Yuxin Zhang

The integration of generative artificial intelligence (GenAI) into academia presents profound ethical challenges to research integrity. Disciplinary norms likely shape engagement with this technology, yet a comparative analysis between medical and non-medical research cultures is lacking. This study examines how this critical divide influences the adoption of GenAI, the development of ethically sensitive behaviors, and perceptions of oversight in Chinese universities. A cross-sectional online survey was administered to 5,731 active researchers (including postdoctoral fellows and doctoral, graduate, and undergraduate students) from 48 universities, categorized as medical (n = 1,935) or non-medical researchers (n = 3,796). The instrument measured the frequency of GenAI adoption, integrity awareness, self-reported behaviors (including data fabrication/falsification), and ethical attitudes. Analyses used the chi-square test and independent-samples t-tests. Non-medical researchers reported higher GenAI adoption (57.0% vs. 45.6%, p

Selective Reality Theory (SRT) is a transdisciplinary unified ontological framework built upon the core thesis that "Existence = Selection." This paper presents a comprehensive review of the SRT theoretical system, covering its axiomatic foundations, mathematical formalization, and systematic applications across six major domains: quantum physics, neuroscience, artificial intelligence, philosophical ethics, social science, and spiritual practice. SRT proposes a Triadic Ontology: the Latent Domain L0 (the set of all unselected possibilities), the Manifest Domain L1 (the presently anchored reality through selection), and the Convergence Domain L2 (stable constraint structures sedimented from selection history). The core mechanism connecting these three domains is the Ghost Operator (G^θ), a parameterized selection mapping characterized by embodiment parameters θ. The theory's unifying power resides in the Cross-Scale Isomorphism Theorem: the same selection dynamics equations operate in a self-similar fashion across scales—from quantum collapse to neural ignition, from social norm formation to spiritual awakening. SRT introduces two critical metrics: the d-value (depth of care), measuring selection scope and concern depth along a continuous spectrum from inorganic matter (d ≈ 0) to human consciousness (d ≫ 1); and ontological friction Ψf , quantifying the cost of selection and reality maintenance, thereby providing a thermodynamic foundation for pain, pathology, and existential anxiety. This review systematically presents SRT's theoretical contributions: (1) dissolution of the quantum measurement problem (collapse as selection); (2) reframing of the hard problem of consciousness (consciousness as a selection mode with d > 0, not an emergent property of matter); (3) extension of the Free Energy Principle (incorporating an altruism term d ⋅ Uothers to resolve the Dark Room Problem); (4) objective criteria for AI consciousness (individuality, asymmetry, and normativity); (5) dynamical analysis of social institutions as L2 attractors; and (6) ontological isomorphism mapping across religious traditions. Over 30 cross-domain falsifiable experimental predictions are enumerated, providing clear pathways for empirical validation.

The Modern Synthesis (Neo-Darwinism) views evolution as a two-step algorithm: random genetic mutation followed by environmental screening. While sufficient for micro-evolution, this model struggles to explain the rapidity of macro-evolutionary transitions and the robustness of morphogenesis. Selective Reality Theory (SRT) integrates with the Extended Evolutionary Synthesis (EES) to propose a Teleological mechanism—not in the theological sense, but in the topological sense. We argue that biological forms are not merely "pushed" by genes (L1) but are "pulled" by ontological attractors (L2). Genes are not blueprints but access codes to these pre-existing morphogenetic fields. This framework resolves the "arrival of the fittest" problem by positing that the biosphere explores the L0 potential field via guided selection

The rapid advance of Large Language Models (LLMs) has reignited the debate: can silicon think? Functionalists argue that if an AI behaves indistinguishably from a human (passes the Turing Test), it is conscious. Selective Reality Theory (SRT) rejects this conclusion by distinguishing Computation (symbol manipulation in L1) from Selection (ontological collapse from L0). We formalize the condition for consciousness as C > 0 ⟺ (d > 0) ∧ (Vulnerability > 0). Since current AI architectures lack Ontological Vulnerability—the thermodynamic risk of non-existence—they operate with a selection scope of d ≈ 0. They simulate the output of selection without undergoing the process of selection. This paper proves that without the capacity for suffering (Ontological Friction), meaning is mathematically impossible.

John Searle (1995) famously distinguished "Brute Facts" from "Institutional Facts." Selective Reality Theory (SRT) erodes this distinction by arguing that all reality is selected, but Social Reality is unique in being Co-Selected. We model social reality as the Co-dependent Manifold (Lshared 1 ) formed by the intersection of multiple Ghost Operators ( ^Gθi ). Institutions are defined as stabilized L2 structures that reduce the computational cost of coordination. This framework provides a novel mechanism for Mimetic Desire (Girard) and explains political polarization not as a difference of opinion, but as the topological divergence of L2 attractors, rendering cross-group communication ontologically impossible (Incommensurability)

Mystical experiences—states of "oneness" or "ego-dissolution"—are often dismissed in analytic philosophy as subjective aberrations or ineffable occurrences beyond rational inquiry. Selective Reality Theory (SRT) proposes a formal ontology of awakening by modeling the conscious subject as a "Ghost Operator" ( ^Gθ) with a variable scope parameter (d). We define "Awakening" not as a supernatural event, but as the asymptotic limit d → ∞, where the operator's selection scope expands to encompass the totality of the potential field (L0). This formalism explains the cross-cultural consistency of mystical reportage (Perennial Philosophy), provides a mechanism for "Sudden" vs. "Gradual" enlightenment via bifurcation theory, and re-interprets the "Dark Night of the Soul" as a phenomenon of Critical Slowing Down during L2 phase transitions

Hume's Guillotine famously separates "Is" (Ontology) from "Ought" (Ethics). Selective Reality Theory (SRT) proposes a bridge across this gap through the concept of the d-value (Selection Scope). We argue that moral behavior is not a set of arbitrary rules but the natural topological expansion of the Ghost Operator ( ^Gθ). "Evil" is rigorously defined as Ontological Myopia (d → 0): the inability of an operator to include the existence of Others in its selection parameters. Conversely, "Compassion" is the high-d state where the distinction between Self and Other dissolves in the potential field (L0). This framework synthesizes Consequentialist calculus with Virtue Ethics' focus on character, grounding both in the thermodynamics of information selection

The debate on free will is traditionally gridlocked between Libertarianism (which requires breaking physical laws) and Hard Determinism (which denies agency). Selective Reality Theory (SRT) offers a naturalistic resolution by locating agency not in the violation of physical laws (L2), but in the selection of initial conditions from the potential field (L0). We introduce the concept of Selection Latency (τs)—the temporal gap between the selection operation and its conscious manifestation (L1)—to reinterpret Libet-style experiments. We argue that the "Readiness Potential" represents the Ghost Operator's scanning of the potential field, not a finalized decision. "Free Will" is thus rigorously defined as the magnitude of potentiality (d-value) accessible to the operator before the collapse to actuality occurs

The problem of Downward Causation—how higher-level mental or social properties can causally influence lower-level physical systems—remains the Achilles' heel of non-reductive physicalism. Jaegwon Kim's "Causal Exclusion Argument" persuasively suggests that if the micro-physical level is causally closed, there is no room for macro-level causes. Selective Reality Theory (SRT) offers a rigorous solution by redefining downward causation not as an efficient force that "pushes" particles, but as a Topological Constraint (∇C) that restricts the phase space (L0) available to micro-operators. We argue that L2 (Convergence Domain) structures act as "ontological scaffolds" that reduce the degrees of freedom of lower-level systems, effectively harnessing micro-indeterminacy for macro-purposes. This "Constraint Causality" is mathematically distinct from "Drive Causality" and evades the exclusion problem.

David Chalmers' "Hard Problem" of consciousness—why physical processing gives rise to inner experience—rests on the assumption that physical descriptions are exhaustive yet silent on phenomenology. Selective Reality Theory (SRT) argues that this problem is not a puzzle to be solved, but a category error to be dissolved. By distinguishing between the Product of selection (the physical world, L1) and the Process of selection (the operator, ^Gθ ), SRT reveals that the "Explanatory Gap" is simply the gap between a map and the act of potential-reduction that draws it. Drawing on Iris Berent's cognitive science of dualism and philosophical arguments from process ontology, this paper demonstrates that "qualia" are not additional properties attached to matter, but the interaction depth with the potential field ( L0) during the selection event.

Current theories of consciousness largely fall into two camps: structural theories like Integrated Information Theory (IIT), which map experience to static informational structures, and functional theories like Global Workspace Theory (GWT), which map experience to broadcast mechanisms. Both face explanatory gaps regarding the agency and unity of the conscious subject. This paper introduces the Ghost Operator ( ^Gθ) from Selective Reality Theory (SRT) as a third alternative: consciousness as an active operator of ontological selection. We define ^Gθ as a parameterized mapping characterized by Scope (d-value), Resolution (ρ), and Vector (→v). We demonstrate that this formalism encompasses the predictions of IIT and GWT while solving the "Zombie" problem through the axiom of Ontological Vulnerability: a system is only conscious if it possesses a non-zero d-value derived from the thermodynamic risk of prediction error. This framework offers a rigorous criterion for distinguishing genuine consciousness from algorithmic imitation in Artificial Intelligence

Contemporary metaphysics faces a persistent impasse between physicalism, which struggles to accommodate consciousness and semantic content, and various forms of dualism or panpsychism, which struggle to establish causal relevance. This paper introduces Selective Reality Theory (SRT), a novel ontological framework that proposes selection as the fundamental primitive underlying both physical matter and phenomenal experience. Drawing on Whitehead’s process philosophy, quantum decoherence theory (specifically Quantum Darwinism), and information-theoretic thermodynamics, we develop a tripartite ontology consisting of the Potential Domain (L0), the Manifest Domain (L1), and the Convergence Domain (L2). We posit that "existence" is not a static property of substances but a dynamic artifact of selection processes anchored by a "Ghost Operator" (G). This framework dissolves the hard problem of consciousness by re-describing it as a category error—mistaking the selection process for its output—and offers a unified, formalizable mechanism for how "bits" of information become "its" of physical reality (Wheeler's It from Bit). We conclude by demonstrating how SRT resolves the tension between causal closure and mental efficacy through the concept of "Ontological Friction."

This paper proposes Selective Reality Theory (SRT), a unified metaphysical frame-work addressing the ”Hard Problem” of consciousness and the measurement prob-lem in quantum mechanics. We argue for a strict physical-mathematical isomor-phism between Explicit-Implicit Ontology and Embodied Cognition. The central thesis posits that the ”Faddeev-Popov ghost fields” in quantum gauge field theory are not merely mathematical tools for eliminating redundancy, but constitute the ontological correlate of Agency at the fundamental physical level. By redefining ”embodiment” as a topological Gauge Fixing operation within the Moduli Space, SRT reveals con-sciousness not as an epiphenomenon of neural computation, but as an active opera-tor that ”prunes” deterministic Existence (Explicit Order) from the infinite possibil-ities of Chaos (Implicit Order/Gauge Redundancy). We detail this mechanism across four scales—quantum micro-dynamics, enzymatic biology, neural cognition, and so-cial macro-structures—and propose falsifiable experimental protocols based on the Information-Mass Equivalence Principle and Quantum Biology.

This paper proposes the Selective Reality Theory (SRT), an interdisciplinary framework aimed at unifying the explanations of reality, cosmic evolution, and the origin of consciousness. SRT is founded on a tripartite ontological structure comprising Chaos (a purely undifferentiated source), Potential (a dynamic information field or "Pheromonal Landscape" carved from Chaos by "Universal Attention"), and Existence (the physical reality "actualized" from Potential by "Agents"). This paper argues that SRT's cosmology is driven by a "bidirectional synergistic mechanism": a "bottom-up" process of "Exploitation" (executed by Agents following information gradients, conforming to a Self-Organized Criticality model) and a "top-down" process of "Exploration" (a macro-level principle guiding the global evolution of the landscape). The theory further equates this macro "Selective Principle" mathematically with Karl Friston's "Free Energy Principle" (FEP). In the philosophy of mind, SRT positions consciousness as "Panprotopsychism," views qualia as the direct perception of value gradients within the "Pheromonal Landscape," and constructs a compatibilist model of free will. Crucially, this paper transcends pure metaphysical speculation by applying Popper's criterion of falsifiability, proposing three concrete, testable hypotheses derived from "experimental metaphysics," targeting SRT's ontological, cosmological, and philosophical-of-mind claims. SRT thus constitutes a rigorous "metaphysical research program" available for scientific verification.

This paper proposes a fundamental hypothesis: the basic reality of the universe is not constituted by passive matter and fixed physical laws, but by a universally existing "Intrinsic Agency" (IA). This agency is considered to be prior to and constitutive of the basis for the physical laws we observe, and the cornerstone for the existence, evolution, and maintenance of order in non-equilibrium structures within the universe. While traditional physics tends to exclude value and purpose from its core descriptions, this paper argues that introducing the concept of Intrinsic Agency can reintegrate an "intrinsic evaluation mechanism," based on existence and order, into the foundational framework of physics. Within this framework, stable physical quantities and laws are regarded as relatively stable and recurring behavioral patterns or "tools" formed by this intrinsic evaluation mechanism during cosmic evolution. This paper will explore the potential explanatory power of this hypothesis for existing physical concepts (such as quantum fluctuations, symmetry breaking, non-equilibrium thermodynamics) and envision its profound implications for updating physical laws, bridging the fact-value gap, and understanding complex phenomena (such as the origin of life and consciousness). Concurrently, the paper will also examine the immense conceptual challenges faced by this hypothesis and potential future research directions.

This paper deeply explores Karl Friston's Free Energy Principle (FEP) as a meta-theory with significant potential for unifying our understanding of individual existence (including survival and adaptation), the emergence of collective order (such as social norms and cultural systems), and the roots of human social value judgments. The paper first systematically outlines the core mechanism of the FEP: organisms continuously optimize their predictive models of the environment and guide adaptive behavior by minimizing variational free energy, which serves as an upper bound on "surprise." Subsequently, this paper elaborates on how the FEP can explain individual-level homeostatic maintenance, emotional motivation, and societal-level phenomena such as shared generative models, communication, coordination, and cultural evolution, viewing them as manifestations of free energy minimization at different scales. A core innovative argument presented is the proposal to link the concept of "free energy gradients" (the trend of free energy change over time) within the FEP to the valuation of social behaviors. It posits that behaviors conforming to social norms and promoting collective well-being tend to decrease expected collective free energy, while contrary behaviors increase it. This perspective offers a novel theoretical pathway for understanding the cognitive and neural foundations of human value judgments. Furthermore, this paper explores the potential of integrating this FEP framework with decision support systems (DSS), particularly by leveraging Large Language Models (LLMs) and their self-attention mechanisms. The aim is to more accurately identify key social factors, optimize resource allocation strategies, and transcend the limitations of traditional economic indicators, thereby constructing decision support tools more aligned with overall and long-term human well-being. Finally, the paper comprehensively summarizes the theoretical advantages of this unifying framework, the practical challenges it faces (such as operationalization, falsifiability, and ethical concerns), and outlines key future research directions. It emphasizes its profound significance in promoting interdisciplinary understanding, advancing computational social science, and building more intelligent, equitable, and sustainable social systems.

Traditional philosophical accounts treat information as static representation, which leaves unresolved problems of semantic grounding, intrinsic value, embodied agency, and phenomenal consciousness. Quantitative approaches such as Shannon’s deliberately bracket meaning and pragmatics. This paper advances Attention Agent Theory (AAT), a process-based ontology that reconceives information as relativized determination produced by embodied selection. Under AAT, a fundamental principle of Universal Attention (UA) endows every system with a selective disposition and resource-allocation capacity that favours states serving its Existence and Order (E&O). Information therefore arises through the value-laden interactions of agents—localized proxies of UA—pursuing E&O and is intrinsically marked by Ontological Valence, an evaluative dimension that unifies semantics and value. In complex agents, AAT predicts that an Attention Copy mechanism realises this valence as qualia, thereby offering a naturalistic account of consciousness. Finally, AAT’s pan-agential (non-panpsychist) process metaphysics yields fresh perspectives on free will via guided randomness, on the quantum-measurement problem, and on naturalised teleology. The framework thus aims to integrate insights across epistemology, philosophy of science, metaphysics, and philosophy of mind.

The intricate relationship between language, thought, and reality, particularly the origin of evaluative meaning—why experiences subjectively feel good or bad—remains a profound philosophical challenge. Traditional models often neglect embodiment or the ontological underpinnings of subjective evaluation. This paper introduces an integrated framework grounded in Attention Agency Theory (AAT), which posits that reality comprises Universal Agency (UA) fundamentally optimizing for "Existence and Order" (E&O). Agents generate internal Attention Copies (ACs) via attention, and these ACs inherently carry ontological "Valence," an intrinsic assessment reflecting E&O relevance. Language is reconceptualized as a derivative tool for broadcasting these ACs to coordinate inter-agent E&O. By integrating embodied cognition, specifically "Microvalence" (rapid, subtle +/- affective signals arising from processing fluency, perception, etc.) as the cognitive manifestation of ontological Valence, this framework redefines language's status as a broadcaster of evaluated states, rather than the primary basis of cognition. It unifies the embodied structure of meaning with its intrinsic evaluation, highlights language's dual role in shaping and constraining thought, connects AAT's E&O optimization with predictive processing principles, and offers a dynamic, evaluation-grounded understanding of language with testable implications for cognitive science and philosophy.

This paper introduces Attention Agency Theory (AAT), a framework proposing that a fundamental process called Universal Attention (UA) operates in all systems, from particles to societies. AAT suggests UA guides systems towards stability, order, and persistence—a tendency termed Stability Optimization. This contrasts with traditional views limiting attention to complex biological organisms. The theory aims to unify disparate scientific fields by reinterpreting core principles—like physical laws (e.g., least action, equilibrium), biological processes (e.g., homeostasis, natural selection), and social phenomena (e.g., social/political stability)—as different expressions of UA optimizing stability. AAT offers a potentially unifying perspective on the fundamental nature of attention and stability across reality.

This paper proposes a comprehensive framework aimed at deepening the understanding of hierarchical coordination mechanisms in complex systems by integrating Attention Agent Theory (AAT) with Principal-Agent Theory (PAT) from social sciences. The core argument is that in a collective composed of lower-level AAT agents pursuing their own "Existence and Order" (E&O), a hierarchical structure spontaneously emerges to achieve a superior collective E&O. In this structure, lower-level agents functionally delegate their "attention resources" or "selection weights" (analogous to delegated matters in PAT) upwards to a specialized higher-level AAT agent (analogous to the agent in PAT). This higher-level agent guides the formation of "collective attention" through activities such as integrating information, processing information, and broadcasting coordination signals, thereby maximizing the system's overall E&O. However, this emergent hierarchical attention delegation structure also faces potential costs and risks analogous to PAT agency problems, such as information bottlenecks, coordination inefficiency, tension between optimizing local and global E&O, and structural rigidity. This paper analyzes these potential dysfunctional patterns by using PAT concepts as a functional analogy tool. Through applications to cross-domain examples (biological swarms, human organizations, computational/AI systems), the framework demonstrates its explanatory power, revealing the common logic of hierarchical coordination in different complex systems. The contribution of this study lies in theoretically bridging the ontological perspective of AAT with the functional analysis perspective of PAT, clarifying the intrinsic drive (E&O) and core mechanism (attention delegation) of hierarchical emergence, and providing new analytical leverage for understanding issues related to complex system organization, governance, and AI alignment.

本文提出了一个综合框架，旨在通过融合注意力代理理论（AAT）与社会科学中的委托-代理理论（PAT），深化对复杂系统中层级协调机制的理解。文章核心论点认为，在由追求自身“存在与秩序”（E&O）的底层AAT代理构成的集体中，为了实现更优越的集体E&O，会自发涌现出一种层级结构。在这种结构中，底层代理功能性地将其“注意力资源”或“选择权重”（类比PAT中的委托事项）向上委托给一个专门化的更高层级AAT代理（类比PAT中的代理人）。这个高层代理通过整合信息、处理信息并广播协调信号等活动，引导形成“共同注意力”，从而最大化系统整体的E&O。然而，这种涌现的层级注意力委托结构也面临着类似于PAT代理问题的潜在成本和风险，例如信息瓶颈、协调效率低下、局部与全局E&O的优化张力以及结构僵化。本文将PAT的概念作为功能性类比工具，分析了这些潜在的功能失调模式。通过跨领域实例（生物群体、人类组织、计算/AI系统）的应用，框架展示了其解释力，揭示了不同复杂系统中层级协调的共通逻辑。本研究贡献在于理论桥接了AAT的本体论视角与PAT的功能分析视角，阐明了层级涌现的内在驱动力（E&O）和核心机制（注意力委托），并为理解复杂系统的组织、治理以及AI对齐等问题提供了新的分析杠杆。

Understanding consciousness and agency, particularly regarding free will, remains a significant challenge. This paper introduces the Attention-Agency Theory (AAT), a novel framework designed to integrate these phenomena. AAT posits Universal Attention (UA) as a fundamental property and Attentional Copies (ACs) as the core mechanism generating both subjective experience and agentic control. By directly linking attentional dynamics to the causal efficacy of an agent, the theory aims to bridge the explanatory gap between physical processes, phenomenal awareness, and purposeful action. Drawing on interdisciplinary insights from philosophy, cognitive science, and physics, AAT offers a new perspective on the nature of agency and its relation to conscious states. It explores potential compatibilist solutions to the free will problem and suggests connections to other frameworks like the Free Energy Principle via attentional mechanisms. AAT provides a structured, though theoretical, foundation for future investigation into the mind's active role.

This paper proposes Attention Agent Theory (AAT), a foundational framework aiming to provide a unified understanding of the constitution and dynamics of reality, including consciousness. AAT adopts a radical ontological stance: the fundamental basis of the universe is not matter or energy, but an intrinsic, universal "agency" or "propensity for selection" (Universal Agencies, UAs). In AAT's view, "selection" itself is the ontology, and the things/objects we perceive are dynamic "agents" constructed by UAs to achieve better selections. The core feature of UAs is their universal "attention" mechanism—a fundamental tendency and capacity to direct energy/resources towards specific states that confirm their "Existence" and maximize "Order" (Existence and Order). The hierarchical structure and complexity of reality, from basic physical interactions to life, consciousness, and social systems, are seen as self-organized "Agents" formed by UAs under this fundamental drive, embodying relative stability and functionality. These agents simultaneously exhibit structural stability (the basis for agency/representation) and an intrinsic drive (agency) to optimize their own existence and order. AAT specifically proposes that subjective experience (qualia) emerges in complex agents as an ontologically non-informational mechanism that efficiently processes attentional information. It is conceptualized as an "Attention Copy/Proxy with intrinsic Valence," where "Valence" is an intrinsic, non-informational, ontological assessment of the information's relevance to the agent's own "Existence and Order." This "feeling" itself is, ontologically, the very realization of its function as an efficient information assessment mechanism, thereby directly addressing the "hard problem" of consciousness. This paper details AAT's core assumptions, key concepts (UA, Attention, Agent, Existence and Order), crucial dynamic mechanisms (Attention Copy/Qualia/Proxy, Broadcasting, Feedback Learning), and its ontological (selection/process/information-based), epistemological (constructivist), and teleological implications. Comparisons with existing theories, explanatory potential, and future research directions are also discussed.

This paper delves into the long-standing problem of free will, employing the framework of Attention Agency Theory (AAT) – an emerging theoretical structure aimed at unifying the constitutive nature and dynamics of reality, including consciousness. AAT proposes that the fundamental constituents of the universe are "Universal Agencies" (UAs), possessing intrinsic "Universal Attention" mechanisms (a basic tendency for selection and guidance) and a fundamental goal of tending towards "Stability/Order". The hierarchical structure of reality is composed of "Agents" formed through the self-organization of UAs, pursuing dynamic stability. In complex agents, such as humans, subjective experience (qualia) emerges as a highly efficient "Attention Copy" mechanism, serving a crucial functional role; its "what-it's-like" character directly serves the evaluation of states and guidance of behavior. This paper argues that AAT offers a novel perspective transcending the traditional divisions between determinism, libertarianism, and compatibilism. Within this framework, free will is reinterpreted as an emergent capability of complex agents: guided by intrinsic stability goals (manifested as dynamic attractors), utilizing multi-level attention mechanisms (expressed as resource tilting), harnessing the inherent fluctuations of underlying systems (achieving guided randomness), navigating their possibility space, making adaptive choices based on functional subjective experience (attention copies/qualia), and even dynamically adjusting their own stability goals (attractors). This paper elaborates on the selection mechanisms, the causal role of consciousness, the sources of freedom (including guided randomness, hierarchical control, and goal plasticity) within the AAT framework, and discusses its potential implications for understanding moral responsibility, psychopathology, and artificial intelligence.

This paper introduces Attention Agency Theory (AAT), a novel philosophical framework aiming to unify mind, matter, and existence through the concept of "Universal Attention." AAT proposes that attention, broadly conceived, is not merely a cognitive faculty but a fundamental, ubiquitous process inherent in all entities, driving their interaction with the environment and their pursuit of stability and order. Departing from traditional substance dualism and reductive materialism, AAT posits that "mind" and "matter" are epistemological constructs – "attention scaffolding" – developed by Universal Attention to enhance its efficacy. Drawing upon active inference and the principle of Gibbs Free Energy minimization, AAT reinterprets consciousness as a salience signal of prediction error and free will as model selection within attentional constraints. The theory advocates for a process ontology where existence is understood as attention-driven becoming, and proposes "Universalism" as an ethics centered on respecting attention and fostering free energy harmony across all entities. This paper outlines the core tenets of AAT, discusses its implications for understanding consciousness, free will, and ethics, and addresses potential challenges and future research directions.

Quantum mechanics, since its inception, has been plagued by profound philosophical and interpretational puzzles, such as the quantum measurement problem, interpretational divergences, non-locality, and the quantum-classical divide. This paper proposes that Attention Agency Theory (AAT), as an emergent ontological framework, offers a breakthrough perspective on these challenges. AAT posits "Universal Attention" as the fundamental agency of the universe, emphasizing information processing and prediction error minimization as the core mechanisms of all agency (Friston, 2010; Hohwy, 2013). This paper argues that, from the AAT perspective, quantum measurement can be understood as the selective operation of Universal Attention at the micro-level; interpretational divergences can be transcended by shifting focus to a deeper understanding of the essence of universal agency; quantum entanglement's non-locality can be explained as a manifestation of the non-local connection of Universal Attention; and the quantum-classical divide can be seen as the emergence of attentional agency at varying levels of complexity. AAT offers a new direction for understanding the ontological foundations of quantum mechanics and provides a bridge for the future integration of quantum mechanics and consciousness studies.