doi.org/10.5334/joc.308 Risemberg, R. I. C. et al. (2026a). A importância da metodologia científica no desenvolvimento de artigos científicos. E-Acadêmica, 7(1), e0171675. https://eacademica.org/eacademica/article/view/675. Risemberg, R. I. C., Wakin, M., & Shitsuka, R. (2026b). Qualitativa, quantitativa ou quali-quantitativa: Como identificar de modo rápido a abordagem de um artigo científico? E-Acadêmica, 7(2), e0172702. https://doi.org/10.52076/eacad-v7i2.702 Rizzolatti, G., & Craighero, L. (2004). The https://doi.org/10.1146/annurev.neuro.27.070203.144230 mirror-neuron system. Annual Review of Neuroscience, 27, 169-192. Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274(5294), 1926-1928. https://doi.org/10.1126/science.274.5294.1926 Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. Harvard University Press. Tsay, J. S., Kim, H. E., McDougle, S. D., Taylor, J. A., Haith, A., Avraham, G., Krakauer, J. W., Collins, A. G., & Ivry, R. B. (2024). Fundamental processes in sensorimotor learning: Reasoning, refinement, and retrieval. eLife, 13, e91839. https://doi.org/10.7554/eLife.91839 Whiteside, S. P., Dyson, L., Cowell, P. E., & Varley, R. A. (2015). The relationship between apraxia of speech and oral apraxia: Association or dissociation? Archives of Clinical Neuropsychology, 30(7), 670-682. https://doi.org/10.1093/arclin/acv051 13