Table of contents

• Abstract
• 1. Introduction
• 2. The Dominant Mono-Agential Paradigm
• 3. Observable Unexplained Decouplings
• 4. Resolution of Agential Tensions: Comparison Between DAVArch and Models of Internal Conflict, Physiological Emergence, and Collective Decoupling
• 4.1. Explicit Conviction and Contrary Behavior
• 4.2. Stated Truth, Flagrant Lie, and Parapraxis
• 4.3. Emotion, Endocrine Modulation, and Narrative Sovereignty
• 4.4. Collective Declaration and Effective Behavior
• 4.5. Comparative Synthesis
• 5. DAVArch and the SSI Model: Dual-Agent Architecture
• 5.1. Domains and Fields of Action of the Two SER
• 5.1.1. Latency and Capture
• 5.1.2. Archive and Evaluative Base
• 5.1.3. Speech and Language
• 5.1.4. Socialization
• 5.1.5. Subsistence
• 6. Cross-Metacognition
• 7. Clinical, Relational, and Moral Implications
• 7.1. Clinical Implications
• 7.2. Relational Implications
• 7.3. Moral Implications
• 7.4. Toward a Less Punitive Practice
• 8. Falsifiability and Empirical Pathways
• 9. Discussion and Conclusion
• References

Abstract

The dominant narrative of the human social system operates under a foundational assumption: mono-agency, the idea that each human organism is governed by a single agent that feels, decides, promises, explains, and acts. That presupposition does not account for most observable decouplings. Clinical experience, relational conflict, the divergence between statement and behavior, the recognized impact of intense affective states and psychoactive substances on decision-making, and a range of neuropsychological findings all reveal decouplings that the dominant paradigm does not explain so much as conceal through increasingly complex conceptual patches.

This paper challenges that assumption and proposes DAVArch (Dual-Agent Vital Architecture) as an architectural model for describing the ongoing negotiation between two agential systems coupled within the same organism. The human organism is governed by the continuous negotiation between two SER (Sistema Enactor Relatador): a SERbasal, oriented toward biological continuity, adaptive regulation, and the basal presuppositions and dispositions for socialization; and a SERsocial, oriented toward social coherence, identity, normativity, prestige, and relational coherence within the framework of a given individual in a concrete context and over the course of a lifespan segment within the terrestrial living system. Cross-metacognition does not create this architecture: it makes it explicit. (Throughout this paper, relato is retained untranslated from the Spanish; it designates the operative constitution of a world by a living being, and is not synonymous with narration — see section 5.)

From this perspective, mono-agency is not the architectural foundation of human functioning but the simplified form through which that architecture became socially intelligible. Many phenomena habitually treated as lying, weakness, incoherence, self-deception, or betrayal are better understood as ordinary outcomes of misalignment between coupled evaluative agencies. For the SERsocial, the explicitation of DAVArch can be understood as a milestone of cross-metacognition, analogous, on another level, to the human capacity to read its own genetic code explicitly. This paper develops the conceptual foundations of that architecture, contrasts it with the dominant mono-agential paradigm, and proposes lines for future empirical inquiry, including the study of endocrine correlates, non-verbal dissonances, divergences between declared attitudes and effective behavior, and a reinterpretation of split-brain findings.

1. Introduction

The attribution of intention, responsibility, and personal continuity in modern psychology presupposes that each human organism is governed by a single agent. This is especially notable in the efforts to reinforce it through categories tied to the mono-agent. This assumption is so deeply embedded in ordinary language, attribution practices, and much of psychological theory that it is typically presented not as a hypothesis under examination but as an immediate description of reality. That naturalization does not express a descriptive virtue but rather the effect of a historical and functional simplification. The mono-agential model makes human behavior socially and narratively manageable by unifying agency, self-explanation, and attribution, but it does so at the cost of masking the organism's effective architecture. By gathering under a single instance processes, tensions, decisions, and commitments that emerge from an internally dual organization, it induces systematic errors both in the assignment of responsibility and in the interpretation of psychic and behavioral life.

Clinical practice, relational conflict, the divergence between statement and behavior, the recognized impact of intense affective states and psychoactive substances on decision-making, as well as a range of neuropsychological findings, point to recurrent decouplings that the dominant paradigm does not adequately explain. These decouplings are increasingly recognized and studied. However, rather than revising the architectural assumption of mono-agency, contemporary thought has tended to surround it with corrective layers: unconscious processes, automatic systems, cognitive dissonance in a mono-agent, defense mechanisms, executive failures, subpersonal dynamics, implicit biases, affective predominance, and related constructs (see, among others, Kahneman, 2011; Festinger, 1957). Such additions do capture genuine phenomena, but they also leave the foundational assumption formally intact. The result is a dominant paradigm that grows increasingly complex while leaving its central premise unexamined.

This paper argues that the problem is not merely that the dominant mono-agential paradigm explains too little, but that it explains by concealing what it cannot structurally accommodate. Many phenomena habitually described as weakness, self-deception, betrayal, irrationality, ambivalence, or inconsistency are better understood not as failures of a single self but as ordinary consequences of misalignment between coupled evaluative agencies operating within the same life. In this sense, what appears as a defect under mono-agency appears as an expected tension within a different architecture.

We argue that DAVArch (Dual-Agent Vital Architecture) describes the architecture of agency in the human organism. Human life, on this account, is organized through the continuous negotiation between two SER (Sistema Enactor Relatador): a SERbasal, oriented toward biological continuity, adaptive regulation, and the basal predispositions for socialization; and a SERsocial, oriented toward social coherence, identity, normativity, prestige, and relational coherence within the trajectory of a given individual in a concrete context. The claim is not that the human being contains "two persons," nor that the model merely adds another layer to the mono-agential view. The claim is more radical: mono-agency itself is the simplified narrative interface through which the SERsocial negotiates without full access to the basal program.

From this perspective, the historical persistence of mono-agency is neither surprising nor neutral. It reflects the SERsocial's self-perception by design prior to cross-metacognition. That is, mono-agency functioned as the constitutive narrative through which dual-agent organisms became socially intelligible to one another without perceiving, from their social agent, their own internal architecture. The contemporary proliferation of conceptual patches does not indicate the progressive refinement of a sufficient model but rather the growing tension of a foundational narrative that can no longer hold.

DAVArch therefore does not seek to refine the dominant mono-agential paradigm from within. It seeks to expose its descriptive limits and to propose a different architectural foundation. From this standpoint, the explicitation of DAVArch for the SERsocial marks a milestone of cross-metacognition: a moment at which life begins to observe its own operative architecture, much as human beings came to read their own genetic code explicitly. The aim of this paper is to develop that thesis in a concise and testable form. First, the dominant mono-agential paradigm is defined together with the observable decouplings it leaves unexplained. Second, different ways of resolving those tensions are compared across models of internal conflict, physiological emergence, and collective decoupling. Third, DAVArch is presented as a dual-agent architecture grounded in the interaction among SERbasal, SERsocial, and the Physical Interface (see section 5). Finally, its implications and the empirical pathways through which the model can be tested are explored.

2. The Dominant Mono-Agential Paradigm

The dominant mono-agential paradigm assumes that a human organism is governed, effectively, by a single agent. Under this assumption, a single instance is expected to feel, evaluate, decide, promise, explain, and retain responsibility for the continuity of those acts over time. This premise is not limited to academic psychology. It structures ordinary language, moral judgment, legal imputability, therapeutic expectation, and the everyday organization of social life. People are asked what they "really want," whether they "truly meant" what they said, why they "failed themselves," or whether they "broke their word," as though a single unified internal authority were the natural seat of all evaluation and behavior.

This paradigm has strong practical appeal. It simplifies coordination among organisms by allowing each individual to appear as a socially negotiable unit. A promise can be attributed to a single speaker, a contract to a single signatory, a betrayal to a single responsible party, and a symptom to a single conflicted self. In that sense, mono-agency functions as a narrative compression device: it reduces the organism to a single representative voice. That compression is efficient for administration, morality, and rapid negotiation, but efficiency should not be confused with descriptive adequacy.

The problem begins when observable human phenomena cease to fit the premise of unified sovereignty. Instead of revising the architectural assumption itself, the dominant paradigm tends to preserve mono-agency by surrounding it with auxiliary categories. The organism continues to be treated as fundamentally unitary, but its inconsistencies are distributed among corrective notions: unconscious motives, automatic processes, cognitive dissonance, defense mechanisms, executive dysfunction, affective hijacking, implicit bias, compulsions, subpersonal conflicts, and related constructs. These notions often capture real regularities, but they typically function as patches around a preserved center rather than as a basis for revising that center. The notion of cognitive dissonance (Festinger, 1957) illustrates this procedure well: it presupposes a single agent who experiences tension between incompatible cognitions, when what may be occurring is a negotiation between agencies with different criteria.

This produces a paradox. The dominant paradigm continues to speak the language of a single agent while depending increasingly on explanatory devices that imply an incomplete sovereignty of the self. A person may sincerely declare an intention and enact another; may justify a behavior she did not effectively govern; may be legally evaluated with diminished responsibility under intense affective states; may be clinically described as split between what she says and what she does; and may show, from neuropsychology, narratives that do not fully coincide with the source of action. In all these cases, mono-agency is not abandoned. It is protected through conceptual supplementation.

For this reason, the dominant mono-agential paradigm should not be understood simply as a false theory awaiting correction. It is better understood as the constitutive narrative of a social order built upon organisms that, at the level of their social agency, experience themselves as unitary. Its persistence reflects not only institutional inertia but a deeper structural condition: the spontaneous self-perception of the SERsocial in the absence of cross-metacognition. Mono-agency is, in that sense, historically intelligible. But that intelligibility does not make it sufficient.

The purpose of the present argument is not to deny that mono-agency has had organizational utility. It is to show that this utility has had a cost. By treating the organism as though a single agent were its full author, the dominant paradigm repeatedly converts observable internal decouplings into moral or personal defects: lying, weakness, incoherence, self-deception, betrayal, irrationality. What it cannot describe structurally, it tends to moralize, pathologize, or rename. The following section examines the observable decouplings where this limitation becomes most evident.

3. Observable Unexplained Decouplings

Observable decouplings are not marginal anomalies of human functioning. They are among its most persistent features. The dominant mono-agential paradigm does not entirely ignore them, but neither does it explain them architecturally. Instead, it tends to register them as exceptions, failures, defenses, biases, or interferences affecting an agent who remains, at base, singular. From DAVArch, by contrast, these phenomena do not appear as awkward residues of an imperfectly achieved unity but as expected expressions of a negotiation between coupled evaluative agencies operating with different criteria within the same living organism. This analysis refers, in principle, to organisms without significant lesions or states of acute intoxication that decisively alter the functional integrity of their interfaces or biosoft.

A first recurrent decoupling is the divergence between promise and behavior. An individual declares that he will stop overeating, that he will no longer seek out a certain person, that he will not react violently again, or that he will commit stably to a project. Yet his subsequent behavior repeatedly contradicts that statement. Under the dominant paradigm, this situation is usually described as lying, weakness of will, self-deception, or inconsistency. But that reading presupposes that the same agent who promised is the same one who decided later under equivalent conditions. What is called incoherence may better describe, from DAVArch, a misalignment between the SERsocial that formulates a narrative strategy and the SERbasal that continues evaluating according to other indicators of continuity.

It is important, however, to introduce here a decisive distinction. DAVArch does not deny the existence of deliberate lies, conscious concealments, strategic manipulations, or moral economies based on guilt, debt, or victimhood. Such phenomena exist and can be organized from both the SERsocial and the SERbasal. What DAVArch questions is that every divergence between statement and behavior should be automatically interpreted as a lie. The dominant paradigm tends to equate deliberate lying with unexplicated inter-agential misalignment. DAVArch compels the distinction between both phenomena. In some cases, an organism deceives with calculation; in others, the SERsocial sincerely promises something that the SERbasal does not accompany, or ceases to accompany. Confusing both situations not only impoverishes description: it also distorts clinical practice, morality, and attribution.

A second observable decoupling appears between declared attitude and final behavior. This phenomenon is well known in clinical, social, electoral, relational, and consumer contexts. People say they prefer one thing, defend a value, or reject an option, and yet act otherwise. The dominant paradigm responds to this problem with notions such as bias, rationalization, social desirability, or internal conflict. All of them may capture something of the phenomenon, but they do not alter the underlying premise: a single agent would have failed to represent or execute itself correctly. From a dual-agent architecture, by contrast, the divergence ceases to be read as representational error of a single self and becomes compatible with the existence of coupled evaluative agencies whose criteria do not fully coincide (cf. LaPiere, 1934).

A third important decoupling is retrospective rationalization. Frequently, the organism acts first and explains afterward. This occurs not only in everyday life but also in clinical and experimental settings. The person feels compelled to produce a consistent justification for a behavior already enacted, even though she does not have full access to the architecture that generated it. From DAVArch, this phenomenon is not interpreted simply as a defense or an introspective defect but as a structural consequence: the SERsocial is frequently left in charge of narrating decisions or inclinations whose logic it did not sovereignly define. The subsequent narration is not always deliberately false; many times it is an attempt to recompose social continuity over a decision produced under an unexplicated negotiation. The experimental evidence that subjects confabulate causal explanations about their own behaviors was documented early on by Nisbett and Wilson (1977) and reformulated in terms of moral intuition by Haidt (2012).

Another relevant decoupling is the recognized impact of intense affective states and psychoactive substances on the capacity to decide. Clinical practice, psychiatry, law, and common sense all acknowledge that intense fear, anger, anguish, euphoria, alcohol, sedatives, stimulants, and other substances alter behavior and modify the organism's range of effective control. Yet this recognition coexists with a mono-agential language of attribution. A single sovereign agent is still assumed, except in moments of exception. DAVArch proposes a distinction between two planes here. On the one hand, the biochemical modulation induced by the SERbasal is a constitutive part of the organism's ordinary architecture: it does not accidentally disturb a pre-existing self but expresses basal evaluations that orient, bias, or intensify the social agent's decisions. In its extreme forms, such as so-called violent emotion or "amygdala hijacking," that modulation can be understood as an intensified basal response to an evaluation of vital risk (cf. Porges, 2011, on the hierarchical regulation of the autonomic nervous system as a mediator of social and defensive responses). On the other hand, exogenous psychoactive substances can intervene in that architecture by altering the balance of forces between coupled agencies and modifying the way the SERsocial can narrate, inhibit, or accompany what the SERbasal drives or blocks.

Relational conflict also shows decouplings that the dominant paradigm treats linearly. In the couple, for example, a verbal commitment may be read as a total expression of the person, and its subsequent non-fulfillment as direct betrayal. However, much of amorous, sexual, or family conflict shows a more complex structure: a SERsocial that sustains promises, pacts, roles, or ideals of coherence, and a SERbasal that re-evaluates bonds, attractions, alarms, rejections, or mandates of continuity from another register. When this architecture is not made explicit, conflict tends to translate into guilt, grievance, and litigation. The dominant paradigm facilitates that translation because it presupposes internal transparency where there is opaque negotiation.

Neuropsychological findings, particularly phenomena associated with the split brain, reinforce this difficulty. When different evaluative and narrative processes become partially decoupled, behavior can show a logic not entirely coincident with the subsequent verbal explanation (Sperry, 1982; Gazzaniga, 2011; McGilchrist, 2009). The dominant paradigm tends to absorb these phenomena as neurological oddities or as extreme limits of the unified self. DAVArch considers them especially valuable because they visibly exhibit a condition that, in milder and more ordinary form, pervades general human life: the lack of full coincidence between coupled evaluative agencies that produce narrative, orientation, and behavior according to criteria that do not entirely converge.

The hypothesis that follows from these decouplings is that the problem does not reside solely in the lack of integration of processes within a single agent but in the architectural insufficiency of the mono-agential assumption for describing how the world is effectively defined within the human organism. The following section compares different modes of resolving these tensions and prepares the positive presentation of that architecture.

4. Resolution of Agential Tensions: Comparison Between DAVArch and Models of Internal Conflict, Physiological Emergence, and Collective Decoupling

A useful route for distinguishing DAVArch from mono-agential models lies not in comparing general vocabularies or opposing entire traditions but in observing how different frameworks resolve recurrent controversies in which the hypothesis of a unitary agency shows explanatory limits. In multiple domains, from individual clinical practice to collective analysis of consumption, systematic decouplings appear between what an organism or group declares, what it experiences, what it modulates physiologically, and what it finally enacts. The decisive question is not merely to describe those decouplings but to determine which functional architecture renders them intelligible.

Various traditions have recognized phenomena incompatible with a simple mono-agential sovereignty. However, they frequently reinscribe them within corrected unitary architectures through notions of internal conflict, physiological emergence, automatic interference, contextual bias, or non-transparent zones of the same subject. DAVArch proposes another reading: such tensions are not peripheral anomalies of an underlying unity but expected effects of a dual-agent architecture in negotiation, in which functionally differentiated agents participate with unequal access to explicitation, physiological modulation, and behavioral execution.

4.1. Explicit Conviction and Contrary Behavior

A paradigmatic case is one in which an individual affirms with conviction that he will not perform a certain behavior and yet ends up performing it. The situation of someone who declares that he will not eat and shortly afterward eats does not constitute a marginal oddity; it expresses a type of decoupling widely recognized between explicit formulation and enactive resolution.

In a moral-voluntarist reading, this phenomenon is usually interpreted as weakness of character, inconsistency, or insufficient discipline. The contradiction is lodged within a single will that fails to sustain its own mandate. The problem is that this formulation normatively names the failure but does not clarify its architecture. It describes the distance between rule and action, but it does not explain why the explicit rule lacks effective sovereignty over the final resolution.

Dynamic-intrapsychic models introduce greater complexity here. The contrary behavior can be read as the return of a repressed tendency, a compromise formation, the predominance of an instance not aligned with the declared self, or the expression of an unresolved internal conflict. This family of approaches better captures the functional heterogeneity of behavior, but it usually preserves an ultimate architectural unity of the subject: the conflict occurs within a single structural entity, even if that entity is not transparent to itself. Freud (1923) was the one who most clearly risked categorizing that internal heterogeneity, proposing instances (Id, Ego, Super-Ego) that allowed the conflict to be localized rather than dissolved into a single agent. DAVArch does not argue against that intuition but updates it: what Freud described as instances of a single stratified subjectivity, DAVArch redistributes into two enacting agencies of equivalent status and non-coincident criteria, which compels a strong reconsideration of inherited clinical categories.

Cognitive-executive models offer a third variant. Here, decoupling is translated into competition among deliberative control, automatization, habit, anticipated reward, temporal discounting, or prioritization biases. This formulation has the virtue of operationalizing relevant components, but it often retains as a background assumption the idea that there is a unitary agent whose deliberative system attempts to govern subordinate processes that appear as automatic, preconfigured, or insufficiently regulated (cf. Kahneman, 2011; Friston, 2010).

Predictive coding models, particularly the free energy principle (Friston, 2010), formalize this logic as a single system that minimizes prediction error; DAVArch does not duplicate the predictive paradigm but displaces it. The issue is not that there are two systems predicting where others see only one, but that life is not limited to anticipating a given world: it operatively defines that world in order to live within it.

What is at stake, then, is not a competition between better and worse forecasts inside a single architecture, but a negotiation between agencies that define relevance, threat, opportunity, and continuity from irreducible criteria.

DAVArch therefore proposes a different resolution. The divergence between explicitly formulated conviction and effectively enacted behavior is not interpreted primarily as the failure of a single agency against itself, but as the result of a negotiation between a social agent, specialized in definitional explicitation, narrative commitment, and public traceability, and a basal agent, specialized in adaptive evaluation, regulatory pressure, and modulation of the organism from registers not equivalent to the verbal-predominant language of the social agent. In this reading, the sentence "I am not going to eat" does not exhaust the architecture that will decide the behavior. It expresses a socially formulated definition of a course of action, but it does not by itself guarantee the convergence of the whole system.

The explanatory value of DAVArch does not lie in replacing the word conflict with new terminology, but in redistributing causal agency. What in mono-agential models appears as weakness, interference, self-control failure, or insufficient regulation can here be read as the outcome of an asymmetric negotiation between agents with different languages, priorities, and fields of action.

4.2. Stated Truth, Flagrant Lie, and Parapraxis

A second family of cases concerns the tensions between verbal articulation, access to information, and control of the expressive channel. Under this heading can be gathered, with due distinctions, the flagrant lie, the parapraxis, and certain so-called drive irruptions that do not manage to translate stably into explicit language.

In the flagrant lie, the problem is not only that a subject says something false but that he does so under conditions in which he seems to know, wholly or partially, the falsity of what he states. Mono-agential models usually treat this phenomenon as self-deception, conscious dissimulation, disinhibition, or inconsistency between belief and emission. In all these cases, the figure of a single agency that possesses the information, distorts it, or fails to manage it coherently is preserved.

Dynamic-intrapsychic models complicate this scene by introducing notions of repression, disavowal, compromise formation, or the irruption of unassumed contents. The parapraxis, in particular, acquires interpretive value here because it shows that the verbal surface does not monopolize the causality of emission. However, even in these more sophisticated formulations, the recognized functional plurality is usually reinscribed within a single structural subjectivity—split or stratified, but ultimately one.

Cognitive-executive models, for their part, tend to read these episodes as monitoring errors, inhibition failures, competitive intrusions, interference from prepotent responses, or conflicts between declared intention and processing automatisms. Again, the problem is formulated as perturbation or internal limitation of a background agency that fails to fully order its own channels (cf. Nisbett & Wilson, 1977).

DAVArch shifts the question. The issue is no longer only why the same subject "fails" to tell the truth or to sustain his own verbal line, but which sector of the architecture had access to which information, in what format it could process it, and through which channel it could impose it. In this perspective, verbal articulation does not constitute the organism's sovereign window onto itself. The social agent has privileged capacities for public formulation, narrative coherence, and the management of symbolic commitment, but it does not exhaustively control all the evaluations or all the pressures that participate in behavioral and expressive output. At the same time, the basal agent does not need to propositionally formulate a public version of the world in order to bias, induce, block, or precipitate certain emissions.

This allows for a more precise rereading of the difference between lying, parapraxis, and drive pressure. A lie is not merely a falsehood emitted by a single will; it can be a strategic resolution adopted by the social agent under conditions of negotiation, pressure, or informational decoupling. Parapraxis is not merely an "error" of the same speaker but the expressive irruption of an architecture not monopolized by social articulation. And drive no longer appears as dark energy or irrational residue; it can be reconsidered as evaluative pressure from an agent that does not speak in the same way yet effectively shapes the system's orientation.

4.3. Emotion, Endocrine Modulation, and Narrative Sovereignty

A third controversy arises around the status of emotions. In numerous contemporary approaches, emotions are described as emergent states of the physiological system, patterns of endocrine regulation, rapid evaluation biases, or global modulations of the organism. These formulations have allowed increasingly precise description of relevant mechanisms. However, the description of emotion as mere emergence raises a decisive architectural question: emergence from what, with what criteria, and with what capacity for directing behavior and the definition of the world? (see Damasio, 1994, 1996; Panksepp, 1998).

In moral-voluntarist versions, emotion usually figures as a perturbation of correct control, a disorder of judgment, or an interference from an agency that should govern itself more stably. In dynamic-intrapsychic models, it can be read as the return of conflict, affective condensation, displacement, or the expression of an underlying structural tension. In cognitive-executive and neuroregulatory models, emotion frequently appears as a salience signal, rapid valuation, adaptive prioritization, or control modulation.

Each of these approaches captures genuine aspects of the phenomenon, but many times leaves the architectural status of the modulating instance underdefined. If an emotion systematically alters attention, redefines urgencies, reorganizes priorities, modifies the tone of the world, biases interpretation, and changes the probability of certain behaviors, then describing it only as "emergence" or "system state" may prove insufficient. Physiological modulation is not a collateral detail; it actively participates in resolving the situation (cf. Damasio, 1994).

DAVArch proposes that emotions are the tools the basal agent uses in negotiation with the rest of the organism. This does not imply reducing emotion to "chemistry" or assuming a rudimentary dualism between a system that feels and one that thinks. It implies, rather, recognizing that endocrine, visceral, and tonic modulation functions as a channel of evaluation and influence of an agency that does not operate primarily in the propositional register but that decisively alters the horizon of relevance and available action for the social agent. In this architecture, emotion is neither mere noise nor simple collateral effect of the system; it is one of the modes of participation of basal agency in defining the situation.

In this way, DAVArch does not deny endocrine or neuroregulatory findings but relocates them. Where certain approaches describe modulation mechanisms without fully specifying the architecture that integrates them, DAVArch proposes that such mechanisms are part of the basal agent's field of action in its negotiation with the social agent. It is not the substance in itself that determines the function, but the organism's architecture that assigns a function to a given substance. The chemical mediator matters as an operational vehicle, but not as the ultimate foundation of the functional meaning of the process. In this framework, the explanation must not stop at which molecule acts, but at what architectural role it fulfills within the system.

4.4. Collective Declaration and Effective Behavior

The problem of decoupling is not limited to the individual scale. In collective analysis as well, systematic divergences appear between declared preferences, normative identities, expressed values, and effectively performed behaviors. A frequent case is the distance between what surveys capture about tastes, opinions, or consumption intentions and what records of actual purchase, adherence, or aggregate behavior finally show. This decoupling between declared attitude and effective behavior was documented as early as LaPiere (1934) and constitutes one of the most replicated findings in social psychology.

Sociological and social-analytic approaches usually explain these discrepancies by appealing to desirability biases, contextual constraints, response performativity, normative pressure, measurement error, or inconsistency between discourse and practice. These tools are valuable and in many cases indispensable. However, they often continue to assume, implicitly, that the collective can be treated as a basic functional unit that "says one thing" and "does another," subsequently requiring corrective patches to that unit.

DAVArch suggests that these decouplings can be read differently. Without affirming here a finished theory of collective agents homologous to individual ones, it can be maintained that certain aggregate phenomena replicate an analogous structure. Surveys frequently capture better the definitional, normative, or publicly administrable layer of the collective; effective purchases and other recorded behaviors capture better the negotiated enactive output. This parallel does not pretend to equate individual and collective without qualification but rather to show that the logic of decoupling between declaration, modulation, and behavior is not restricted to the individual case and reinforces the insufficiency of the mono-agential schema as the sole matrix of intelligibility.

This point extends the scope of the general argument. If at the individual level the contradiction between word, emotion, and behavior has historically been treated as a deviation from an underlying unity, something similar may occur at the collective level when the distance between declared preference and effective behavior is processed as a mere anomaly of the same social subject. DAVArch allows, at least as a working hypothesis, rereading these tensions as indications of more complex negotiation architectures than those admitted by classical mono-agential models.

4.5. Comparative Synthesis

What these controversies share is not their thematic content but their structure. In all cases, tensions emerge between explicitation, modulation, and execution that mono-agential models tend to resolve through notions of weakness, internal conflict, automatism, emergence, or contextual bias. DAVArch does not deny the reality of those phenomena; it questions whether they should be understood as peripheral corrections to a primary architectural unity.

Its hypothesis is stronger: such tensions become more intelligible when they are treated as ordinary consequences of a dual-agent architecture. In this architecture, the social agent has advantages in explicitation, narrative formulation, the public administration of commitment, and the symbolic definition of a course of action; the basal agent has advantages in adaptive evaluation, physiological modulation, tonic pressure, and enactive influence on behavior. The organism's eventual resolution does not necessarily coincide with the voice best able to articulate it.

From this perspective, the difference between DAVArch and the models compared here does not reside merely in the former adding a new vocabulary for naming already-known conflicts. It resides in the fact that every controversy analyzed reveals the same underlying structure: a social agent that formulates, promises, or declares, and a basal agent that evaluates, modulates, or reorients from another register. In the unfulfilled conviction, the social agency formulates and the basal does not accompany. In the parapraxis, the basal agency irrupts into a channel that the social does not monopolize. In emotion, the basal agency redefines the horizon of action without passing through social narrative. In the collective decoupling, the declarative layer of the group captures the social voice while effective behavior expresses negotiations that voice does not govern.

5. DAVArch and the SSI Model: Dual-Agent Architecture

If the decouplings between explicitation, modulation, and execution, as well as the divergences between goals within the same individual, are not residual anomalies but recurrent conflicts, then the relevant question becomes which functional architecture can explain them without resorting to a proliferation of patches. DAVArch answers this question by holding that the human organism is not governed by a single agent; rather, its agency is organized as a dual-agent architecture.

This architecture should not be understood as a sum of "parts" or as a clinical metaphor for describing internal conflicts. It describes the functional structure of the agency of the human organism. Within it, two SER in tension—SERbasal and SERsocial—operate upon the same living organism, together with a Physical Interface (PI) that is also enactive. This ensemble constitutes, in the human, the SSI architecture (SERbasal – SERsocial – Physical Interface).

SER stands for Sistema Enactor Relatador. The original Spanish term is retained because relatador here does not refer merely to verbal narration, but to the organism's capacity to enact and organize relationally structured courses of sense, valuation, and action. This indicates that both agents are enactive and evaluative, and that both organize courses of action in distinct registers. Both produce orientation, relato, and strategy, though they do so with different priorities and with unequal access to verbal explicitation. The difference between them is not that one acts while the other interprets, but that they organize the organism's continuity according to criteria that do not fully coincide. The concept of enaction structuring this formulation is inspired by Varela, Thompson, and Rosch (1991), although DAVArch redistributes it within a dual-agency architecture. The term "enactor" here designates an agency that operatively brings forth a world; it is not simply equivalent to the enactive organism in Varela's sense, but rather to each of the two systems that, within the same organism, enact according to their own criteria.

In DAVArch, however, the concept of relato departs decisively from its dominant usage. In this semantic reconstruction, relato does not designate a verbal narration, a conscious story, or a discourse about the world. It designates the way in which a living being operatively defines its environment. This is why relato and narration are not synonymous. Relato occurs in the organism that defines, not in the one that offers indicators. What is offered by the environment, by another organism, or by a part of the system itself by way of narration does not circulate as already-constituted meaning but as potential material for a new definition. In this sense, the SER do not exchange finished relatos: they produce their own relato from negotiations with indicators coming from the environment, from other organisms, and from their own internal architecture. To produce relato, in DAVArch, is not to tell the world but to operatively constitute it in order to live in it.

To constitute operatively means to define, and to define in the triple sense that the Spanish verb carries: to project a course of action (definir una estrategia — to set a strategy), to delimit what is relevant and what falls outside (definir un contorno — to delineate a boundary), and to materially produce what is needed (as a sculptor defines a form in marble). The SER do not perform one of these operations at a time: all three occur simultaneously in every enaction. What drives this triple operation is not the minimization of prediction error nor the correction of deviations from an anticipatory model, but the will to live oriented by the belief that life must continue — what, with affinities and differences, can be recognized as conatus. A system that predicts generates a model and corrects what does not fit; a system that defines constitutes the world in which it is going to live, and in doing so determines what counts as fit, as risk, as opportunity and as continuity. This distinction is decisive against predictive coding models (Friston, 2010), which formalize the entire activity of the organism as a single system minimizing free energy. In DAVArch there is not one system correcting error but two agencies that define world from mutually irreducible criteria, driven by a force that does not seek to reduce surprise but to sustain and expand life.

This redefinition is decisive and expands the notion of relato beyond the plane of explicit language. Both the SERbasal and the SERsocial are relators. Both define environment, priority, risk, opportunity, and continuity. They do so, however, through different grammars. The SERsocial can verbalize an important part of its definitions and present them as justification, project, or commitment. The SERbasal does not need to do so propositionally: it produces relato through biases, impulses, alarms, appetites, rejections, intensifications, activation rhythms, and modulations that reorganize the organism's operative field through neurochemistry and other non-verbal mechanisms of negotiation.

The SERbasal is oriented toward biological continuity, adaptive regulation, and the basal predispositions for socialization. It evaluates in terms of survival, energy, attachment, alarm, reproduction, compatibility, safety, and vital continuity. Its mode of expression is not primarily propositional. It expresses itself by inducing attentional biases, impulses, alarms, attractions, rejections, appetites, and activation rhythms in the SERsocial and the PI through biochemical modulations that orient the rest of the organism prior to discursive articulation. It also participates in basal forms of relating: attachment, hierarchy, cooperation, mating, caregiving, and other modes of basal socialization. Consequently, the SERbasal is not a mere physiological background, the base of an iceberg, or a reservoir of impulses; it is an enactive-relator agency that defines a world from the logic of vital continuity as shaped by evolution.

The SERsocial is oriented toward social coherence, identity, normativity, prestige, and relational coherence within the trajectory of a given individual in a concrete context. It is the agency that habitually appears under the pronoun "I." It produces relato, promises, justifies, plans, answers for the organism before others, documents commitments, and organizes the individual's continuity within a specific social milieu. Its mode of expression is more readily verbalizable, more sensitive to norms, and more suited to explicit negotiation. In evolutionary terms, DAVArch proposes that the animal social interface, present in rudimentary form in many species, in humans develops to the point of acquiring the status of SER: it ceases to be merely an interface and becomes SERsocial - that is, an enactive-relator agency with its own competencies in the social field. At this point, the proposal retains partial affinities with Mead (1934), for whom the self emerges in the social process, although DAVArch redistributes that emergence within a dual architecture of agency.

The Physical Interface (PI) is the enactive-physiological domain of the organism: the active bodily field where the strategies of both SER are executed, registered, and limited. It does not constitute a passive support. It is enactive. It carries the anatomy, sensoriality, motricity, energetic availability, endocrine response, inscription of damage, and the material execution of much of the negotiation between both SER. Through it, the organism touches, avoids, resists, becomes exhausted, enjoys, is injured, recovers, and conditions what each agency can sustain. The PI does not decide in the same sense as the SER, but neither does it function as a mere neutral instrument: it introduces limits, costs, possibilities, and retroactions that actively participate in the system's dynamics. In that sense, the PI is not only a vehicle of execution; it also participates in constituting the environment as the organism can effectively inhabit it.

The bilaterality of the organism offers in DAVArch a relevant structural antecedent, but it does not by itself amount to a simple hemispheric localization or an exhaustive neuroanatomical partition of functions. Its value in this model is architectural and evolutionary, not topographic in a narrow sense. Bilaterality is presented as the evolutionary basis of a negotiation architecture that allows the coexistence of non-identical perspectives within the same system. Expanding and redefining what McGilchrist (2009) proposes—for whom the hemispheres produce two qualitatively distinct ways of attending to the world—DAVArch does not reduce agential duality to hemispheric lateralization but takes it as a possible antecedent.

The emotional dimension makes this architecture especially clear. In DAVArch, the SERbasal does not primarily experience emotions in the phenomenological register in which the SERsocial can articulate them; rather, it induces and modulates them as part of its regulatory field. It generates biochemical modulations in the SERsocial and the PI that bias, orient, or intensify the organism's behavior according to its evaluation of context. The SERsocial and the PI undergo these modulations as lived experience. In this way, emotion appears neither as simple chemical emergence nor as an exclusive property of a narrative self, but as the result of an architecture in which one agency regulates and other embodied instances undergo, use, or negotiate that regulation. In its extreme forms, such as violent emotion, this logic intensifies.

DAVArch describes, in sum, an architecture of coexistence and not of fusion. The human organism is not organized as a single self, nor as a higher unity that absorbs all internal difference, but as the continuous negotiation between two coupled SER and an enactive Physical Interface upon the same life. From this perspective, describing the human requires abandoning the image of an agent that first defines and then, in any case, fails. It requires, instead, mapping the architecture through which different relatos, produced within the same organism, negotiate which world is defined as habitable, desirable, risky, binding, or intolerable. On this basis, the following section examines the concept of cross-metacognition and its role in the explicitation of this architecture.

5.1. Domains and Fields of Action of the Two SER

Reciprocal Genesis and the Non-Resolution of Origin

The SSI architecture does not posit three independent components that were assembled at some point, nor a genetic sequence in which one preceded the others. The basal and the social do not exist without physical support; the physical is the crystallization of basal and social enactions; the basal is crystallized social strategy; the social is a tool of rapid adaptation to the concrete environment of a concrete organism. This circularity is not a defect of the model: it is its most honest description. Bacterial quorum sensing—social coordination among individual organisms that may eventually crystallize as the structure of a multicellular organism—illustrates how what initially operates as negotiation among agents can lose degrees of freedom until it materializes as non-negotiable architecture. DAVArch does not claim to resolve which domain is original. It holds that all three co-produce one another, and that deciding which came first would amount to locating a starting point in a circular process that has been operating since the earliest living systems.

In the case of the two SER agents, the bilaterality of the human brain offers the relevant structural antecedent already anticipated above: the physical substrate already carries a dual organization with documented functional asymmetries. DAVArch does not, however, claim that the SERbasal and the SERsocial are agents localized in different hemispheres or discrete brain regions. The two SER are not homunculi housed in anatomical compartments: they are regimes of evaluation and operation that may share substrates, be variably distributed across structures, and be reconfigured according to context and demand. That certain functions associable with a SER may predominate in one hemisphere does not imply that the hemisphere is that SER. Just as a software program is not defined by the processor on which it runs—and may run partially on several processors, or share one with other programs—the agents in DAVArch are defined by their evaluative criteria, temporal regimes, and fields of operation, not by their location in hardware. Split-brain evidence does not demonstrate that each hemisphere contains an agent; it demonstrates that when negotiation between substrates is interrupted, divergent evaluations emerge that the intact architecture normally integrates through continuous co-production.

5.1.1. Latency and Capture

The SERbasal operates with detection latencies on the order of milliseconds. It captures the equivalent of a video frame: it evaluates a sensory signal, determines whether it constitutes threat or opportunity, and orders chemical responses before the SERsocial has even begun to process the situation. This regime of capture explains why the organism can react with startle, attraction, or rejection before any narrative explanation is available. The SERbasal does not process argumentative sequences: it evaluates instantaneous configurations and responds chemically.

The SERsocial, by contrast, operates with latencies on the order of seconds. It does not capture isolated frames: it builds symbolic coherence by integrating sequences, context, expectations, and prior agreements. Its processing is necessarily slower because its domain—relational coherence—requires the integration of indicators of greater combinatorial complexity. The SERsocial has no direct access to the signals the basal system evaluated in milliseconds: it receives the effects of that evaluation as emotional states or sensations already chemically induced, and processes them according to its own logic.

This produces a constitutive temporal asymmetry. By the time the SERsocial formulates an explanation of why the organism feels what it feels, the SERbasal has already evaluated, responded, and quite possibly readjusted its strategy. Much of what is called retrospective rationalization is the SERsocial constructing a coherent relato over decisions that the SERbasal has already made.

5.1.2. Archive and Evaluative Base

The SERbasal operates on a phylogenetic archive: strategies crystallized over millions of years of evolutionary history. This archive is not broad in the number of discrete items it contains, but in its temporal depth. Each entry is the result of enactions that proved consistently viable and lost degrees of freedom until they became non-negotiable mandates. The cost of revising a basal belief is high, and its windows of update are long. The SERbasal is extremely cautious in establishing changes of criterion because every criterion it carries was stabilized by selective pressure over generations.

The SERsocial operates on an ontogenetic archive that begins nearly empty and is built during the life of the concrete organism. Its elasticity is maximal: it is designed to absorb rapidly the specific conditions of the social environment into which it is born, to learn its rules, read its signals, and create new configurations. What it loses in temporal depth it gains in adaptive speed. In the human case, the invention of systems of external inscription—from clay tablets to computers—decisively amplified this archive, not only in storage capacity but in a specific function: allowing the SERsocial to observe its own relatos in delayed form. This delayed observation is the condition of possibility for cross-metacognition. Without the externalization of relato, the SERsocial cannot examine itself and, by extension, cannot distinguish its own evaluations from those of the SERbasal.

5.1.3. Speech and Language

The SERbasal understands linguistic structure but does not produce speech. Evidence from disconnected hemispheres shows that the hemisphere not dominant for verbal production can understand instructions, evaluate semantics, and recognize grammatical structures. If the SERbasal is functionally mapped onto that register, its relation to language is not one of ignorance but of comprehension without production. Its influence on speech is exerted by electrochemical induction: by modulating emotional states, urgencies, and dispositions, the SERbasal can affect the selection of words, the construction of phrases, and the expressive intensity of the SERsocial without having to formulate propositions of its own. This influence is not limited to the lexicon: under intense emotional states basally induced, the SERsocial produces syntactic structures it would not have chosen under lower chemical pressure—broken phrases, imperative constructions, compulsive repetitions—which shows that basal induction operates on the very architecture of the utterance, not only on its content.

The slip of the tongue, the word that emerges against declared intention, can be read within this framework as an irruption of basal criteria into the social expressive channel.

The SERsocial speaks without needing to rationalize grammar. It produces speech as enaction: it does not consciously apply rules; speech emerges as the fluent execution of acquired competences. But speaking is not only production for others: it is also a physical act that generates proprioceptive and somatosensory feedback—laryngeal vibration, orofacial muscular activation, bone resonance, auditory feedback—that the SERbasal can evaluate. When the SERsocial verbalizes an argument, it simultaneously produces a bodily event that the SERbasal registers and upon which it can operate. In this way, speech functions as a bidirectional channel: the SERsocial offers content to the social environment and, at the same time, exposes that content to the SERbasal through the physical interface.

5.1.4. Socialization

The SERbasal carries phylogenetically preset social strategies. Nursing, mating, attachment predispositions, the evaluation of biological compatibility (MHC/HLA, microbiota, olfactory signals), and dispositions toward roles such as leadership, resource distribution, or the assessment of scarcity and abundance are functions of the basal domain. These strategies do not operate as explicit instructions, but as chemical orientations that induce emotional states in the SERsocial and motor dispositions in the PI.

But the social domain of the SERbasal includes one function that generates highly visible dissonances: the improvement of the lineage's limitations. The SERbasal evaluates the inherited deficits of the lineage—immunological, temperamental, or social-structural—and orients the organism toward configurations that compensate for them. Partner choice is the paradigmatic case. The SERsocial may declare a preference for one type of partner while the SERbasal directs attraction toward another type that compensates for lineage deficits. Under the mono-agential paradigm, this is read as self-deception, compulsive repetition, or a neurotic pattern. Under DAVArch, these are two agents operating simultaneously over the same field with different criteria of selection. Studies with disconnected hemispheres show this without mediation: each hemisphere chooses a different profession, selects different objects, and evaluates situations according to different criteria. This is not ambivalence within a single self: it is two evaluators who, when no longer required to negotiate, reveal agendas that had always been different.

The SERsocial, for its part, manages belonging to the group as an existential priority. Social exclusion is not a mere narrative inconvenience: the SERbasal reads it as a real vital threat and responds with the same chemistry it would mobilize in the face of a predator. Belonging determines access to resources, protection, and reproduction. The SERsocial weaves group networks, distinguishes the in-group from out-groups, learns trades and uses of tools, transmits knowledge, and organizes the dynamics of competition, cooperation, and aggression within the framework of prevailing social agreements.

5.1.5. Subsistence

The SERbasal manages preset homeostasis: thermal, energetic, immunological, and reproductive regulation. Its protective strategies include cooperation with internal agents (symbiotic microorganisms, immune responses) and responses to external threats (other animals, other humans, adverse environmental conditions). These strategies operate without requiring the participation of the SERsocial and may even operate against its explicit indications, as when the organism becomes ill to resolve an invasion the SERsocial had not registered.

6. Cross-Metacognition

Cross-metacognition is the process by which one of the SER begins to observe and explicitly map the architecture it shares with the other. In the human case, it is principally the SERsocial that gains this capacity, since it is the agency with privileged access to verbal articulation and symbolic formulation. However, cross-metacognition does not create the architecture: it makes it explicit. The architecture was already operative; what changes is that the SERsocial begins to distinguish its own evaluations from those of the SERbasal.

This distinction is not merely theoretical. It has practical consequences for clinical work, relational conflict, and moral judgment. When the SERsocial operates without cross-metacognition, it tends to attribute all evaluations to itself, to treat basal pressures as personal failures, and to moralize what it cannot explain. When cross-metacognition begins to operate, the SERsocial can begin to map the negotiation rather than simply undergo it. This does not eliminate the tension between agencies, but it can reduce the unnecessary suffering that arises from interpreting structural decoupling as personal defect.

The condition of possibility for cross-metacognition lies in the externalization of relato. As long as the SERsocial's relatos remain internal, they cannot be examined at a distance. The invention of external inscription systems—from clay tablets to digital text—allowed the SERsocial to observe its own productions in delayed form and, eventually, to notice that not all of its evaluations originate in itself. In this sense, cross-metacognition is not a spiritual achievement or a therapeutic technique but a structural consequence of the development of external inscription.

7. Clinical, Relational, and Moral Implications

7.1. Clinical Implications

If the architecture of the human organism is dual rather than mono-agential, then clinical practice must revise several of its foundational assumptions. The first is that the patient is a single agent who fails, resists, or deceives. Under DAVArch, many clinical phenomena—relapse, ambivalence, broken commitments, compulsive behavior, contradictory statements—can be reread as expressions of the negotiation between SERbasal and SERsocial rather than as defects of a single self. This does not abolish clinical categories, but it redistributes their explanatory weight.

The second revision concerns the therapeutic relationship itself. If the therapist addresses a single agent, she may inadvertently reproduce the mono-agential compression that generates part of the suffering. If she begins to map the negotiation between agencies, she can help the patient distinguish between what the SERsocial articulates and what the SERbasal evaluates, reducing the moralization of structural decoupling.

7.2. Relational Implications

In relational conflict, DAVArch introduces a more sober reading of broken promises, erotic ambivalence, and the distance between what is said and what is done. The dominant paradigm tends to translate every divergence into betrayal, weakness, or deception. DAVArch distinguishes between deliberate strategies and unexplicated misalignments, allowing a less punitive and more architecturally precise reading of relational dynamics.

This does not eliminate accountability. It demands a more rigorous prior description before assigning guilt. The distinction between a deliberate lie and a structural misalignment is not a subtlety: it determines whether the response should be moral condemnation, relational renegotiation, or architectural mapping.

7.3. Moral Implications

On the moral plane, the deepest consequence is that DAVArch questions the naturalness of a dual morality founded on linear oppositions: truth or lie, loyalty or betrayal, will or weakness, authenticity or repression, coherence or hypocrisy. Such oppositions do not disappear entirely, but they cease to be sufficient categories for describing the dynamics of the organism. The dominant paradigm moralizes where it cannot map. It converts inter-organic decouplings into faults of a unitary self. What it does not comprehend as negotiation, it translates as fault (see Haidt, 2012, on the primacy of moral intuition over rational justification).

DAVArch does not therefore propose a naïve suspension of morality or a generalized absolution. It does not claim that every conflictive behavior is justified by the existence of two agencies. It asserts something more demanding: that moral evaluation must begin with an architecturally more precise description. Only after distinguishing between deliberate deception, structural misalignment, basal reconfiguration, social capture of debt, or unexplicated conflict does attribution cease to rely on a unitary fiction and become more rigorous.

This moral reconsideration becomes still more precise if one distinguishes between tools of abundance and tools of scarcity. DAVArch holds that living systems have adaptive strategies for both registers: in contexts of abundance, the circle of cooperation expands, while in contexts of scarcity, competitive, aggressive, or predatory tools emerge. The problem with the dominant paradigm is not recognizing both but moralizing them dually: prestiging the former and concealing the latter. Thus, when the SERbasal narrates scarcity and resorts to more aggressive tools, the SERsocial tends to deny them or to produce justificatory narratives that preserve its normative image, deepening the dissonance between agents. DAVArch does not canonize any of these tools; it reinscribes them as contextual and demands that their use be evaluated according to harm, necessity, and coherence with vital continuity.

This also modifies the status of guilt. Under mono-agency, guilt usually functions as a linear record of a fault of the subject against itself or against another. Under DAVArch, guilt can be read as one of the ways in which certain decouplings become documented and administered within moral and relational economies. It does not disappear, but it ceases to be automatic evidence of a simple moral truth. It can be a pressure device of the SERsocial, a basal record of debt, a tool of victimhood, or an indicator of unmapped conflict. That plurality demands a more sober and less automatic reading.

7.4. Toward a Less Punitive Practice

The joint consequence of these transformations is a more precise clinical, relational, and moral practice. Where the dominant paradigm asks who failed, DAVArch asks what each SER evaluated and why their conclusions did not converge. Where mono-agency reads the broken promise as betrayal, DAVArch distinguishes whether the SERsocial promised something the SERbasal never accompanied or whether a subsequent basal reconfiguration altered the conditions of the pact. Where moral judgment condemns the organism for not behaving as one, DAVArch points out that unity was a narrative interface, not a description of the architecture.

This does not reduce the ethical demand. It displaces it. Honesty no longer consists solely in "telling the truth" from a supposedly transparent self but in learning to recognize when one is facing a deliberate strategy, when facing an instrumentalized debt, when facing a misalignment between SERbasal and SERsocial, and when facing an effective transformation of the organism that the inherited language can no longer describe. In that sense, DAVArch does not impoverish responsibility: it makes it less immediate, less theatrical, and more faithful to the architecture of human agency.

8. Falsifiability and Empirical Pathways

Precisely because DAVArch is proposed as a description of the architecture of agency of the human organism, it must open predictions, observable frontiers, and possible conditions of refutation. This section outlines some of those pathways for empirical contrast.

The first condition of falsifiability is architectural. DAVArch would lose force if the decouplings that it reads as tensions among SERbasal, SERsocial, and PI could be described sufficiently, stably, and non-moralistically from an effectively mono-agential model, without multiplying auxiliary patches or displacing the anomaly into supplementary categories. DAVArch currently has no patches. If a single-agent paradigm were to explain with equal or greater precision the divergence between social articulations and basal evaluations, the retrospective rationalization of the SERsocial, the non-coincidence between endocrine-regulatory language and rationalized language, the differential impact of intense affective states and psychoactive substances on internal negotiation, and relational conflict without relapsing into a fiction of unitary sovereignty, then DAVArch would be weakened. The dual-agent thesis stands not by novelty but by comparative explanatory power. Symmetrically, DAVArch must explain not only divergence between agencies but also convergence: if the architecture is dual, why does the organism operate most of the time as though it were one? The model's answer is that functional convergence does not indicate fusion but successful negotiation: when both SER evaluate compatibly, the organism produces coherent behavior without needing to articulate the duality. Apparent mono-agency is, in this sense, the ordinary outcome of a well-resolved negotiation, not proof of an underlying unity. Conversely, divergence destabilizes the mono-agential explanation and exposes its descriptive limits.

The second condition is inter-organic. The model presupposes that relatively distinguishable frontiers exist between basal evaluations, social articulations, and the constraints or records of the Physical Interface. If those frontiers were completely unobservable in principle - if they could not be inferred from patterns of behavioral reorganization, non-verbal leakage, physiological markers, or clinical regularities - then the proposal would come too close to a global metaphor without empirical anchorage. DAVArch requires the opposite: the architecture must leave traces - systematic differences between what the SERsocial articulates, what the SERbasal induces or reconfigures through its own language, and what the PI registers as cost, fatigue, activation, pleasure, rejection, injury, or limit.

A third pathway of contrast is endocrine and neuroregulatory. The proposal holds that the biochemical modulation induced by the SERbasal is not unspecific noise or a mere chemical accompaniment, but one of the ordinary languages of vital agency. From this follows an empirical expectation: it should be possible to detect consistent configurations linking basal evaluations, their hormonal or autonomic correlates, and the subsequent articulations, misalignments, or reorganizations of the SERsocial and the PI. This does not imply reducing agency to chemistry; it implies the opposite. It requires showing that chemistry forms part of a regulatory architecture with functional directionality. In this framework, the combined study of endocrinology, micro-expression, postural tone, avoidance, activation rhythms, and behavioral reorganization becomes especially relevant.

A fourth pathway is comparative and behavioral. DAVArch predicts that the non-coincidence between agential languages should not appear as random residue but as a systematic pattern wherever the SERsocial formulates commitments or articulations that do not coincide with basal evaluation, or wherever the SERbasal narrates scarcity, alarm, attachment, or attraction in a register that the SERsocial fails to translate precisely. Empirically, this opens a concrete agenda: comparing social strategies with basal modulations, PI reorganizations, and effective trajectories of enaction; observing contexts of scarcity and abundance; recording how enacted tools change when the SERbasal narrates risk or availability; and examining whether subsequent social strategy tends to deny, justify, or resignify those variations. Along this line, the shift from a dual morality to an adaptive ethics is not merely normative; it also makes it possible to formulate hypotheses about when the organism canonizes tools of abundance and when it activates tools of scarcity under semantizations that do not recognize them as such.

A fifth pathway of contrast comes from split-brain phenomena and, more broadly, from any situation in which different evaluative and articulatory processes become partially decoupled. DAVArch does not need to reduce itself to those extreme cases, but it can find in them a privileged methodological window (Gazzaniga, 2011). If, under such conditions, it could be shown that the non-coincidence between articulations, orientations, and reorganizations is only apparent or can be fully reabsorbed by a well-specified mono-agential model, the dual-agent thesis would lose support. If, however, such phenomena continue to show that the SERsocial's articulations do not fully coincide with basal orientation and with the SSI's overall enactive organization, then the model gains architectural plausibility.

The sixth pathway is clinical and longitudinal. DAVArch should not be validated merely by sounding intuitive but by producing differentially fruitful interventions. This requires observing whether a practice based on mapping the negotiation between SERbasal, SERsocial, and PI achieves better results than a mono-agential practice in domains such as relapse, compulsion, overeating, erotic ambivalence, chronic guilt, broken promises, or relational litigiousness. Here the criterion should not be merely subjective or narrative. It should include indicators such as tension reduction, cost redistribution, relapse reduction, changes in avoidance patterns, modification of relational biases, and greater capacity to distinguish between deliberate deception, instrumentalized debt, and structural decoupling.

A seventh pathway, more audacious, is technological and modeling-oriented. If the agential-regulatory duality posited by DAVArch were completely arbitrary, it should not be possible to functionally emulate it. However, the development of artificial endocrine layers, neuromodulation, and programmable circuits opens the possibility of designing systems that replicate, at least partially, an analogous organization: a basal regulatory layer, a social articulation layer, and an active bodily interface. In strong terms, if such an artificial architecture could be built without the patterns DAVArch expects appearing, or if a well-designed mono-agential system reproduced the same phenomena with equal power, the proposal would need to be revised.

Finally, there is also a condition of theoretical sobriety. DAVArch would fail if it limited itself to renaming with neologisms what is already sufficiently described by other models, or if it could not distinguish between architecture, substrate, intoxication, injury, and ordinary conflict. In other words, it is not enough for the model to be suggestive; it must carve the phenomenon better, produce new distinctions, and tolerate risk. For this reason, falsifiability here does not consist in a single decisive experiment but in a constellation of contrasts: physiological, behavioral, clinical, neuropsychological, and technological.

Understood in this way, falsifiability does not weaken the proposal. It makes it workable. If the dual-agent architecture is more than a verbal reorganization of human conflict, then it must leave observable traces, open predictions, and clearly accept the conditions under which it could prove insufficient. The final section takes up these implications and situates the general scope of the proposed displacement.

9. Discussion and Conclusion

The trajectory developed above allows a more precise account of the scope of the proposed thesis. DAVArch is not offered as a new vocabulary for naming familiar human conflicts, nor as a clinically useful metaphor within a framework that would remain, in its basis, mono-agent. Its claim is stronger: many of the phenomena that modern psychology, clinical practice, ordinary morality, and legal language have treated as failures of a unitary self become more intelligible once one abandons the assumption of a single sovereign agency per organism and adopts a dual-agent architecture.

From this perspective, the main limitation of the dominant paradigm is not simply that it "explains too little," but that it preserves its central premise by covering its inadequacies with ever more numerous supplements. The proliferation of notions such as internal conflict, automatism, bias, defense, pulsional irruption, dissonance, or affective hijacking does not invalidate the phenomena those notions attempt to capture; on the contrary, it shows that decoupling is persistent enough to require continual correction. What DAVArch calls into question is the idea that those corrections should continue to orbit around a fiction of unitary sovereignty. If explicitation, modulation, and execution do not stably converge in a single instance, then the problem is no longer one of integration within one agent, but of architecture.

The model enables a unified rereading of phenomena that usually appear dispersed across separate domains: broken promises, retrospective rationalization, erotic ambivalence, relational conflict, decision-making altered by intense affective states, decoupling between declared attitude and effective behavior, and certain neuropsychological findings in which narrative and orientation do not fully coincide. The gain does not lie in denying the specificity of each case, but in offering a more parsimonious redistribution of causal agency. Where the mono-agential paradigm tends to moralize, pathologize, or proliferate supplements, DAVArch proposes mapping the negotiation between coupled agencies with different access to explicitation, physiological modulation, and behavioral resolution.

This displacement has direct consequences for the status of responsibility. The argument does not lead to a generalized absolution or a relativistic dissolution of moral judgment. Nor does it allow reducing every divergence to mere innocent misalignment. What it demands is a more rigorous prior description. Only after distinguishing between deliberate deception, inter-agential decoupling, basal reconfiguration, social capture of debt, or unexplicated conflict does attribution cease to rest on a fictitious unity and become more precise. At this point, DAVArch does not weaken the ethical demand; it makes it less automatic, less theatrical, and more dependent on the effective architecture of the organism.

The proposal also has a clinical and relational scope that should not be underestimated. If structural decoupling continues to be read as moral defect of the self, the cost is not merely conceptual: it translates into unnecessary guilt, subjective burden, avoidable relational litigiousness, and clinical interventions that confuse mapping with reproach. DAVArch suggests that part of human suffering derives not solely from the conflict between agencies but from the inadequate descriptive framework with which that conflict has been interpreted.

That said, the strength of DAVArch cannot be decided by philosophical affinity, rhetorical force, or intuitive clinical resonance. Its value depends on exposure to contrast. If an effectively mono-agential model were to explain, with equal or greater precision, the decouplings gathered here without reverting to a multiplication of patches or to moral over-attribution, then the dual-agent proposal would have to be revised. Likewise, if no relatively consistent empirical traces could be found linking basal evaluations, social articulations, PI modulations, and behavioral reorganizations, the model would lose support. Here, falsifiability does not weaken the central thesis; it imposes discipline.

It is worth situating this proposal against the most sophisticated contemporary attempts to address the same territory. Solms (2021) explicitly sets out to reconnect the Freudian program with neuroscience through Friston's free energy principle: the most visible attempt to update Freud with available biology. However, Solms reintegrates the Freudian instances within a single predictive system that minimizes free energy; the resulting architecture remains mono-agential, albeit biologically enriched. DAVArch does not deny the utility of the mechanisms Solms describes, but holds that reabsorbing them into a single evaluative system reproduces the problem it purports to solve.

Seth (2021) proposes that conscious experience is a controlled hallucination generated by the organism's predictive system. His formulation is neuroscientifically rigorous and represents one of the most articulate positions in contemporary neuroscience of consciousness. Nevertheless, it remains entirely within a single-agent architecture that generates models and corrects deviations; the evaluative duality that DAVArch posits has no place in that framework.

Thompson (2007) and the 4E cognition tradition (embodied, embedded, enacted, extended) constitute the closest interlocutors to DAVArch on a decisive point: they share the notion that the organism does not represent a prior world but constitutes it enactively. DAVArch adopts that premise but redistributes it: where enactivism maintains a single enactive organism, DAVArch unfolds it into two enactors with mutually irreducible criteria. The difference is not one of vocabulary but of architecture.

Damasio (2010) advanced significantly by showing that the self is constructed in successive layers of complexity, from a bodily proto-self to an autobiographical self. That layering captures a real functional heterogeneity, but organizes it as stratification within a single subject, not as negotiation between agents. DAVArch does not deny the stratification Damasio describes; it holds that such stratification becomes more intelligible when reread as the outcome of an architecture in which two agencies define world from non-coincident registers. In each of these cases, the sophistication of the contemporary framework does not suffice to abandon the mono-agential assumption. It surrounds it, nuances it, complexifies it, but does not revise it. DAVArch distinguishes itself from all of them not by ignoring them but by identifying the exact point at which they stop.

However, the problem can no longer be formulated as though mono-agency remained a neutral and sufficient foundation. What this paper attempts to show is that its historical persistence has owed less to its architectural adequacy than to its utility for rendering socially complex organisms manageable. Mono-agency simplifies narration, identity, attribution, and negotiation; but that very simplification compresses decisive functional differences and transforms structural decouplings into personal defects. As long as that compression remained operative, it could be mistaken for descriptive truth. The tipping point appears when the explanatory, clinical, and moral cost of sustaining it begins to exceed its administrative advantage.

The final thesis, then, can be stated as follows: describing a dual-agent architecture as though it were a defectively integrated mono-agent system is no longer a legitimate simplification, but a descriptive reduction that loses precision precisely where precision is most needed. It loses theoretical precision because it compresses the architecture of human agency; clinical precision because it turns into defect what first ought to be mapped; and relational and moral precision because it over-attributes responsibility where unitary sovereignty does not adequately describe the process that produced the behavior. If DAVArch does in fact describe human architecture more precisely, then continuing to interpret the organism as though it were governed by a single agency will cease to be a mere theoretical preference and will instead appear as an unnecessary compression of human architecture, with avoidable clinical, relational, moral, and institutional consequences.

References

Damasio, A. R. (1994). Descartes' Error: Emotion, Reason, and the Human Brain. Putnam.

Damasio, A. R. (1996). The somatic marker hypothesis and the possible functions of the prefrontal cortex. Philosophical Transactions of the Royal Society of London B, 351(1346), 1413–1420.

Damasio, A. R. (2010). Self Comes to Mind: Constructing the Conscious Brain. Pantheon Books.

Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford University Press.

Freud, S. (1920). Jenseits des Lustprinzips [Beyond the Pleasure Principle]. Internationaler Psychoanalytischer Verlag.

Freud, S. (1923). Das Ich und das Es [The Ego and the Id]. Internationaler Psychoanalytischer Verlag.

Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.

Gazzaniga, M. S. (2011). Who's in Charge?: Free Will and the Science of the Brain. Ecco/HarperCollins.

Haidt, J. (2012). The Righteous Mind: Why Good People Are Divided by Politics and Religion. Vintage Books.

James, W. (1902). The Varieties of Religious Experience: A Study in Human Nature. Longmans, Green, and Co.

Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.

LaPiere, R. T. (1934). Attitudes vs. Actions. Social Forces, 13(2), 230–237.

Maturana, H. R., & Varela, F. J. (1980). Autopoiesis and Cognition: The Realization of the Living. D. Reidel Publishing.

McGilchrist, I. (2009). The Master and His Emissary: The Divided Brain and the Making of the Western World. Yale University Press.

Mead, G. H. (1934). Mind, Self, and Society. University of Chicago Press.

Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84(3), 231–259.

Panksepp, J. (1998). Affective Neuroscience: The Foundations of Human and Animal Emotions. Oxford University Press.

Porges, S. W. (2011). The Polyvagal Theory: Neurophysiological Foundations of Emotions, Attachment, Communication, and Self-Regulation. W. W. Norton.

Seth, A. (2021). Being You: A New Science of Consciousness. Dutton.

Solms, M. (2021). The Hidden Spring: A Journey to the Source of Consciousness. W. W. Norton.

Sperry, R. W. (1982). Some effects of disconnecting the cerebral hemispheres. Science, 217(4566), 1223–1226.

Thompson, E. (2007). Mind in Life: Biology, Phenomenology, and the Sciences of Mind. Harvard University Press.

Varela, F. J., Thompson, E., & Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human Experience. MIT Press.