Submitted:
27 August 2026
Posted:
28 August 2026
You are already at the latest version
Abstract
This paper provides a presentation and structural overview of the book Evolving Intelligence. The Generative Path from Information and Computation to Cognition and Intelligence. The monograph is forthcoming with World Scientific in 2026. Evolving Intelligence develops a naturalistic, generative framework for understanding how intelligence emerges from the organization of nature. Rather than beginning with human reasoning or with artificial intelligence, the book traces a continuous path from physical and chemical organization in nature through life and cognition to human symbolic culture, artificial intelligence, and hybrid human-AI systems. Its central framework, ICON (Information-Computation-Cognition Naturalism), connects these domains through morphology, interaction, observer-agency, information, computation, cognition, consciousness, and intelligence.The guiding question is not simply “What is intelligence?” but “How can intelligence arise at all within nature?” The book therefore shifts attention from intelligence as an isolated human faculty to the generative processes through which increasingly complex forms of organization, sensing, memory, anticipation, agency, meaning, and problem solving emerge.
Keywords:
information
; computation
; cognition
; consciousness
; intelligence
; AI
; computer science
; computing
; information studies
Nature as evolving morphology and observer-agent networks
The first distinctive move is to treat nature as evolving physical morphology: structured matter-energy whose organization is continuously transformed through interaction. At any level of organization, systems that can affect and be affected by other systems function as observer-agents. Observation is the state change induced by an incoming signal; agency is the capacity to produce outgoing changes. Observation and agency are therefore two sides of the same interactional process.
This concept generalizes the observer beyond the human epistemic subject. Physical and chemical systems instantiate minimal observer-agency through selective interaction, without thereby being cognitive or conscious. With life, observer-agency becomes autopoietic, viability-regulated, and therefore cognitive. Human and technological observer-agents are later, more complex realizations within the same evolving natural world.
Information and computation as complementary aspects of morphology
ICON framework gives information and computation a common physical grounding. Morphology exists independently of any particular observer-agent and constitutes potential information. A difference becomes actual information when it becomes effective for a particular observer-agent in interaction. Information is therefore relationally actualized morphology: physically grounded, but agent-relative in its realization.
Computation is the transformation of morphology/information through time. Information and computation are thus complementary categories: information is structure or being, computation is process or becoming. Structure constrains future transformations, while transformations generate new structure. Their relation is recursive.
This is based on a general concept of natural computation with nature as a network of networks of computational processes. Digital Turing-style computation is a model of one specialized, engineered regime. Natural computation may be continuous, distributed, concurrent, asynchronous, morphological, self-modifying, biological, virtual, or hybrid. Different computational regimes are distinguished by their substrate, organization, memory, temporality, closure, and source of goals.
The transition from computation to cognition
The book makes a clear distinction between computation in general and cognition. Not every information-processing system is cognitive. Cognition begins when information and computation are integrated into autopoietic, viability-regulated living organization. The living cell with its sensing and problem solving, goal directed behavior provides the basic threshold for cognition.
The book adopts the idea that cognition is coextensive with life, but develops it explicitly as scale-relative. Cellular cognition, somatic cognition, neural cognition, human symbolic cognition, and collective cognition differ in mechanisms, integration, temporal horizon, memory, and problem space. This avoids both anthropocentrism and the opposite mistake of calling every sufficiently complex computation “cognition.”
Consciousness and intelligence as distinct dimensions of living cognition
The book addresses processes and structures involved in cognition, consciousness, and intelligence. Cognition is viability-regulated living activity. Consciousness is rooted in embodied self-sensing: the organism registers its own condition in relation to what supports or threatens continued existence. At the most basic biological level, this appears as valence-based self-sensing, a minimal “feeling of existence.” Later on in evolution, nervous systems integrate and expand this into richer perceptual, emotional, and reflective forms.
Intelligence, in its minimal form, is flexible problem solving and goal-directedness. It is the capacity of a cognitive system to reorganize morphology, memory, action, tools, environment, and relational or problem space in response to changing conditions and goals. Intelligence is an organizational capacity that develops in degrees.
Artificial intelligence as real, synthetic, and derivative
This distinction makes it possible to characterize contemporary AI without either dismissing or anthropomorphizing it. Artificial systems can be genuinely intelligent in the sense of performing powerful problem solving, while lacking biological cognition and consciousness. Their goals, relevance structures, evaluation criteria, and conditions of operation are externally supplied by human beings and sociotechnical institutions. Current AI is therefore synthetic and derivative intelligence: real as problem-solving capacity, but not autopoietic, self-grounded, or viability-regulated.
Human intelligence and AI form hybrid intelligent systems. AI operates on externalized products of human cognition: language, texts, images, code, models, institutions, and accumulated cultural knowledge, and can reorganize these at a new scale. This places AI within the natural history of intelligence without equating it with living cognition.
Evolution as expansion of complexity through agency, memory, and problem space
Evolution is interpreted as the historical transformation of living observer-agency. Morphology embodies accumulated interaction history as memory. Memory shapes anticipation; anticipation guides agency; agency modifies organisms and environments; and these changes alter future possibilities. Organisms therefore participate in constructing the spaces in which selection operates. Natural selection remains essential, but living systems also generate and reorganize the variation, developmental possibilities, and niches on which selection acts.
Intelligence evolves through this expansion of organization. As communication, memory, anticipation, and coordination extend across larger spatial, temporal, social, and conceptual domains, systems acquire larger problem spaces and more flexible ways of navigating them. Human language and culture externalize memory and meaning, allowing cognition to become collective and cumulative across generations.
A generative architecture connecting the domains
The book’s main novelty lies in the architecture that connects different steps in the evolution of intelligence. ICON proposes a generative sequence in which later organizational regimes arise through transformations and recursive integrations of earlier ones:
Morphology → memory → computation → interaction → information → agency → life/cognition → consciousness and intelligence → language and collective cognition → artificial and hybrid intelligence → morphology
The sequence is not a simple one-way chain. Each stage feeds back on the conditions that produced it. Processes stabilize into structures; structures constrain subsequent processes. Living systems actively reconstruct themselves; cognition reorganizes environments; culture transforms cognition; and AI now feeds back into human symbolic and technological organization. The book therefore describes intelligence as an evolving, recursively organized phenomenon within nature.
Contributions of the book
Intelligence as an Evolving Natural Phenomenon
Evolving Intelligence: The Generative Path from Information and Computation to Cognition and Intelligence develops a naturalistic framework for understanding intelligence as the outcome of a long history of evolving organization. Its point of departure is that intelligence cannot be adequately understood as an isolated capacity appearing suddenly in human beings or, more recently, in artificial systems. Rather, its origins must be traced through the successive organizational transformations by which physical processes give rise to increasingly complex forms of interaction, memory, agency, life, cognition, consciousness, intelligence, culture, and technology.
The central contribution of the book is the Information–Computation–Cognition Naturalist framework, ICON, which provides a common conceptual architecture for describing this generative continuity while preserving the differences among physical, biological, cognitive, artificial, and hybrid systems.
Nature as Evolving Morphology
The constructive (generative) framework begins with nature understood as evolving physical morphology. Morphology is used in a broad physically grounded sense: it refers to organized structure at a particular level of description, together with the physical properties that constrain possible interactions and transformations.
Nature is therefore not conceived as passive matter upon which organization is subsequently imposed. Physical systems already exhibit structured dynamics, constraints, path dependence, material memory, and the capacity to generate new organization through interaction. Chemistry enriches this basic physical organization through molecular specificity, selective bonding, catalysis, reaction networks, and combinatorial construction. Life introduces a further organizational transition when these physical and chemical resources become recursively integrated into autopoietic systems capable of maintaining their own existence.
Observer-Agency as the Basic Relational Principle
A central conceptual innovation of ICON is the observer-agent. Classical scientific descriptions often presuppose an observer without explaining how observers themselves arise within nature. ICON instead treats observer-agency as a natural phenomenon that develops across levels of organization.
An observer-agent is not primarily a conscious spectator. In its minimal form, it is a system that can be changed through interaction and can, through its changed state, induce changes in other systems. Observation and agency are thus understood as complementary aspects of interaction. An incoming signal induces a state change in a system, which constitutes observation. The resulting reorganization may produce action and outgoing signals, which constitutes agency. The observer-agent is therefore simultaneously affected by and affecting its environment.
Information as Relationally Actualized Morphology
The notion of observer-agency makes it possible to formulate information relationally without abandoning realism. Physical morphology exists independently of any particular observer-agent. However, a physical difference becomes information only when it becomes effective in interaction for some organized system.
Information is therefore not a free-standing substance or an abstract entity distributed throughout the universe. It is relationally actualized morphology: structured difference that makes a difference for an observer-agent. The same physical world may consequently constitute different informational worlds for different agents because their morphologies, histories, capacities, and modes of interaction differ. A chemical gradient, for example, has one informational significance for a bacterium, another for a scientific instrument, and none at all for a system unable to register it.
Morphological Computation as Process and Transformation
Computation is introduced as the inherent property of morphology and processual counterpart of information. If information concerns structured organization, computation concerns its transformation through time. Information and computation can consequently be understood as complementary aspects of the same evolving morphology: information as structure or being, computation as process or becoming.
This differs both from views that treat computation exclusively as formal symbol manipulation and from digital ontologies that identify nature with a universal digital computer. Turing computation remains fundamental within its proper domain, but it is one computational regime among many. Natural computation includes distributed, concurrent, continuous, morphological, chemical, biological, quantum, interactive, and hybrid transformations in which the physical organization of the system participates directly in the information processing.
The relation between information and computation is recursive. Structure constrains what transformations are possible, while transformations generate new structure. The accumulated consequences of previous interactions become embodied in morphology as material memory, and this changed morphology conditions subsequent interactions.
From Memory to Life
History thereby becomes physically effective. Memory does not first appear with brains or symbolic representation. Its precursors exist in path-dependent physical and chemical organization, while living systems transform such historical traces into actively maintained resources for regulation, anticipation, adaptation, and learning.
Morphology can therefore be understood as accumulated interaction history, while computation is the continuing process through which that history is extended and reorganized.
The emergence of life marks a decisive organizational threshold. ICON does not equate all information processing with cognition. Physical and chemical systems may exhibit information, computation, memory-like effects, and minimal observer-agency without being cognitive. Cognition begins when these processes become recursively integrated into an autopoietic organization whose interactions are regulated relative to its own continued existence.
The cell is therefore the fundamental biological observer-agent. It senses, regulates, repairs, maintains boundaries, communicates, reorganizes its internal processes, and acts in ways that contribute to its persistence. At this level, differences in the environment no longer merely produce causal effects; they acquire relevance because some conditions support the continuation of the living system while others threaten it.
Cognition as Life at a Given Level of Organization
ICON adopts and develops the idea that cognition is coextensive with life. Its distinctive formulation is explicitly scale-relative: cognition is the process of life at a given level of organization.
Cellular cognition, multicellular coordination, neural cognition, human symbolic cognition, and collective cognition are not identical phenomena, but they instantiate progressively more complex organizational forms of sensing, memory, anticipation, regulation, and action.
This avoids two symmetrical errors. The first is anthropocentrism, according to which cognition begins only with brains, language, or reflective reasoning. The second is an unrestricted computationalism in which every sufficiently complex information-processing system is described as cognitive. Or panpsychism in which the universe as a whole is conscious. ICON locates the threshold in autopoietic, viability-regulated living organization.
Consciousness as Living Self-Sensing
The biological grounding provides the basis for the book's account of consciousness. Rather than beginning with reflective human awareness, ICON asks what the minimal biological preconditions of consciousness might be.
Its proposal is that basal consciousness consists in viability-regulated self-sensing: the capacity of a living system to register its own condition in terms of valence, that is, in terms of what is favorable or unfavorable for its continued organization. In this minimal sense, consciousness is the feeling of existence of a living system.
Nervous systems do not create self-sensing from nothing; they integrate, accelerate, differentiate, and extend capacities already present in living organization. Neural and reflective consciousness are consequently higher-order developments of a more fundamental biological self-relation. The book presents this explicitly as a hypothesis concerning the evolutionary roots of consciousness, rather than as a settled result of consciousness science.
Intelligence as Flexible Problem Solving and Goal-Directed Behavior
Intelligence is distinguished both from cognition and from consciousness. In living systems, intelligence is treated as a capacity of cognition: the flexible ability to solve problems by reorganizing morphology, memory, action, tools, environment, and relational space.
Intelligence therefore exists in degrees and in different forms. Its development can be understood as the progressive expansion of the spaces within which an agent can detect problems, anticipate possible states, coordinate actions, and generate alternative solutions.
This shifts attention away from intelligence as a static faculty or performance score toward intelligence as an organizational capacity for adaptive problem solving.
Artificial Intelligence as Synthetic and Derivative Intelligence
This distinction provides a conceptual basis for understanding contemporary artificial intelligence. ICON describes current AI as genuine intelligence but recognizes its fundamental difference from living cognition. Artificial systems possess real and powerful problem-solving capacities without being living or conscious. Their intelligence is therefore described as synthetic and derivative. It is synthetic because it is realized in engineered computational architectures; it is derivative because the goals, relevance structures, training materials, evaluation criteria, and social meanings within which those systems operate originate in living human cognition and human institutions.
Artificial intelligence can consequently be genuinely intelligent without possessing autopoietic cognition, intrinsic biological normativity, feeling, or consciousness.
This distinction is particularly important in the context of generative AI. Contemporary systems operate on vast bodies of externalized human symbolic production. Language, writing, mathematics, scientific theories, images, code, databases, and other cultural structures are products of collective human cognition accumulated across generations. Once externalized, these informational morphologies can be transformed computationally by systems that do not themselves possess the living organization from which their semantic and pragmatic grounding originally arose.
Artificial intelligence is not external to nature or opposed to natural intelligence. It is a recent development within the evolutionary history of nature, arising through living systems capable of constructing technologies that extend their own cognitive capacities.
Human, Collective, and Hybrid Intelligence
The book consequently situates human culture and technology within the same natural continuum. Human cognition expands beyond the biological individual through language, tools, institutions, scientific practices, and technological infrastructures. These constitute higher-order informational morphologies in which memory, models, and problem-solving capacities can persist beyond individual lifetimes. Contemporary AI represents a further stage in this externalization and reorganization of cognition.
Human–AI systems are therefore most naturally understood as hybrid cognitive architectures in which biological and artificial capacities are coupled without being conflated.
Evolution as the Expansion of Agency and Problem-Solving Capacity
A further contribution of the framework concerns evolution. Evolution is interpreted not only as the selection of variants but as the historical transformation of organized agency.
Living systems participate actively in generating the variations and environmental conditions upon which selection acts through development, plasticity, niche construction, cellular regulation, symbiosis, genomic restructuring, and other forms of organismal activity.
What evolves is therefore not only morphology but also ways of sensing, forms of memory, degrees of agency, temporal horizons, modes of coordination, and capacities for problem solving. Intelligence is one expression of this continuing expansion of organized interaction.
The Generative Path
Taken together, these ideas provide the generative architecture of the book:
Morphology → memory → computation → interaction → information → agency → life/cognition → consciousness and intelligence → language and collective cognition → artificial and hybrid intelligence → morphology
As already mentioned, the sequence should not be understood as a linear chain. Each stage emerges through recursive interaction with previous levels, while earlier organizational forms remain active within later ones. Structure generates process, process transforms structure, and the results of previous transformations become conditions for further development.
The book therefore describes intelligence not as a property added to nature, but as the outcome of increasingly complex forms of organization generated throughout natural history.
The Novelty of the ICON Framework
The novelty of Evolving Intelligence does not lie primarily in introducing each element of this generative sequence independently. Many of its components have important intellectual predecessors in physics, cybernetics, information theory, natural computing, autopoiesis, enactivism, evolutionary theory, basal cognition, morphological computation, and philosophy of information.
The distinctive contribution is their integration into a single generative naturalistic framework organized around observer-agency.
ICON makes it possible to describe continuity across nature without erasing organizational differences. Physical interaction is not cognition. Cognition is not identical with intelligence. Intelligence does not imply consciousness. Artificial intelligence need not reproduce biological organization in order to constitute genuine problem solving.
The observer-agent concept provides the linking mechanism. It connects information to interaction, computation to transformation, cognition to autopoietic viability, consciousness to living self-sensing, and intelligence to flexible problem solving. In this way, phenomena usually treated in separate disciplines become different organizational regimes within one evolving natural process.
Evolving Intelligence
The resulting picture is one of nature progressively generating systems capable not only of interacting with the world but of remembering, learning, regulating, solving problems, anticipating, constructing shared symbolic worlds, and eventually building artificial systems that participate in these processes.
Intelligence is thus neither a mysterious late arrival nor an abstract capacity detachable from its history. It is an evolving organizational achievement of nature.
In short, Evolving Intelligence offers a naturalistic account of how nature generates systems capable not only of acting and adapting, but eventually of knowing, reflecting, creating, and constructing new forms of intelligence.
References
- Dodig-Crnkovic, G. Evolving Intelligence. The Generative Path from Information and Computation to Cognition and Intelligence; World Scientific, 2026. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.