Study: Consciousness as recursive information alignment in self-sustaining systems

Summary

A testable model of Holistic Information Theory proposes that consciousness arises from recursive information matching within self-sustaining clusters. This model suggests that consciousness is a dynamic process involving memory, evaluation, and adaptive positioning rather than merely a possession of beings. It distinguishes between functional, self-model, and phenomenal consciousness and introduces a Recursive Alignment Index (RAI) for empirical testing. The framework aims to link various fields such as neuroscience, AI research, and the theory of consciousness.

Study at a glance

Abstract

A testable framework model of Holistic Information Theory (GIT) Revised Scientific 2.0 Version with testable framework, measurable variables and interdisciplinary research programme (Work III) Author: Dieter LiedtkeYear: 1970 - 2026Licence: CC BY 4.0 Abstract The question of consciousness is one of the fundamental unsolved problems of science. Despite significant advances in neuroscience, cognitive science, computer science, evolutionary biology and physics, there is still no generally accepted theory that explains why systems not only process information, but also develop subjective perspective, self-reference, learning ability, meaning and future orientation from it. Within the framework of Holistic Information Theory (HIT), this study formulates a testable model in which consciousness is understood as recursive information matching within self-sustaining clusters. A cluster is any structured unit that receives information, compares it with stored states, evaluates it, integrates it and makes it usable for future actions or state changes. The basis is the thesis that consciousness arises gradually where new information is linked to memory, self/non-self distinction, evaluation and adaptive positioning . The study distinguishes between functional consciousness, self-model consciousness and phenomenal consciousness. A Recursive Alignment Index (RAI) is proposed for empirical testing. The model does not claim that atoms, molecules or technical systems possess human consciousness. Rather, it proposes a scale of precursors and degrees of organisation in consciousness-like information processing. This creates a research framework that links neuroscience, systems biology, AI research, theory of consciousness and information physics. 1. Introduction In modern research, consciousness is usually investigated from three perspectives: firstly as a neural product of the brain, secondly as a subjective experience, and thirdly as an information-processing integration process. The present study primarily expands upon the third approach. It assumes that information is not merely passively stored or processed, but organises itself, remembers, corrects, networks and becomes effective in self-sustaining systems. The basic thesis is: Consciousness is the recursive alignment of new information within a self-sustaining cluster, utilising memory, experience, self/non-self distinction and future positional choice. This definition ties in with the GIT thesis formulated in the source document, according to which consciousness is understood as “the comparison of new information within a cluster, utilising memory and experience to maintain and further develop the system”. 2. Problem Statement Consciousness research faces several unresolved problems: Why does subjective perspective arise from neural activity? How is a multitude of individual pieces of information integrated into a unified experience? Why does a sense of self persist despite constant material change? What role do memory, evaluation and self-boundaries play? Are there rudimentary forms of consciousness in non-human or non-neuronal systems? Can AI systems develop functional structures of consciousness? Is consciousness an abrupt leap or a gradual organisational process? The GIT does not answer these questions with a ready-made metaphysical assertion, but through a research model: consciousness should not be understood primarily as a possession of a being, but as a process. 3. Key Concepts 3.1 Information In this model, information means not only data, but any distinguishable difference in state, relationship or meaning that can have an effect within a system. 3.2 Cluster A cluster is a distinct but interconnected unit of elements that can exchange, store, regulate or transform states. The source document cites as examples particle structures, molecular assemblies, biological cells, neural networks, individuals, social communities and artificial systems. 3.3 Memory Memory is stored information of past states. It can be genetic, immunological, neuronal, cultural or technical. 3.4 Experience Experience is not merely storage, but organised memory with an evaluative function. Experience means: recognising patterns, reducing errors, improving decisions and preparing for the future. 3.5 Ability to take a position Consciousness is not merely perception, but the ability to adopt new positions within the information space: changing perspectives, recognising errors, reordering relationships and generating innovation. 4. Core model: Consciousness as recursive information comparison The model can be represented in a process formula: Information → Networking → Cluster → Alignment → Memory → Experience → Self-boundary → Position selection → Degree of consciousness The source document summarises this idea in a similar way: information generates networking, networking generates alignment, alignment generates memory, memory generates perspective, perspective generates consciousness, consciousness generates new information. Consciousness is thus understood as a dynamic process. It arises not solely from the quantity of information, but through the recursive linking of information to a self-sustaining system. 5. Three levels of consciousness 5.1 Level I: Functional consciousness This level describes measurable precursors: Information intake Comparison with previous states Error correction Adaptive response Self/non-self discrimination Examples: immune system, cells, plant responses, swarm coordination, simple autonomous systems. 5.2 Level II: Self-model awareness Here, a system possesses a more stable model of its own boundaries and capacity for action: Inner/outer distinction Self-state model Goal orientation Prioritisation Model of the future Agency Examples: animals, humans, complex autonomous systems. 5.3 Level III: Phenomenal consciousness This level concerns subjective experience: Pain Colour Sense of time sense of self Internal perspective This study does not claim that phenomenal consciousness is fully explained. However, it proposes a bridge: phenomenal consciousness could be a highly integrated form of recursive information alignment. 6. Recursive Alignment Index (RAI) A Recursive Alignment Index is proposed for empirical testing. 6.1 Definition The RAI measures the degree to which a system recursively links new information with memory, self-boundaries, evaluation and future action. 6.2 Variables Information intakeDiversity and relevance of incoming signals. Memory integrationAbility to compare new states with stored states. Error correctionAbility to detect deviations and correct behaviour. Self/non-self discriminationThe ability to distinguish one’s own states from environmental states. Evaluation / RelevanceThe ability to rank information according to its importance, danger, benefit or future value. Future modellingThe ability to simulate or prepare for possible future states. Adaptive positioningAbility to realign behaviour, structure or strategy. 6.3 Scale RAI 0: no discernible recursive integration RAI 1: simple stimulus-response RAI 2: memory-based adaptation RAI 3: Self/non-self comparison RAI 4: flexible self-model RAI 5: reflexive consciousness with symbolic self-description 7. Falsifiable hypotheses Hypothesis 1 Systems with recursive memory matching demonstrate better adaptation to changing environmental conditions than systems without memory matching. Hypothesis 2 Markers of consciousness in the brain correlate more strongly with recursive self-reference and network complexity than with mere activity levels. Hypothesis 3 Non-neuronal biological systems can exhibit low but measurable RAI values. Hypothesis 4 AI systems with a persistent self-state model exhibit higher functional consciousness-likeness than stateless models. Hypothesis 5 Under certain conditions, collective systems can develop a higher functional RAI than individual agents. Hypothesis 6 If a system integrates information but lacks a self/non-self distinction, no stable self-model consciousness arises. Hypothesis 7 Phenomenal consciousness only arises when functional integration, self-model, memory and evaluative relevance exceed a certain threshold of complexity. 8. Research tasks for confirmation 1. Single-cell information matching Investigation into whether individual cells not only respond mechanically to stimuli, but also compare new information with stored states. 2. The immune system as a self/non-self model Analysis of whether immunological memory processes can be interpreted as a biological precursor to self-modelling. 3. Slime moulds and non-neuronal navigation Examination of whether non-neuronal organisms form decision-making pathways via memory, error correction and optimisation. 4. Plant communication Investigation into whether plants store and evaluate information via networks and translate it into future responses. 5. Animal agency Measuring the point at which animals do not merely react, but model their own states. 6. States of human consciousness Comparison of waking consciousness, dreaming, anaesthesia, meditation and coma using the RAI. 7. Brain network coherence Examining whether conscious experience correlates with recursive global integration. 8. AI self-modelling Comparison of AI systems with and without persistent self-description. 9. Swarm consciousness Investigation of collective decision-making processes in bees, ants, birds, humans and digital multi-agent systems. 10. Art and visual modelling Investigation into whether complex abstract models of consciousness are better understood, remembered and further developed through images. 9. Research tasks for refutation 1. Stimulus-response reduction If all supposed

Tags