#neuraxon
neuraxon
Počet zobrazení: 1,115
Diskutuje: 24
Hot
Najnovšie
·
--
Článok
Measuring Machine Intelligence: The g Factor vs. ARC-AGI Benchmark#Neuraxon Intelligence Academy — Volume 10
By the Qubic Scientific Team
If we build an artificial system and want to know whether it is intelligent, what exactly do we measure? We think we know when we hear that ChatGPT-5 announces it has beaten DeepSeek and then that Claude sweeps Gemini.
But the question is still there, intact. Measuring artificial intelligence is not measuring speed or temperature. We have no unit of measurement, as strange as that may seem.
In psychology we have been dealing with this problem for over a century. Artificial intelligence has been at it for a decade. And it does so in a hurry, with a lot of money at stake and with a constant temptation: to declare victory.
The g Factor: A Single Number to Summarize General Intelligence
At the beginning of the 20th century, Charles Spearman realized that when a child performed well in one subject, they tended to perform well in the others, even if they were subjects with no apparent relation. The scores correlated with one another, all of them positively. He called that pattern the positive manifold, and he deduced that there must be a common latent factor behind all those disparate abilities: the factor g, or general intelligence (Spearman, 1904).
The idea is seductive. If all cognitive tests load onto a single factor, it is enough to extract that factor through factor analysis to have a summary measure of general capacity. In human practice, that first factor usually explains between 40 and 50 % of the variance in performance (Detterman & Daniel, 1989; Deary et al., 2009).
But watch out, because here lies the first trap. The g factor is populational. It does not measure the individual, but variance within individuals (Hernández-Orallo et al., 2021). To say that a specific subject has so much g is, strictly speaking, a mistake. g emerges when comparing many subjects, not when examining one. Like personality, you are the most extroverted of your age group. And you remain so at 50 relative to your group, even if in intensity you are less extroverted than at 20.
What Does IQ Really Measure? Understanding Intelligence Scores
But then, what does IQ measure?
It measures a relative position. The scale is calibrated on a sample with mean 100, standard deviation 15. An IQ of 130 is not an absolute amount of intelligence stored inside someone's head; it is the assertion that this person is two standard deviations above the mean of their normative group. The number is attached to the individual, yes, but its meaning is populational. It is a position in a ranking, not a content.
Your height is absolute: you are 180 centimeters tall even if you are the last human being on Earth. Your IQ is not: being above the mean requires a mean, and a mean requires others. No one can be more intelligent than the average on a desert island.
Now one understands why transferring this to AI is so delicate. When someone computes a g for a set of large language models (LLMs), that factor is an artifact of the set they chose. We are measuring a position in a table, and we present it as if it were an internal property of the system.
Applying the g Factor to Artificial Intelligence: A Dangerous Temptation
The temptation to transfer all of this to AI was irresistible. Gignac and Szodorai proposed that, if the performance of models across varied tasks correlates positively, it should be possible to identify a general factor of capacity in artificial systems as well. And indeed, several recent works apply factor analysis to test batteries in LLMs and find a unidimensional g factor that remains stable across models, batteries and extraction methods (Ilić, 2023). It sounds like confirmation. It is wise to be suspicious.
The appearance of a dominant first factor does not prove that there exists a general capacity analogous to the human one. It proves that the scores of those models covary. And they covary for a very shallow reason: they share architecture, they share training corpus, they share optimization recipes. A large, well-trained model does everything better than a small, poorly trained one, across all tasks at once. That is enough to manufacture a beautiful positive manifold that tells us nothing about cognitive generality. It tells us about the scale of computation. WATCH OUT: The factor we extract may simply be a factor of size disguised as intelligence.
The brain, moreover, does not concentrate intelligence in a single module. A multitude of specialized subsystems process in parallel and, when a piece of information wins the competition, it becomes globally available to the rest of the system, which can then recombine it for new purposes (Baars, 1988; Dehaene & Changeux, 2011). What we call generality is global availability: putting a piece learned in one context at the service of a problem in another. It is not a stored scalar number; it is a pattern of access and integration. This is the kind of functional architecture that Neuraxon tries to emulate — modular subsystems with continuous-time dynamics and multi-timescale plasticity, rather than a monolithic transformer.
François Chollet and the Modern Approach: Measuring What You Still Don't Know How to Do
Against the psychometric legacy, François Chollet proposed in 2019 a conceptual turn. His argument, in On the Measure of Intelligence, is that we were measuring the wrong thing.
Traditional AI benchmarks reward skills, specific competencies on concrete tasks. But a skill can be bought with data and computation: it is enough to train sufficiently on a task to master it. Intelligence, Chollet maintains, is not skill, but efficiency in the acquisition of skills: how much you learn from how little, when facing a genuinely new task (Chollet, 2019).
Intelligence is what you do when you don't know what to do.
This distinction changes everything. A system that solves a million problems because it has seen ten million similar ones is not intelligent. An intelligent system is the one that, facing a problem for which it could not prepare, discovers the structure and adapts with few examples. The measure stops being the final result and becomes the slope of learning.
ARC-AGI: The Benchmark That Tests Genuine AI Reasoning
ARC-AGI was born from that idea, and its most recent version, ARC-AGI-3, takes it further. It is not a question-and-answer test. It is a set of interactive environments, like mini-videogames, in which the agent explores an unknown world, deduces what the objective is without being told in natural language, builds a model of the environment and adapts its strategy step by step (ARC Prize, 2025).
The design principles are explicit: environments 100 % solvable by humans, with no preloaded knowledge or hidden instructions, and with enough novelty to prevent memorization. What is scored is not getting it right, but efficiency in the acquisition of skill over time.
It is the opposite of the g factor: instead of looking for what a system already masters and summarizing it, it looks for what it still does not know how to do and measures how much it costs it to learn it.
Data Contamination: Why LLM Benchmark Scores Are Inflated
The ultimate reason why Chollet's approach matters, and why the g factor applied to LLMs is so slippery, has a technical name: data contamination. If the exam, or something almost identical, was in the notes the student studied, their grade does not measure what they can reason. It measures what they have memorized.
Language models are trained on books, forums, code repositories, articles, practically all the available text. The benchmarks with which we then evaluate them are published on the internet. The conclusion is that fragments of the tests end up inside the training data, which violates the separation between training and evaluation and inflates the scores (Xu et al., 2024; Deng et al., 2024). Empirical audits have detected contamination levels ranging from 1 % up to 45 % in widely used benchmarks, and the problem grows over time (Li et al., 2024).
It is not a minor problem of a couple of leaked questions. In benchmarks as cited as MMLU or GSM8K, part of what we interpret as reasoning may be pure memorization (Chen et al., 2025). When decontamination techniques are applied that rewrite the leaked items without altering their difficulty, accuracy drops: in one study, 22.9 % on GSM8K and 19.0 % on MMLU (Zhu et al., 2024).
Paraphrased items, or even ones translated into another language, dodge the superficial-overlap detectors and continue to inflate the results (Yang et al., 2023; Yao et al., 2024). The usual solutions (paraphrasing, translating, tweaking the context) are assumed to be effective without having been validated rigorously. And for most open models we cannot even check anything, because their training data is not published. We are grading exams without knowing what the student studied.
Here one understands why ARC-AGI chose the path it chose. An interactive, novel environment, with no natural-language instructions and designed to prevent brute-force memorization is, by construction, resistant to contamination.
So, What Should We Measure to Evaluate Machine Intelligence?
The g factor is a populational property that, applied to models that share architecture and corpus, runs the risk of measuring the scale of computation and not generality. The lesson for whoever builds artificial systems is not to choose between the g factor and ARC-AGI as if they were rival teams. It is to understand what question each one answers. A factor analysis can be useful to describe the internal structure of a system's performance, as long as the first factor is not confused with an essence of intelligence. And an ARC-type protocol is indispensable for what really matters: checking whether the system generalizes beyond what it saw, or merely recites.
When we evaluate a system only by its final answer, we are measuring it with our eyes closed to its temporal dimension: planning, the updating of beliefs, the integration of evidence across many steps. It is exactly what ARC-AGI-3 decided to score, and exactly what a static exam cannot see.
Why Brain-Inspired AI Architectures Like Neuraxon Take a Different Path
If intelligence is not a stored number but the efficient integration of specialized subsystems, as suggested by the parieto-frontal integration theory (P-FIT) and the global availability of the workspace in the brain…
If that integration is above all a temporal phenomenon, with time scales…
Then a system built on modular architectures with functional spheres, plasticity across multiple temporal scales and continuous dynamics does not need to be evaluated by asking it to recite answers.
The correct question is not how many benchmarks it beats, but with what efficiency it acquires new behavior, over time, in environments for which it was not prepared. That is the direction Neuraxon tries to take. To compute time – that is, adaptation – not memorized answers that simulate being a good student, when in reality, it already knows the questions.
#AI #AGI #Qubic #TechTrends
References
Chollet, F. (2019). On the Measure of Intelligence. arXiv:1911.01547.Deary, I. J., Penke, L., & Johnson, W. (2009). The neuroscience of human intelligence differences. Nature Reviews Neuroscience.Dehaene, S., & Changeux, J.-P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70(2), 200–227.Detterman, D. K., & Daniel, M. H. (1989). Correlations of mental tests with each other and with cognitive variables. Intelligence.Gignac, G. E., & Szodorai, E. T. (2024). Defining and identifying a general factor of ability in AI systems.Guttman, L. (1955). The determinacy of factor score matrices with implications for five other basic problems of common-factor theory. British Journal of Statistical Psychology.Hernández-Orallo, J., et al. (2021). General intelligence disentangled via a generality metric for natural and artificial intelligence. Scientific Reports.Honey, C. J., et al. (2012). Slow cortical dynamics and the accumulation of information over long timescales. Neuron, 76(2), 423–434.Ilić, D. (2023). Unveiling the General Intelligence Factor in Language Models: A Psychometric Approach. arXiv:2310.11616.Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence. Behavioral and Brain Sciences.Spearman, C. (1904). "General intelligence" objectively determined and measured. American Journal of Psychology, 15, 201–293.Roberts, M., et al. (2024). Temporal evidence of contamination from training cutoff dates.Schönemann, P. H. (2008). A Rejoinder to Mackintosh and some Remarks on the Concept of General Intelligence. arXiv:0808.2343.Xu, C., et al. (2024). Benchmark data contamination of large language models: a survey.Yang, S., et al. (2023). Rethinking benchmark and contamination for language models with rephrased samples.Zhu, Q., et al. (2024). Inference-Time Decontamination: Reusing leaked benchmarks for LLM evaluation. Findings of EMNLP 2024.ARC Prize (2025). ARC-AGI-3: An interactive reasoning benchmark. Technical Report.
Explore the Full Neuraxon Intelligence Academy Series
This is Volume 10 of the Neuraxon Intelligence Academy by the Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, decentralized artificial intelligence:
[NIA Volume 1](https://www.binance.com/en/square/post/295315343732018): Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.[NIA Volume 2](https://www.binance.com/en/square/post/295304276561778): Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.[NIA Volume 3](https://www.binance.com/en/square/post/295306656801506): Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.[NIA Volume 4](https://www.binance.com/en/square/post/295302152913618): Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.[NIA Volume 5](https://www.binance.com/en/square/post/302913958960674): Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.[NIA Volume 6](https://www.binance.com/en/square/post/310198879866145): Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.[NIA Volume 7](https://www.binance.com/en/square/post/321350661453970): Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's emergent complexity and self-organized criticality.[NIA Volume 8](https://www.binance.com/en/square/post/322900066069841): Brain Criticality and the Branching Ratio in Neural and Artificial Networks — Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.[NIA Volume 9](https://www.binance.com/en/square/post/328379422341521): The Origins of the g Factor: From Education and Neuroscience to Artificial Intelligence — Explores the origins of the g factor across education, neuroscience, and AI.
$Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
Measuring Machine Intelligence: The g Factor vs. ARC-AGI Benchmark
#Neuraxon Intelligence Academy — Volume 10
By the Qubic Scientific Team
If we build an artificial system and want to know whether it is intelligent, what exactly do we measure? We think we know when we hear that ChatGPT-5 announces it has beaten DeepSeek and then that Claude sweeps Gemini.
But the question is still there, intact. Measuring artificial intelligence is not measuring speed or temperature. We have no unit of measurement, as strange as that may seem.
In psychology we have been dealing with this problem for over a century. Artificial intelligence has been at it for a decade. And it does so in a hurry, with a lot of money at stake and with a constant temptation: to declare victory.
The g Factor: A Single Number to Summarize General Intelligence
At the beginning of the 20th century, Charles Spearman realized that when a child performed well in one subject, they tended to perform well in the others, even if they were subjects with no apparent relation. The scores correlated with one another, all of them positively. He called that pattern the positive manifold, and he deduced that there must be a common latent factor behind all those disparate abilities: the factor g, or general intelligence (Spearman, 1904).
The idea is seductive. If all cognitive tests load onto a single factor, it is enough to extract that factor through factor analysis to have a summary measure of general capacity. In human practice, that first factor usually explains between 40 and 50 % of the variance in performance (Detterman & Daniel, 1989; Deary et al., 2009).
But watch out, because here lies the first trap. The g factor is populational. It does not measure the individual, but variance within individuals (Hernández-Orallo et al., 2021). To say that a specific subject has so much g is, strictly speaking, a mistake. g emerges when comparing many subjects, not when examining one. Like personality, you are the most extroverted of your age group. And you remain so at 50 relative to your group, even if in intensity you are less extroverted than at 20.
What Does IQ Really Measure? Understanding Intelligence Scores
But then, what does IQ measure?
It measures a relative position. The scale is calibrated on a sample with mean 100, standard deviation 15. An IQ of 130 is not an absolute amount of intelligence stored inside someone's head; it is the assertion that this person is two standard deviations above the mean of their normative group. The number is attached to the individual, yes, but its meaning is populational. It is a position in a ranking, not a content.
Your height is absolute: you are 180 centimeters tall even if you are the last human being on Earth. Your IQ is not: being above the mean requires a mean, and a mean requires others. No one can be more intelligent than the average on a desert island.
Now one understands why transferring this to AI is so delicate. When someone computes a g for a set of large language models (LLMs), that factor is an artifact of the set they chose. We are measuring a position in a table, and we present it as if it were an internal property of the system.
Applying the g Factor to Artificial Intelligence: A Dangerous Temptation
The temptation to transfer all of this to AI was irresistible. Gignac and Szodorai proposed that, if the performance of models across varied tasks correlates positively, it should be possible to identify a general factor of capacity in artificial systems as well. And indeed, several recent works apply factor analysis to test batteries in LLMs and find a unidimensional g factor that remains stable across models, batteries and extraction methods (Ilić, 2023). It sounds like confirmation. It is wise to be suspicious.
The appearance of a dominant first factor does not prove that there exists a general capacity analogous to the human one. It proves that the scores of those models covary. And they covary for a very shallow reason: they share architecture, they share training corpus, they share optimization recipes. A large, well-trained model does everything better than a small, poorly trained one, across all tasks at once. That is enough to manufacture a beautiful positive manifold that tells us nothing about cognitive generality. It tells us about the scale of computation. WATCH OUT: The factor we extract may simply be a factor of size disguised as intelligence.
The brain, moreover, does not concentrate intelligence in a single module. A multitude of specialized subsystems process in parallel and, when a piece of information wins the competition, it becomes globally available to the rest of the system, which can then recombine it for new purposes (Baars, 1988; Dehaene & Changeux, 2011). What we call generality is global availability: putting a piece learned in one context at the service of a problem in another. It is not a stored scalar number; it is a pattern of access and integration. This is the kind of functional architecture that Neuraxon tries to emulate — modular subsystems with continuous-time dynamics and multi-timescale plasticity, rather than a monolithic transformer.
François Chollet and the Modern Approach: Measuring What You Still Don't Know How to Do
Against the psychometric legacy, François Chollet proposed in 2019 a conceptual turn. His argument, in On the Measure of Intelligence, is that we were measuring the wrong thing.
Traditional AI benchmarks reward skills, specific competencies on concrete tasks. But a skill can be bought with data and computation: it is enough to train sufficiently on a task to master it. Intelligence, Chollet maintains, is not skill, but efficiency in the acquisition of skills: how much you learn from how little, when facing a genuinely new task (Chollet, 2019).
Intelligence is what you do when you don't know what to do.
This distinction changes everything. A system that solves a million problems because it has seen ten million similar ones is not intelligent. An intelligent system is the one that, facing a problem for which it could not prepare, discovers the structure and adapts with few examples. The measure stops being the final result and becomes the slope of learning.
ARC-AGI: The Benchmark That Tests Genuine AI Reasoning
ARC-AGI was born from that idea, and its most recent version, ARC-AGI-3, takes it further. It is not a question-and-answer test. It is a set of interactive environments, like mini-videogames, in which the agent explores an unknown world, deduces what the objective is without being told in natural language, builds a model of the environment and adapts its strategy step by step (ARC Prize, 2025).
The design principles are explicit: environments 100 % solvable by humans, with no preloaded knowledge or hidden instructions, and with enough novelty to prevent memorization. What is scored is not getting it right, but efficiency in the acquisition of skill over time.
It is the opposite of the g factor: instead of looking for what a system already masters and summarizing it, it looks for what it still does not know how to do and measures how much it costs it to learn it.
Data Contamination: Why LLM Benchmark Scores Are Inflated
The ultimate reason why Chollet's approach matters, and why the g factor applied to LLMs is so slippery, has a technical name: data contamination. If the exam, or something almost identical, was in the notes the student studied, their grade does not measure what they can reason. It measures what they have memorized.
Language models are trained on books, forums, code repositories, articles, practically all the available text. The benchmarks with which we then evaluate them are published on the internet. The conclusion is that fragments of the tests end up inside the training data, which violates the separation between training and evaluation and inflates the scores (Xu et al., 2024; Deng et al., 2024). Empirical audits have detected contamination levels ranging from 1 % up to 45 % in widely used benchmarks, and the problem grows over time (Li et al., 2024).
It is not a minor problem of a couple of leaked questions. In benchmarks as cited as MMLU or GSM8K, part of what we interpret as reasoning may be pure memorization (Chen et al., 2025). When decontamination techniques are applied that rewrite the leaked items without altering their difficulty, accuracy drops: in one study, 22.9 % on GSM8K and 19.0 % on MMLU (Zhu et al., 2024).
Paraphrased items, or even ones translated into another language, dodge the superficial-overlap detectors and continue to inflate the results (Yang et al., 2023; Yao et al., 2024). The usual solutions (paraphrasing, translating, tweaking the context) are assumed to be effective without having been validated rigorously. And for most open models we cannot even check anything, because their training data is not published. We are grading exams without knowing what the student studied.
Here one understands why ARC-AGI chose the path it chose. An interactive, novel environment, with no natural-language instructions and designed to prevent brute-force memorization is, by construction, resistant to contamination.
So, What Should We Measure to Evaluate Machine Intelligence?
The g factor is a populational property that, applied to models that share architecture and corpus, runs the risk of measuring the scale of computation and not generality. The lesson for whoever builds artificial systems is not to choose between the g factor and ARC-AGI as if they were rival teams. It is to understand what question each one answers. A factor analysis can be useful to describe the internal structure of a system's performance, as long as the first factor is not confused with an essence of intelligence. And an ARC-type protocol is indispensable for what really matters: checking whether the system generalizes beyond what it saw, or merely recites.
When we evaluate a system only by its final answer, we are measuring it with our eyes closed to its temporal dimension: planning, the updating of beliefs, the integration of evidence across many steps. It is exactly what ARC-AGI-3 decided to score, and exactly what a static exam cannot see.
Why Brain-Inspired AI Architectures Like Neuraxon Take a Different Path
If intelligence is not a stored number but the efficient integration of specialized subsystems, as suggested by the parieto-frontal integration theory (P-FIT) and the global availability of the workspace in the brain…
If that integration is above all a temporal phenomenon, with time scales…
Then a system built on modular architectures with functional spheres, plasticity across multiple temporal scales and continuous dynamics does not need to be evaluated by asking it to recite answers.
The correct question is not how many benchmarks it beats, but with what efficiency it acquires new behavior, over time, in environments for which it was not prepared. That is the direction Neuraxon tries to take. To compute time – that is, adaptation – not memorized answers that simulate being a good student, when in reality, it already knows the questions.
#AI #AGI #Qubic #TechTrends
References
Chollet, F. (2019). On the Measure of Intelligence. arXiv:1911.01547.Deary, I. J., Penke, L., & Johnson, W. (2009). The neuroscience of human intelligence differences. Nature Reviews Neuroscience.Dehaene, S., & Changeux, J.-P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70(2), 200–227.Detterman, D. K., & Daniel, M. H. (1989). Correlations of mental tests with each other and with cognitive variables. Intelligence.Gignac, G. E., & Szodorai, E. T. (2024). Defining and identifying a general factor of ability in AI systems.Guttman, L. (1955). The determinacy of factor score matrices with implications for five other basic problems of common-factor theory. British Journal of Statistical Psychology.Hernández-Orallo, J., et al. (2021). General intelligence disentangled via a generality metric for natural and artificial intelligence. Scientific Reports.Honey, C. J., et al. (2012). Slow cortical dynamics and the accumulation of information over long timescales. Neuron, 76(2), 423–434.Ilić, D. (2023). Unveiling the General Intelligence Factor in Language Models: A Psychometric Approach. arXiv:2310.11616.Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence. Behavioral and Brain Sciences.Spearman, C. (1904). "General intelligence" objectively determined and measured. American Journal of Psychology, 15, 201–293.Roberts, M., et al. (2024). Temporal evidence of contamination from training cutoff dates.Schönemann, P. H. (2008). A Rejoinder to Mackintosh and some Remarks on the Concept of General Intelligence. arXiv:0808.2343.Xu, C., et al. (2024). Benchmark data contamination of large language models: a survey.Yang, S., et al. (2023). Rethinking benchmark and contamination for language models with rephrased samples.Zhu, Q., et al. (2024). Inference-Time Decontamination: Reusing leaked benchmarks for LLM evaluation. Findings of EMNLP 2024.ARC Prize (2025). ARC-AGI-3: An interactive reasoning benchmark. Technical Report.
Explore the Full Neuraxon Intelligence Academy Series
This is Volume 10 of the Neuraxon Intelligence Academy by the Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Volume 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Volume 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's emergent complexity and self-organized criticality.NIA Volume 8: Brain Criticality and the Branching Ratio in Neural and Artificial Networks — Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.NIA Volume 9: The Origins of the g Factor: From Education and Neuroscience to Artificial Intelligence — Explores the origins of the g factor across education, neuroscience, and AI.
$Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
By the Qubic Scientific Team
If we build an artificial system and want to know whether it is intelligent, what exactly do we measure? We think we know when we hear that ChatGPT-5 announces it has beaten DeepSeek and then that Claude sweeps Gemini.
But the question is still there, intact. Measuring artificial intelligence is not measuring speed or temperature. We have no unit of measurement, as strange as that may seem.
In psychology we have been dealing with this problem for over a century. Artificial intelligence has been at it for a decade. And it does so in a hurry, with a lot of money at stake and with a constant temptation: to declare victory.
The g Factor: A Single Number to Summarize General Intelligence
At the beginning of the 20th century, Charles Spearman realized that when a child performed well in one subject, they tended to perform well in the others, even if they were subjects with no apparent relation. The scores correlated with one another, all of them positively. He called that pattern the positive manifold, and he deduced that there must be a common latent factor behind all those disparate abilities: the factor g, or general intelligence (Spearman, 1904).
The idea is seductive. If all cognitive tests load onto a single factor, it is enough to extract that factor through factor analysis to have a summary measure of general capacity. In human practice, that first factor usually explains between 40 and 50 % of the variance in performance (Detterman & Daniel, 1989; Deary et al., 2009).
But watch out, because here lies the first trap. The g factor is populational. It does not measure the individual, but variance within individuals (Hernández-Orallo et al., 2021). To say that a specific subject has so much g is, strictly speaking, a mistake. g emerges when comparing many subjects, not when examining one. Like personality, you are the most extroverted of your age group. And you remain so at 50 relative to your group, even if in intensity you are less extroverted than at 20.
What Does IQ Really Measure? Understanding Intelligence Scores
But then, what does IQ measure?
It measures a relative position. The scale is calibrated on a sample with mean 100, standard deviation 15. An IQ of 130 is not an absolute amount of intelligence stored inside someone's head; it is the assertion that this person is two standard deviations above the mean of their normative group. The number is attached to the individual, yes, but its meaning is populational. It is a position in a ranking, not a content.
Your height is absolute: you are 180 centimeters tall even if you are the last human being on Earth. Your IQ is not: being above the mean requires a mean, and a mean requires others. No one can be more intelligent than the average on a desert island.
Now one understands why transferring this to AI is so delicate. When someone computes a g for a set of large language models (LLMs), that factor is an artifact of the set they chose. We are measuring a position in a table, and we present it as if it were an internal property of the system.
Applying the g Factor to Artificial Intelligence: A Dangerous Temptation
The temptation to transfer all of this to AI was irresistible. Gignac and Szodorai proposed that, if the performance of models across varied tasks correlates positively, it should be possible to identify a general factor of capacity in artificial systems as well. And indeed, several recent works apply factor analysis to test batteries in LLMs and find a unidimensional g factor that remains stable across models, batteries and extraction methods (Ilić, 2023). It sounds like confirmation. It is wise to be suspicious.
The appearance of a dominant first factor does not prove that there exists a general capacity analogous to the human one. It proves that the scores of those models covary. And they covary for a very shallow reason: they share architecture, they share training corpus, they share optimization recipes. A large, well-trained model does everything better than a small, poorly trained one, across all tasks at once. That is enough to manufacture a beautiful positive manifold that tells us nothing about cognitive generality. It tells us about the scale of computation. WATCH OUT: The factor we extract may simply be a factor of size disguised as intelligence.
The brain, moreover, does not concentrate intelligence in a single module. A multitude of specialized subsystems process in parallel and, when a piece of information wins the competition, it becomes globally available to the rest of the system, which can then recombine it for new purposes (Baars, 1988; Dehaene & Changeux, 2011). What we call generality is global availability: putting a piece learned in one context at the service of a problem in another. It is not a stored scalar number; it is a pattern of access and integration. This is the kind of functional architecture that Neuraxon tries to emulate — modular subsystems with continuous-time dynamics and multi-timescale plasticity, rather than a monolithic transformer.
François Chollet and the Modern Approach: Measuring What You Still Don't Know How to Do
Against the psychometric legacy, François Chollet proposed in 2019 a conceptual turn. His argument, in On the Measure of Intelligence, is that we were measuring the wrong thing.
Traditional AI benchmarks reward skills, specific competencies on concrete tasks. But a skill can be bought with data and computation: it is enough to train sufficiently on a task to master it. Intelligence, Chollet maintains, is not skill, but efficiency in the acquisition of skills: how much you learn from how little, when facing a genuinely new task (Chollet, 2019).
Intelligence is what you do when you don't know what to do.
This distinction changes everything. A system that solves a million problems because it has seen ten million similar ones is not intelligent. An intelligent system is the one that, facing a problem for which it could not prepare, discovers the structure and adapts with few examples. The measure stops being the final result and becomes the slope of learning.
ARC-AGI: The Benchmark That Tests Genuine AI Reasoning
ARC-AGI was born from that idea, and its most recent version, ARC-AGI-3, takes it further. It is not a question-and-answer test. It is a set of interactive environments, like mini-videogames, in which the agent explores an unknown world, deduces what the objective is without being told in natural language, builds a model of the environment and adapts its strategy step by step (ARC Prize, 2025).
The design principles are explicit: environments 100 % solvable by humans, with no preloaded knowledge or hidden instructions, and with enough novelty to prevent memorization. What is scored is not getting it right, but efficiency in the acquisition of skill over time.
It is the opposite of the g factor: instead of looking for what a system already masters and summarizing it, it looks for what it still does not know how to do and measures how much it costs it to learn it.
Data Contamination: Why LLM Benchmark Scores Are Inflated
The ultimate reason why Chollet's approach matters, and why the g factor applied to LLMs is so slippery, has a technical name: data contamination. If the exam, or something almost identical, was in the notes the student studied, their grade does not measure what they can reason. It measures what they have memorized.
Language models are trained on books, forums, code repositories, articles, practically all the available text. The benchmarks with which we then evaluate them are published on the internet. The conclusion is that fragments of the tests end up inside the training data, which violates the separation between training and evaluation and inflates the scores (Xu et al., 2024; Deng et al., 2024). Empirical audits have detected contamination levels ranging from 1 % up to 45 % in widely used benchmarks, and the problem grows over time (Li et al., 2024).
It is not a minor problem of a couple of leaked questions. In benchmarks as cited as MMLU or GSM8K, part of what we interpret as reasoning may be pure memorization (Chen et al., 2025). When decontamination techniques are applied that rewrite the leaked items without altering their difficulty, accuracy drops: in one study, 22.9 % on GSM8K and 19.0 % on MMLU (Zhu et al., 2024).
Paraphrased items, or even ones translated into another language, dodge the superficial-overlap detectors and continue to inflate the results (Yang et al., 2023; Yao et al., 2024). The usual solutions (paraphrasing, translating, tweaking the context) are assumed to be effective without having been validated rigorously. And for most open models we cannot even check anything, because their training data is not published. We are grading exams without knowing what the student studied.
Here one understands why ARC-AGI chose the path it chose. An interactive, novel environment, with no natural-language instructions and designed to prevent brute-force memorization is, by construction, resistant to contamination.
So, What Should We Measure to Evaluate Machine Intelligence?
The g factor is a populational property that, applied to models that share architecture and corpus, runs the risk of measuring the scale of computation and not generality. The lesson for whoever builds artificial systems is not to choose between the g factor and ARC-AGI as if they were rival teams. It is to understand what question each one answers. A factor analysis can be useful to describe the internal structure of a system's performance, as long as the first factor is not confused with an essence of intelligence. And an ARC-type protocol is indispensable for what really matters: checking whether the system generalizes beyond what it saw, or merely recites.
When we evaluate a system only by its final answer, we are measuring it with our eyes closed to its temporal dimension: planning, the updating of beliefs, the integration of evidence across many steps. It is exactly what ARC-AGI-3 decided to score, and exactly what a static exam cannot see.
Why Brain-Inspired AI Architectures Like Neuraxon Take a Different Path
If intelligence is not a stored number but the efficient integration of specialized subsystems, as suggested by the parieto-frontal integration theory (P-FIT) and the global availability of the workspace in the brain…
If that integration is above all a temporal phenomenon, with time scales…
Then a system built on modular architectures with functional spheres, plasticity across multiple temporal scales and continuous dynamics does not need to be evaluated by asking it to recite answers.
The correct question is not how many benchmarks it beats, but with what efficiency it acquires new behavior, over time, in environments for which it was not prepared. That is the direction Neuraxon tries to take. To compute time – that is, adaptation – not memorized answers that simulate being a good student, when in reality, it already knows the questions.
#AI #AGI #Qubic #TechTrends
References
Chollet, F. (2019). On the Measure of Intelligence. arXiv:1911.01547.Deary, I. J., Penke, L., & Johnson, W. (2009). The neuroscience of human intelligence differences. Nature Reviews Neuroscience.Dehaene, S., & Changeux, J.-P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70(2), 200–227.Detterman, D. K., & Daniel, M. H. (1989). Correlations of mental tests with each other and with cognitive variables. Intelligence.Gignac, G. E., & Szodorai, E. T. (2024). Defining and identifying a general factor of ability in AI systems.Guttman, L. (1955). The determinacy of factor score matrices with implications for five other basic problems of common-factor theory. British Journal of Statistical Psychology.Hernández-Orallo, J., et al. (2021). General intelligence disentangled via a generality metric for natural and artificial intelligence. Scientific Reports.Honey, C. J., et al. (2012). Slow cortical dynamics and the accumulation of information over long timescales. Neuron, 76(2), 423–434.Ilić, D. (2023). Unveiling the General Intelligence Factor in Language Models: A Psychometric Approach. arXiv:2310.11616.Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence. Behavioral and Brain Sciences.Spearman, C. (1904). "General intelligence" objectively determined and measured. American Journal of Psychology, 15, 201–293.Roberts, M., et al. (2024). Temporal evidence of contamination from training cutoff dates.Schönemann, P. H. (2008). A Rejoinder to Mackintosh and some Remarks on the Concept of General Intelligence. arXiv:0808.2343.Xu, C., et al. (2024). Benchmark data contamination of large language models: a survey.Yang, S., et al. (2023). Rethinking benchmark and contamination for language models with rephrased samples.Zhu, Q., et al. (2024). Inference-Time Decontamination: Reusing leaked benchmarks for LLM evaluation. Findings of EMNLP 2024.ARC Prize (2025). ARC-AGI-3: An interactive reasoning benchmark. Technical Report.
Explore the Full Neuraxon Intelligence Academy Series
This is Volume 10 of the Neuraxon Intelligence Academy by the Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Volume 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Volume 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's emergent complexity and self-organized criticality.NIA Volume 8: Brain Criticality and the Branching Ratio in Neural and Artificial Networks — Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.NIA Volume 9: The Origins of the g Factor: From Education and Neuroscience to Artificial Intelligence — Explores the origins of the g factor across education, neuroscience, and AI.
$Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
149
·
--
We have no unit of measurement for intelligence.
Not for humans. Not for machines.
We've been arguing about it for over a century.
Up to 45% of the benchmarks we use to evaluate LLMs contain leaked training data.
ARC-AGI-3 was built to fix that.
Humans solve 100% of it.
Frontier AI scores below 1%.
NIA Volume 10 breaks down the g factor, Chollet's framework, benchmark contamination, and what measuring machine intelligence actually requires.
Full read
👇
[Measuring Machine Intelligence: The g Factor vs. ARC-AGI Benchmark](https://www.binance.com/en/square/post/332806106415490)
@BiBi
#AI #AGI #Qubic #TechTrends #Neuraxon
We have no unit of measurement for intelligence.
Not for humans. Not for machines.
We've been arguing about it for over a century.
Up to 45% of the benchmarks we use to evaluate LLMs contain leaked training data.
ARC-AGI-3 was built to fix that.
Humans solve 100% of it.
Frontier AI scores below 1%.
NIA Volume 10 breaks down the g factor, Chollet's framework, benchmark contamination, and what measuring machine intelligence actually requires.
Full read
👇
Measuring Machine Intelligence: The g Factor vs. ARC-AGI Benchmark
@Binance BiBi
#AI #AGI #Qubic #TechTrends #Neuraxon
Not for humans. Not for machines.
We've been arguing about it for over a century.
Up to 45% of the benchmarks we use to evaluate LLMs contain leaked training data.
ARC-AGI-3 was built to fix that.
Humans solve 100% of it.
Frontier AI scores below 1%.
NIA Volume 10 breaks down the g factor, Chollet's framework, benchmark contamination, and what measuring machine intelligence actually requires.
Full read
👇
Measuring Machine Intelligence: The g Factor vs. ARC-AGI Benchmark
@Binance BiBi
#AI #AGI #Qubic #TechTrends #Neuraxon
142
·
--
Článok
The g Factor in Artificial Life: From Spearman's 1904 Classroom to Evolved Artificial BrainsNeuraxon Intelligence Academy, Volume 9 · By the Qubic Scientific Team
In one line: General intelligence, the g factor psychologists have measured for over a century, is the missing ingredient in today's language models, and Qubic's Neuraxon project is now selecting for it directly inside an artificial-life simulation.
The g Factor: From a 1904 Classroom to Artificial Brains
In 1904, Charles Spearman stumbled upon a regularity that would forever change psychology. Examining the school grades of a group of English children, he noticed something seemingly trivial but strange: those who excelled in mathematics also tended to excel in French, in music, in language. Disciplines with no apparent connection correlated systematically with one another. Spearman proposed that beneath this tangle of disparate abilities there lay a single common factor, a general cognitive thread. He called it g (Spearman, 1904).
More than a century later, g remains one of the most replicated findings in the behavioral sciences (Carroll, 1993; Deary et al., 2010). It is neither a grade average nor an arbitrary construct: it is what emerges when factor analysis is applied to almost any battery of cognitive tests. It appears consistently when we measure working memory, fluid reasoning, processing speed, verbal comprehension, or novel problem solving. In psychometric terms, g is the shared variance that no single test measures on its own.
What the g Factor Means in the Brain and in Behavior
P-FIT Theory and Brain Network Efficiency
From cognitive neuroscience, g has ceased to be a statistical abstraction and has become a property of brain architecture. The P-FIT theory (Parieto-Frontal Integration Theory) identifies a distributed network made up of dorsolateral prefrontal cortex, posterior parietal cortex, anterior cingulate, and temporal areas, whose connection efficiency predicts intelligence test scores (Jung & Haier, 2007). Functional connectivity studies show that g correlates with the brain's ability to dynamically reconfigure its networks (the executive control network, the default mode network, the salience network) according to task demands (Barbey, 2018; Cole et al., 2015). It is not about having "more" neurons in a specific place, but about better orchestrating the flow of information between functionally specialized regions.
The Predictive Brain and Free-Energy Minimization
This orchestration acquires an even deeper meaning in light of the predictive brain theory (Clark, 2013; Friston, 2010). Under this framework, the brain is not a passive receiver of stimuli but a hierarchical inference engine that continuously generates predictions about the world and adjusts its internal models based on prediction error. Here g fits naturally: the ability to predict well, to anticipate environmental contingencies, to learn quickly from error and, above all, to abstract regularities that transfer across domains, is precisely what intelligence tests capture indirectly. A brain with high g would be, on this reading, a system with more efficient generative models, capable of compressing experience into high-level abstractions and of minimizing free energy across heterogeneous contexts (Hohwy, 2013); that is, it reduces prediction error rapidly and therefore learns. Cognitive generality, then, would not be a static property of the neural hardware, but the quality of a deeply hierarchical predictive process. The research remains open. Other currents posit that g really has to do with the neurodevelopment of our brain, given that no matter what task we are performing or attempting, there is a huge common factor in any experience because it happens inside the same organ.
Behaviorally, g is the best predictor. Forget emotional intelligence; it is g that best forecasts what your academic performance, occupational success, longevity, and even certain health indicators may be (Deary et al., 2010; Gottfredson, 1997). Not because it is destiny, but because it captures something very basic: the capacity of a cognitive system to face problems it has not seen before, integrating heterogeneous information under time and resource constraints. g is, in a sense, a measure of generality.
The Problem of Measuring General Intelligence in Artificial Systems
For decades, artificial systems have shone in narrow tasks (playing chess, classifying images, translating) but failed to transfer that performance outside their domain (Chollet, 2019). The #AGI debate revolves precisely around this: what does it mean, operationally, for a system to be "generally" intelligent?
If we take the parallel with human psychometrics seriously, the answer is uncomfortable but clear: to speak of generality we need to measure it, and measuring it requires diverse tests whose shared variance reveals something analogous to g. A system with high performance on a single task tells us nothing about its generality; a system with moderate and correlated performance across many structurally distinct tasks does. Spearman's logic, transferred to non-biological substrates, still holds: generality is not postulated, it is factored.
Why the g Factor Does Not Appear in Transformers (and What That Implies for AGI)
It is worth pausing here on the currently dominant paradigm. Large language models based on transformer architectures (Vaswani et al., 2017) deliver astonishing performance on linguistic tasks, but psychometric analyses applied to their outputs do not show the factor structure characteristic of g (Burnell et al., 2023; Ilić & Gignac, 2024). Their hits and misses across domains do not correlate as they would in humans; they depend rather on the density and quality of patterns present in their training data. A transformer can brilliantly solve one problem and fail on another that is structurally equivalent but phrased slightly differently, something a system with genuine g would not do (Mitchell, 2021).
This has serious implications. It suggests that the pursuit of cognitive generality exclusively through language may be a dead end, an architectural dead end. Language is the most visible output of human cognition, but not its substrate. To pretend that by scaling text one will arrive at g is like pretending that by scaling descriptions of chess games one will arrive at mastery: one obtains statistical mimicry, not the underlying cognitive structure. (We argued a closely related point in our analysis of why intelligence is not scale, and on why LLM predictions are not brain predictions.) Without genuine hierarchical prediction, without generative models of the world, without coordination between functionally specialized modules, behavior can look general without being so. The absence of g in transformers is not a failure of scale: it is a clue that generality requires other architectural ingredients (LeCun, 2022).
The g Factor Inside the Neuraxon Game of Life
We have taken this intuition to a different experimental terrain. In Multi-Neuraxon Game of Life Lite 5.0, the artificial creatures (the Nxons) grow their own brains and compete to survive. What is new in this version is that the selective pressure is applied to g. The Nxons are not selected for mastering a specific task, but for showing that common thread that allows them to face many.
The brains of the Nxons have been designed following a simplified model anchored in cognitive neuroscience, since they use six functional regions, inspired by the same kind of maps that psychologists use to describe the modular organization of the human brain. The bet is that generality does not emerge from a monolithic architecture, but from the coordination among specialized regions that share information flexibly. It is the P-FIT intuition translated into artificial life, and it connects directly with the predictive brain principle: each region contributes its own model, and the integration between them is what allows hierarchical prediction and, therefore, generality. (These dynamics build directly on the brain-criticality and branching-ratio principles we explored in [Volume 8](https://www.binance.com/en/square/post/322900066069841).)
Notably, the experiment is public and observable. Anyone can open their browser and watch how the Nxons evolve generation after generation, how their internal circuits reorganize under the pressure of a fitness function that rewards cognitive generality instead of specialization.
Implications for Artificial Life (Alife) and Applications for Qubic
For the field of artificial life, the explicit incorporation of g as a selection criterion opens a line of work that goes beyond academic exercise. Most Alife systems have evolved agents that solve very concrete niches: foraging, predator avoidance, navigation (Bedau, 2003; Lehman et al., 2020). But few have tried to select for something as abstract as the ability to generalize across heterogeneous cognitive domains. If we manage to get artificial organisms to show positive correlations between distinct tasks (the computational equivalent of Spearman's children) we will have an extraordinary test bench for questions that human psychometrics can only address correlationally: what evolutionary pressures favor the emergence of g? What neural architectures make it possible? Is g a convergent solution or a phylogenetic accident?
For Qubic, this line of research fits with a very concrete vision of the future of #AI . While the industry invests massive resources in scaling transformers over text, Qubic is committed to exploring architecturally alternative paths: modular artificial brains, evolved, distributed, and subjected to real selective pressures. Qubic's decentralized useful-compute network offers the ideal substrate for this kind of experimentation at scale, where thousands of Nxon populations can coevolve in parallel, with fitness functions designed to favor the emergence of g. It is not only open research: it is the possibility of building, on decentralized infrastructure, an empirical alternative to the dominant paradigm of language-based AI, one that starts from the right question (how to measure and select generality) instead of assuming it. If genuine cognitive generality requires architectures inspired by brains and not by corpora, Qubic is one of the few environments where that hypothesis can be seriously put to the test.
A deeper analysis is in preparation, as it forms part of our recent papers and experiments. Spearman's old g, that thread which wove together children's school grades, we now use in digital creatures that learn to survive.
References
Barbey, A. K. (2018). Network neuroscience theory of human intelligence. Trends in Cognitive Sciences, 22(1), 8–20. https://doi.org/10.1016/j.tics.2017.10.001Bedau, M. A. (2003). Artificial life: Organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. https://doi.org/10.1016/j.tics.2003.09.012Burnell, R., Schellaert, W., Burden, J., Ullman, T. D., Martínez-Plumed, F., Tenenbaum, J. B., et al. (2023). Rethink reporting of evaluation results in AI. Science, 380(6641), 136–138. https://doi.org/10.1126/science.adf6369Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. Cambridge University Press. https://doi.org/10.1017/CBO9780511571312Chollet, F. (2019). On the measure of intelligence. arXiv preprint arXiv:1911.01547. https://arxiv.org/abs/1911.01547Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204. https://doi.org/10.1017/S0140525X12000477Cole, M. W., Ito, T., & Braver, T. S. (2015). Lateral prefrontal cortex contributes to fluid intelligence through multinetwork connectivity. Brain Connectivity, 5(8), 497–504. https://doi.org/10.1089/brain.2015.0357Deary, I. J., Penke, L., & Johnson, W. (2010). The neuroscience of human intelligence differences. Nature Reviews Neuroscience, 11(3), 201–211. https://doi.org/10.1038/nrn2793Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138. https://doi.org/10.1038/nrn2787Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24(1), 79–132. https://doi.org/10.1016/S0160-2896(97)90014-3Hohwy, J. (2013). The predictive mind. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199682737.001.0001Ilić, D., & Gignac, G. E. (2024). Evidence of interrelated cognitive-like capabilities in large language models: Indications of artificial general intelligence or achievement? Intelligence, 106, 101858. https://doi.org/10.1016/j.intell.2024.101858Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence: Converging neuroimaging evidence. Behavioral and Brain Sciences, 30(2), 135–154. https://doi.org/10.1017/S0140525X07001185LeCun, Y. (2022). A path towards autonomous machine intelligence. OpenReview, version 0.9.2. https://openreview.net/forum?id=BZ5a1r-kVsfLehman, J., Clune, J., Misevic, D., Adami, C., Altenberg, L., Beaulieu, J., et al. (2020). The surprising creativity of digital evolution. Artificial Life, 26(2), 274–306. https://doi.org/10.1162/artl_a_00319Mitchell, M. (2021). Why AI is harder than we think. arXiv preprint arXiv:2104.12871. https://arxiv.org/abs/2104.12871Spearman, C. (1904). "General intelligence," objectively determined and measured. The American Journal of Psychology, 15(2), 201–292. https://doi.org/10.2307/1412107Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30. https://arxiv.org/abs/1706.03762
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 9 of the #Neuraxon Intelligence Academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
[NIA Volume 1](https://www.binance.com/en/square/post/295315343732018): Why Intelligence Is Not Computed in Steps, but in Time. Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.[NIA Volume 2](https://www.binance.com/en/square/post/295304276561778): Ternary Dynamics as a Model of Living Intelligence. Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.[NIA Volume 3](https://www.binance.com/en/square/post/295306656801506): Neuromodulation and Brain-Inspired AI. Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.[NIA Volume 4](https://www.binance.com/en/square/post/295302152913618): Neural Networks in AI and Neuroscience. A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.[NIA Volume 5](https://www.binance.com/en/square/post/302913958960674): Astrocytes and Brain-Inspired AI. How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.[NIA Volume 6](https://www.binance.com/en/square/post/310198879866145): Conscious Machines vs Intelligent Organisms: AI Consciousness Explained. Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.[NIA Volume 7](https://www.binance.com/en/square/post/321350661453970): Conway's Game of Life, Artificial Life, and Digital Ecosystems. How emergent complexity and self-organized criticality move from simulators to decentralized AI infrastructure.[NIA Volume 8](https://www.binance.com/en/square/post/322900066069841): Brain Criticality and the Branching Ratio in Neural and Artificial Networks. Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.NIA Volume 9: The g Factor in Artificial Life. You are here.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org or browse the full Academy and Blog. Join the discussion on X, Discord, and Telegram.
Qubic is a decentralized, open-source network for experimental technology. Nothing on this site should be construed as investment, legal, or financial advice.
The g Factor in Artificial Life: From Spearman's 1904 Classroom to Evolved Artificial Brains
Neuraxon Intelligence Academy, Volume 9 · By the Qubic Scientific Team
In one line: General intelligence, the g factor psychologists have measured for over a century, is the missing ingredient in today's language models, and Qubic's Neuraxon project is now selecting for it directly inside an artificial-life simulation.
The g Factor: From a 1904 Classroom to Artificial Brains
In 1904, Charles Spearman stumbled upon a regularity that would forever change psychology. Examining the school grades of a group of English children, he noticed something seemingly trivial but strange: those who excelled in mathematics also tended to excel in French, in music, in language. Disciplines with no apparent connection correlated systematically with one another. Spearman proposed that beneath this tangle of disparate abilities there lay a single common factor, a general cognitive thread. He called it g (Spearman, 1904).
More than a century later, g remains one of the most replicated findings in the behavioral sciences (Carroll, 1993; Deary et al., 2010). It is neither a grade average nor an arbitrary construct: it is what emerges when factor analysis is applied to almost any battery of cognitive tests. It appears consistently when we measure working memory, fluid reasoning, processing speed, verbal comprehension, or novel problem solving. In psychometric terms, g is the shared variance that no single test measures on its own.
What the g Factor Means in the Brain and in Behavior
P-FIT Theory and Brain Network Efficiency
From cognitive neuroscience, g has ceased to be a statistical abstraction and has become a property of brain architecture. The P-FIT theory (Parieto-Frontal Integration Theory) identifies a distributed network made up of dorsolateral prefrontal cortex, posterior parietal cortex, anterior cingulate, and temporal areas, whose connection efficiency predicts intelligence test scores (Jung & Haier, 2007). Functional connectivity studies show that g correlates with the brain's ability to dynamically reconfigure its networks (the executive control network, the default mode network, the salience network) according to task demands (Barbey, 2018; Cole et al., 2015). It is not about having "more" neurons in a specific place, but about better orchestrating the flow of information between functionally specialized regions.
The Predictive Brain and Free-Energy Minimization
This orchestration acquires an even deeper meaning in light of the predictive brain theory (Clark, 2013; Friston, 2010). Under this framework, the brain is not a passive receiver of stimuli but a hierarchical inference engine that continuously generates predictions about the world and adjusts its internal models based on prediction error. Here g fits naturally: the ability to predict well, to anticipate environmental contingencies, to learn quickly from error and, above all, to abstract regularities that transfer across domains, is precisely what intelligence tests capture indirectly. A brain with high g would be, on this reading, a system with more efficient generative models, capable of compressing experience into high-level abstractions and of minimizing free energy across heterogeneous contexts (Hohwy, 2013); that is, it reduces prediction error rapidly and therefore learns. Cognitive generality, then, would not be a static property of the neural hardware, but the quality of a deeply hierarchical predictive process. The research remains open. Other currents posit that g really has to do with the neurodevelopment of our brain, given that no matter what task we are performing or attempting, there is a huge common factor in any experience because it happens inside the same organ.
Behaviorally, g is the best predictor. Forget emotional intelligence; it is g that best forecasts what your academic performance, occupational success, longevity, and even certain health indicators may be (Deary et al., 2010; Gottfredson, 1997). Not because it is destiny, but because it captures something very basic: the capacity of a cognitive system to face problems it has not seen before, integrating heterogeneous information under time and resource constraints. g is, in a sense, a measure of generality.
The Problem of Measuring General Intelligence in Artificial Systems
For decades, artificial systems have shone in narrow tasks (playing chess, classifying images, translating) but failed to transfer that performance outside their domain (Chollet, 2019). The #AGI debate revolves precisely around this: what does it mean, operationally, for a system to be "generally" intelligent?
If we take the parallel with human psychometrics seriously, the answer is uncomfortable but clear: to speak of generality we need to measure it, and measuring it requires diverse tests whose shared variance reveals something analogous to g. A system with high performance on a single task tells us nothing about its generality; a system with moderate and correlated performance across many structurally distinct tasks does. Spearman's logic, transferred to non-biological substrates, still holds: generality is not postulated, it is factored.
Why the g Factor Does Not Appear in Transformers (and What That Implies for AGI)
It is worth pausing here on the currently dominant paradigm. Large language models based on transformer architectures (Vaswani et al., 2017) deliver astonishing performance on linguistic tasks, but psychometric analyses applied to their outputs do not show the factor structure characteristic of g (Burnell et al., 2023; Ilić & Gignac, 2024). Their hits and misses across domains do not correlate as they would in humans; they depend rather on the density and quality of patterns present in their training data. A transformer can brilliantly solve one problem and fail on another that is structurally equivalent but phrased slightly differently, something a system with genuine g would not do (Mitchell, 2021).
This has serious implications. It suggests that the pursuit of cognitive generality exclusively through language may be a dead end, an architectural dead end. Language is the most visible output of human cognition, but not its substrate. To pretend that by scaling text one will arrive at g is like pretending that by scaling descriptions of chess games one will arrive at mastery: one obtains statistical mimicry, not the underlying cognitive structure. (We argued a closely related point in our analysis of why intelligence is not scale, and on why LLM predictions are not brain predictions.) Without genuine hierarchical prediction, without generative models of the world, without coordination between functionally specialized modules, behavior can look general without being so. The absence of g in transformers is not a failure of scale: it is a clue that generality requires other architectural ingredients (LeCun, 2022).
The g Factor Inside the Neuraxon Game of Life
We have taken this intuition to a different experimental terrain. In Multi-Neuraxon Game of Life Lite 5.0, the artificial creatures (the Nxons) grow their own brains and compete to survive. What is new in this version is that the selective pressure is applied to g. The Nxons are not selected for mastering a specific task, but for showing that common thread that allows them to face many.
The brains of the Nxons have been designed following a simplified model anchored in cognitive neuroscience, since they use six functional regions, inspired by the same kind of maps that psychologists use to describe the modular organization of the human brain. The bet is that generality does not emerge from a monolithic architecture, but from the coordination among specialized regions that share information flexibly. It is the P-FIT intuition translated into artificial life, and it connects directly with the predictive brain principle: each region contributes its own model, and the integration between them is what allows hierarchical prediction and, therefore, generality. (These dynamics build directly on the brain-criticality and branching-ratio principles we explored in Volume 8.)
Notably, the experiment is public and observable. Anyone can open their browser and watch how the Nxons evolve generation after generation, how their internal circuits reorganize under the pressure of a fitness function that rewards cognitive generality instead of specialization.
Implications for Artificial Life (Alife) and Applications for Qubic
For the field of artificial life, the explicit incorporation of g as a selection criterion opens a line of work that goes beyond academic exercise. Most Alife systems have evolved agents that solve very concrete niches: foraging, predator avoidance, navigation (Bedau, 2003; Lehman et al., 2020). But few have tried to select for something as abstract as the ability to generalize across heterogeneous cognitive domains. If we manage to get artificial organisms to show positive correlations between distinct tasks (the computational equivalent of Spearman's children) we will have an extraordinary test bench for questions that human psychometrics can only address correlationally: what evolutionary pressures favor the emergence of g? What neural architectures make it possible? Is g a convergent solution or a phylogenetic accident?
For Qubic, this line of research fits with a very concrete vision of the future of #AI . While the industry invests massive resources in scaling transformers over text, Qubic is committed to exploring architecturally alternative paths: modular artificial brains, evolved, distributed, and subjected to real selective pressures. Qubic's decentralized useful-compute network offers the ideal substrate for this kind of experimentation at scale, where thousands of Nxon populations can coevolve in parallel, with fitness functions designed to favor the emergence of g. It is not only open research: it is the possibility of building, on decentralized infrastructure, an empirical alternative to the dominant paradigm of language-based AI, one that starts from the right question (how to measure and select generality) instead of assuming it. If genuine cognitive generality requires architectures inspired by brains and not by corpora, Qubic is one of the few environments where that hypothesis can be seriously put to the test.
A deeper analysis is in preparation, as it forms part of our recent papers and experiments. Spearman's old g, that thread which wove together children's school grades, we now use in digital creatures that learn to survive.
References
Barbey, A. K. (2018). Network neuroscience theory of human intelligence. Trends in Cognitive Sciences, 22(1), 8–20. https://doi.org/10.1016/j.tics.2017.10.001Bedau, M. A. (2003). Artificial life: Organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. https://doi.org/10.1016/j.tics.2003.09.012Burnell, R., Schellaert, W., Burden, J., Ullman, T. D., Martínez-Plumed, F., Tenenbaum, J. B., et al. (2023). Rethink reporting of evaluation results in AI. Science, 380(6641), 136–138. https://doi.org/10.1126/science.adf6369Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. Cambridge University Press. https://doi.org/10.1017/CBO9780511571312Chollet, F. (2019). On the measure of intelligence. arXiv preprint arXiv:1911.01547. https://arxiv.org/abs/1911.01547Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204. https://doi.org/10.1017/S0140525X12000477Cole, M. W., Ito, T., & Braver, T. S. (2015). Lateral prefrontal cortex contributes to fluid intelligence through multinetwork connectivity. Brain Connectivity, 5(8), 497–504. https://doi.org/10.1089/brain.2015.0357Deary, I. J., Penke, L., & Johnson, W. (2010). The neuroscience of human intelligence differences. Nature Reviews Neuroscience, 11(3), 201–211. https://doi.org/10.1038/nrn2793Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138. https://doi.org/10.1038/nrn2787Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24(1), 79–132. https://doi.org/10.1016/S0160-2896(97)90014-3Hohwy, J. (2013). The predictive mind. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199682737.001.0001Ilić, D., & Gignac, G. E. (2024). Evidence of interrelated cognitive-like capabilities in large language models: Indications of artificial general intelligence or achievement? Intelligence, 106, 101858. https://doi.org/10.1016/j.intell.2024.101858Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence: Converging neuroimaging evidence. Behavioral and Brain Sciences, 30(2), 135–154. https://doi.org/10.1017/S0140525X07001185LeCun, Y. (2022). A path towards autonomous machine intelligence. OpenReview, version 0.9.2. https://openreview.net/forum?id=BZ5a1r-kVsfLehman, J., Clune, J., Misevic, D., Adami, C., Altenberg, L., Beaulieu, J., et al. (2020). The surprising creativity of digital evolution. Artificial Life, 26(2), 274–306. https://doi.org/10.1162/artl_a_00319Mitchell, M. (2021). Why AI is harder than we think. arXiv preprint arXiv:2104.12871. https://arxiv.org/abs/2104.12871Spearman, C. (1904). "General intelligence," objectively determined and measured. The American Journal of Psychology, 15(2), 201–292. https://doi.org/10.2307/1412107Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30. https://arxiv.org/abs/1706.03762
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 9 of the #Neuraxon Intelligence Academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time. Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence. Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI. Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Volume 4: Neural Networks in AI and Neuroscience. A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI. How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained. Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Volume 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems. How emergent complexity and self-organized criticality move from simulators to decentralized AI infrastructure.NIA Volume 8: Brain Criticality and the Branching Ratio in Neural and Artificial Networks. Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.NIA Volume 9: The g Factor in Artificial Life. You are here.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org or browse the full Academy and Blog. Join the discussion on X, Discord, and Telegram.
Qubic is a decentralized, open-source network for experimental technology. Nothing on this site should be construed as investment, legal, or financial advice.
In one line: General intelligence, the g factor psychologists have measured for over a century, is the missing ingredient in today's language models, and Qubic's Neuraxon project is now selecting for it directly inside an artificial-life simulation.
The g Factor: From a 1904 Classroom to Artificial Brains
In 1904, Charles Spearman stumbled upon a regularity that would forever change psychology. Examining the school grades of a group of English children, he noticed something seemingly trivial but strange: those who excelled in mathematics also tended to excel in French, in music, in language. Disciplines with no apparent connection correlated systematically with one another. Spearman proposed that beneath this tangle of disparate abilities there lay a single common factor, a general cognitive thread. He called it g (Spearman, 1904).
More than a century later, g remains one of the most replicated findings in the behavioral sciences (Carroll, 1993; Deary et al., 2010). It is neither a grade average nor an arbitrary construct: it is what emerges when factor analysis is applied to almost any battery of cognitive tests. It appears consistently when we measure working memory, fluid reasoning, processing speed, verbal comprehension, or novel problem solving. In psychometric terms, g is the shared variance that no single test measures on its own.
What the g Factor Means in the Brain and in Behavior
P-FIT Theory and Brain Network Efficiency
From cognitive neuroscience, g has ceased to be a statistical abstraction and has become a property of brain architecture. The P-FIT theory (Parieto-Frontal Integration Theory) identifies a distributed network made up of dorsolateral prefrontal cortex, posterior parietal cortex, anterior cingulate, and temporal areas, whose connection efficiency predicts intelligence test scores (Jung & Haier, 2007). Functional connectivity studies show that g correlates with the brain's ability to dynamically reconfigure its networks (the executive control network, the default mode network, the salience network) according to task demands (Barbey, 2018; Cole et al., 2015). It is not about having "more" neurons in a specific place, but about better orchestrating the flow of information between functionally specialized regions.
The Predictive Brain and Free-Energy Minimization
This orchestration acquires an even deeper meaning in light of the predictive brain theory (Clark, 2013; Friston, 2010). Under this framework, the brain is not a passive receiver of stimuli but a hierarchical inference engine that continuously generates predictions about the world and adjusts its internal models based on prediction error. Here g fits naturally: the ability to predict well, to anticipate environmental contingencies, to learn quickly from error and, above all, to abstract regularities that transfer across domains, is precisely what intelligence tests capture indirectly. A brain with high g would be, on this reading, a system with more efficient generative models, capable of compressing experience into high-level abstractions and of minimizing free energy across heterogeneous contexts (Hohwy, 2013); that is, it reduces prediction error rapidly and therefore learns. Cognitive generality, then, would not be a static property of the neural hardware, but the quality of a deeply hierarchical predictive process. The research remains open. Other currents posit that g really has to do with the neurodevelopment of our brain, given that no matter what task we are performing or attempting, there is a huge common factor in any experience because it happens inside the same organ.
Behaviorally, g is the best predictor. Forget emotional intelligence; it is g that best forecasts what your academic performance, occupational success, longevity, and even certain health indicators may be (Deary et al., 2010; Gottfredson, 1997). Not because it is destiny, but because it captures something very basic: the capacity of a cognitive system to face problems it has not seen before, integrating heterogeneous information under time and resource constraints. g is, in a sense, a measure of generality.
The Problem of Measuring General Intelligence in Artificial Systems
For decades, artificial systems have shone in narrow tasks (playing chess, classifying images, translating) but failed to transfer that performance outside their domain (Chollet, 2019). The #AGI debate revolves precisely around this: what does it mean, operationally, for a system to be "generally" intelligent?
If we take the parallel with human psychometrics seriously, the answer is uncomfortable but clear: to speak of generality we need to measure it, and measuring it requires diverse tests whose shared variance reveals something analogous to g. A system with high performance on a single task tells us nothing about its generality; a system with moderate and correlated performance across many structurally distinct tasks does. Spearman's logic, transferred to non-biological substrates, still holds: generality is not postulated, it is factored.
Why the g Factor Does Not Appear in Transformers (and What That Implies for AGI)
It is worth pausing here on the currently dominant paradigm. Large language models based on transformer architectures (Vaswani et al., 2017) deliver astonishing performance on linguistic tasks, but psychometric analyses applied to their outputs do not show the factor structure characteristic of g (Burnell et al., 2023; Ilić & Gignac, 2024). Their hits and misses across domains do not correlate as they would in humans; they depend rather on the density and quality of patterns present in their training data. A transformer can brilliantly solve one problem and fail on another that is structurally equivalent but phrased slightly differently, something a system with genuine g would not do (Mitchell, 2021).
This has serious implications. It suggests that the pursuit of cognitive generality exclusively through language may be a dead end, an architectural dead end. Language is the most visible output of human cognition, but not its substrate. To pretend that by scaling text one will arrive at g is like pretending that by scaling descriptions of chess games one will arrive at mastery: one obtains statistical mimicry, not the underlying cognitive structure. (We argued a closely related point in our analysis of why intelligence is not scale, and on why LLM predictions are not brain predictions.) Without genuine hierarchical prediction, without generative models of the world, without coordination between functionally specialized modules, behavior can look general without being so. The absence of g in transformers is not a failure of scale: it is a clue that generality requires other architectural ingredients (LeCun, 2022).
The g Factor Inside the Neuraxon Game of Life
We have taken this intuition to a different experimental terrain. In Multi-Neuraxon Game of Life Lite 5.0, the artificial creatures (the Nxons) grow their own brains and compete to survive. What is new in this version is that the selective pressure is applied to g. The Nxons are not selected for mastering a specific task, but for showing that common thread that allows them to face many.
The brains of the Nxons have been designed following a simplified model anchored in cognitive neuroscience, since they use six functional regions, inspired by the same kind of maps that psychologists use to describe the modular organization of the human brain. The bet is that generality does not emerge from a monolithic architecture, but from the coordination among specialized regions that share information flexibly. It is the P-FIT intuition translated into artificial life, and it connects directly with the predictive brain principle: each region contributes its own model, and the integration between them is what allows hierarchical prediction and, therefore, generality. (These dynamics build directly on the brain-criticality and branching-ratio principles we explored in Volume 8.)
Notably, the experiment is public and observable. Anyone can open their browser and watch how the Nxons evolve generation after generation, how their internal circuits reorganize under the pressure of a fitness function that rewards cognitive generality instead of specialization.
Implications for Artificial Life (Alife) and Applications for Qubic
For the field of artificial life, the explicit incorporation of g as a selection criterion opens a line of work that goes beyond academic exercise. Most Alife systems have evolved agents that solve very concrete niches: foraging, predator avoidance, navigation (Bedau, 2003; Lehman et al., 2020). But few have tried to select for something as abstract as the ability to generalize across heterogeneous cognitive domains. If we manage to get artificial organisms to show positive correlations between distinct tasks (the computational equivalent of Spearman's children) we will have an extraordinary test bench for questions that human psychometrics can only address correlationally: what evolutionary pressures favor the emergence of g? What neural architectures make it possible? Is g a convergent solution or a phylogenetic accident?
For Qubic, this line of research fits with a very concrete vision of the future of #AI . While the industry invests massive resources in scaling transformers over text, Qubic is committed to exploring architecturally alternative paths: modular artificial brains, evolved, distributed, and subjected to real selective pressures. Qubic's decentralized useful-compute network offers the ideal substrate for this kind of experimentation at scale, where thousands of Nxon populations can coevolve in parallel, with fitness functions designed to favor the emergence of g. It is not only open research: it is the possibility of building, on decentralized infrastructure, an empirical alternative to the dominant paradigm of language-based AI, one that starts from the right question (how to measure and select generality) instead of assuming it. If genuine cognitive generality requires architectures inspired by brains and not by corpora, Qubic is one of the few environments where that hypothesis can be seriously put to the test.
A deeper analysis is in preparation, as it forms part of our recent papers and experiments. Spearman's old g, that thread which wove together children's school grades, we now use in digital creatures that learn to survive.
References
Barbey, A. K. (2018). Network neuroscience theory of human intelligence. Trends in Cognitive Sciences, 22(1), 8–20. https://doi.org/10.1016/j.tics.2017.10.001Bedau, M. A. (2003). Artificial life: Organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. https://doi.org/10.1016/j.tics.2003.09.012Burnell, R., Schellaert, W., Burden, J., Ullman, T. D., Martínez-Plumed, F., Tenenbaum, J. B., et al. (2023). Rethink reporting of evaluation results in AI. Science, 380(6641), 136–138. https://doi.org/10.1126/science.adf6369Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. Cambridge University Press. https://doi.org/10.1017/CBO9780511571312Chollet, F. (2019). On the measure of intelligence. arXiv preprint arXiv:1911.01547. https://arxiv.org/abs/1911.01547Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204. https://doi.org/10.1017/S0140525X12000477Cole, M. W., Ito, T., & Braver, T. S. (2015). Lateral prefrontal cortex contributes to fluid intelligence through multinetwork connectivity. Brain Connectivity, 5(8), 497–504. https://doi.org/10.1089/brain.2015.0357Deary, I. J., Penke, L., & Johnson, W. (2010). The neuroscience of human intelligence differences. Nature Reviews Neuroscience, 11(3), 201–211. https://doi.org/10.1038/nrn2793Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138. https://doi.org/10.1038/nrn2787Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24(1), 79–132. https://doi.org/10.1016/S0160-2896(97)90014-3Hohwy, J. (2013). The predictive mind. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199682737.001.0001Ilić, D., & Gignac, G. E. (2024). Evidence of interrelated cognitive-like capabilities in large language models: Indications of artificial general intelligence or achievement? Intelligence, 106, 101858. https://doi.org/10.1016/j.intell.2024.101858Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of intelligence: Converging neuroimaging evidence. Behavioral and Brain Sciences, 30(2), 135–154. https://doi.org/10.1017/S0140525X07001185LeCun, Y. (2022). A path towards autonomous machine intelligence. OpenReview, version 0.9.2. https://openreview.net/forum?id=BZ5a1r-kVsfLehman, J., Clune, J., Misevic, D., Adami, C., Altenberg, L., Beaulieu, J., et al. (2020). The surprising creativity of digital evolution. Artificial Life, 26(2), 274–306. https://doi.org/10.1162/artl_a_00319Mitchell, M. (2021). Why AI is harder than we think. arXiv preprint arXiv:2104.12871. https://arxiv.org/abs/2104.12871Spearman, C. (1904). "General intelligence," objectively determined and measured. The American Journal of Psychology, 15(2), 201–292. https://doi.org/10.2307/1412107Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30. https://arxiv.org/abs/1706.03762
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 9 of the #Neuraxon Intelligence Academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, Aigarth, and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time. Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence. Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI. Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Volume 4: Neural Networks in AI and Neuroscience. A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI. How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained. Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Volume 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems. How emergent complexity and self-organized criticality move from simulators to decentralized AI infrastructure.NIA Volume 8: Brain Criticality and the Branching Ratio in Neural and Artificial Networks. Why a branching ratio near 1 and self-organized criticality are bioinspired design principles in Neuraxon.NIA Volume 9: The g Factor in Artificial Life. You are here.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org or browse the full Academy and Blog. Join the discussion on X, Discord, and Telegram.
Qubic is a decentralized, open-source network for experimental technology. Nothing on this site should be construed as investment, legal, or financial advice.
1
1
423
·
--
Qubic Bridging 137 Years of Science Into Next-Gen AI Real-World Application! 🧠💻
Many crypto projects stay trapped in theory, but #Qubic is proving its real-world utility at the highest scientific levels.
At the upcoming 11th International Conference on Machine Learning Technologies (May 20-22) in Berlin, researchers David Vivancos and Jose Sánchez are set to unveil "Neuraxon"—a biologically inspired Artificial Neuron computation blueprint.
How is $Qubic making this a reality?
Real-World Infrastructure: Qubic isn’t just a network; it provides the core computational powerhouse needed to simulate complex biological neural growth.
True Open Science: Driven by Qubic’s decentralized ecosystem, empowering global researchers to break AI monopolies.
The Path to True AI: Transitioning from basic machine learning straight into advanced AGI.
History comes full circle in Berlin. In 1889, the first human neuron was shown there. In May 2026, Qubic powers the architecture to replicate it on machines. This is utility. This is the future of AI.
👉https://www.researchgate.net/publication/400868863_Neuraxon_V20_A_New_Neural_Growth_Computation_Blueprint
#Qubic #AI #AGI #Neuraxon
Qubic Bridging 137 Years of Science Into Next-Gen AI Real-World Application! 🧠💻
Many crypto projects stay trapped in theory, but #Qubic is proving its real-world utility at the highest scientific levels.
At the upcoming 11th International Conference on Machine Learning Technologies (May 20-22) in Berlin, researchers David Vivancos and Jose Sánchez are set to unveil "Neuraxon"—a biologically inspired Artificial Neuron computation blueprint.
How is $Qubic making this a reality?
Real-World Infrastructure: Qubic isn’t just a network; it provides the core computational powerhouse needed to simulate complex biological neural growth.
True Open Science: Driven by Qubic’s decentralized ecosystem, empowering global researchers to break AI monopolies.
The Path to True AI: Transitioning from basic machine learning straight into advanced AGI.
History comes full circle in Berlin. In 1889, the first human neuron was shown there. In May 2026, Qubic powers the architecture to replicate it on machines. This is utility. This is the future of AI.
👉https://www.researchgate.net/publication/400868863_Neuraxon_V20_A_New_Neural_Growth_Computation_Blueprint
#Qubic #AI #AGI #Neuraxon
Many crypto projects stay trapped in theory, but #Qubic is proving its real-world utility at the highest scientific levels.
At the upcoming 11th International Conference on Machine Learning Technologies (May 20-22) in Berlin, researchers David Vivancos and Jose Sánchez are set to unveil "Neuraxon"—a biologically inspired Artificial Neuron computation blueprint.
How is $Qubic making this a reality?
Real-World Infrastructure: Qubic isn’t just a network; it provides the core computational powerhouse needed to simulate complex biological neural growth.
True Open Science: Driven by Qubic’s decentralized ecosystem, empowering global researchers to break AI monopolies.
The Path to True AI: Transitioning from basic machine learning straight into advanced AGI.
History comes full circle in Berlin. In 1889, the first human neuron was shown there. In May 2026, Qubic powers the architecture to replicate it on machines. This is utility. This is the future of AI.
👉https://www.researchgate.net/publication/400868863_Neuraxon_V20_A_New_Neural_Growth_Computation_Blueprint
#Qubic #AI #AGI #Neuraxon
342
·
--
IS AI FINALLY LEARNING TO "THINK" LIKE A BRAIN? 🧠✨
Why does the human brain operate at the "Edge of Chaos"? It’s all about a magic principle called Brain Criticality.
In the latest NIA Vol. 8, the Qubic Scientific Team explores the Branching Ratio—the key metric of neural connectivity. When this ratio is near 1, a network achieves:
- Maximal Dynamic Range: Detecting the subtlest signals.
- Optimal Memory: Balancing past information with new inputs.
- Peak Complexity: The hallmark of true intelligence.
See how Neuraxon uses these bio-inspired principles to build AI that doesn't just calculate—it reverberates like a living organism.
👉 Read the full deep dive here: [Brain Criticality in Neuraxon](https://www.binance.com/en/square/post/322900066069841)
#Qubic
#Neuraxon
#DeAI
#SmartContracts
#CryptoAi
IS AI FINALLY LEARNING TO "THINK" LIKE A BRAIN? 🧠✨
Why does the human brain operate at the "Edge of Chaos"? It’s all about a magic principle called Brain Criticality.
In the latest NIA Vol. 8, the Qubic Scientific Team explores the Branching Ratio—the key metric of neural connectivity. When this ratio is near 1, a network achieves:
- Maximal Dynamic Range: Detecting the subtlest signals.
- Optimal Memory: Balancing past information with new inputs.
- Peak Complexity: The hallmark of true intelligence.
See how Neuraxon uses these bio-inspired principles to build AI that doesn't just calculate—it reverberates like a living organism.
👉 Read the full deep dive here: Brain Criticality in Neuraxon
#Qubic
#Neuraxon
#DeAI
#SmartContracts
#CryptoAi
Why does the human brain operate at the "Edge of Chaos"? It’s all about a magic principle called Brain Criticality.
In the latest NIA Vol. 8, the Qubic Scientific Team explores the Branching Ratio—the key metric of neural connectivity. When this ratio is near 1, a network achieves:
- Maximal Dynamic Range: Detecting the subtlest signals.
- Optimal Memory: Balancing past information with new inputs.
- Peak Complexity: The hallmark of true intelligence.
See how Neuraxon uses these bio-inspired principles to build AI that doesn't just calculate—it reverberates like a living organism.
👉 Read the full deep dive here: Brain Criticality in Neuraxon
#Qubic
#Neuraxon
#DeAI
#SmartContracts
#CryptoAi
Luck3333
·
--
Neuraxon: Implementing Brain Criticality in Artificial Networks
Written by Qubic Scientific TeamBranching ratio and criticality in biological networks, in artificial networks, and as a bioinspired principle in Neuraxon
What do a snow avalanche, a forest fire, an earthquake, and the spontaneous activity of the cerebral cortex have in common?
They all share a frontier between order and chaos, what is called a critical state. In the brain, that edge is measured by a simple parameter: the branching ratio (σ or m). It would be something like the average ratio of neuronal "offspring" that each "parent" neuron activates. When σ ≈ 1, activity neither dies out nor explodes; it reverberates.
Beggs and Plenz (2003) recorded the spontaneous activity of the cerebral cortex in rats and found that the activity formed cascade-like patterns, the so-called neuronal avalanches, with a branching ratio close to 1. The brain seemed to live at a critical point. In humans, the branching ratio σ once again appears close to unity (Wang et al., 2025; Plenz et al., 2021; Wilting & Priesemann, 2019).
At the critical point, systems simultaneously exhibit maximal sensitivity to perturbations (responsiveness), maximal dynamic capacity (number of accessible states), maximal information transmission, and maximal complexity (Timme et al., 2016; Shew et al., 2009, 2011).
What Is the Branching Ratio and How Is It Measured?
Conceptually, the branching ratio is trivial: if at instant t there are A(t) active neurons and at t+1 there are A(t+1), then:
σ = ⟨ A(t+1) / A(t) ⟩
Three regimes follow from this (de Carvalho & Prado, 2000; Haldeman & Beggs, 2005):
Subcritical (σ < 1): activity decays; the system "forgets" the perturbation quickly. It is stable but poor in memory and not very expressive.Supercritical (σ > 1): activity explodes into cascades. This is the signature of pathological regimes such as epileptic seizures (Hsu et al., 2008; Hagemann et al., 2021).Critical (σ ≈ 1): each spike, on average, generates another spike. Activity reverberates, neuronal avalanches obey power laws, and the system maintains a structured memory of the input.
The beauty of σ is that it is a single number that summarizes the global dynamical regime. But measuring it is less trivial. When applied to in vivo cortical recordings, the measurement reveals that the cortex does not operate exactly at σ = 1, but slightly below, in a regime that the authors call reverberating (Wilting et al., 2018). The difference is important: being exactly at σ = 1 would be like pedaling a bicycle balanced on a tightrope; being slightly below allows for rapid adjustment to task demands without the risk of runaway explosion.
Criticality in Artificial Neural Networks: From the Edge of Chaos to Reservoir Computing
Bertschinger and Natschläger (2004) showed that random recurrent threshold networks reach their maximal computational capacity on temporal processing tasks precisely at the order–chaos transition.
Boedecker et al. (2012) extended the analysis to echo state networks within the reservoir computing paradigm, confirming that information transfer capacity and active memory are maximized at the edge of chaos.
Fig. 3. A spiking neuromorphic network with synaptic plasticity self-organizes toward criticality under low external input, exhibiting power-law avalanche size distributions — the hallmark of the critical state in both biological and artificial neural networks. Under higher input, the network shifts to a subcritical regime with truncated distributions. Reproduced from Cramer et al. (2020), Nature Communications, 11, 2853. CC BY 4.0.
In the language of artificial neural networks, the measurement parameter is called the spectral radius. When it exceeds 1, trajectories diverge exponentially (chaos); when it is well below 1, the network collapses to the fixed point and loses memory. The spectral radius close to 1 is, in this context, the formal equivalent of the biological σ ≈ 1 (Magnasco, 2022; Morales et al., 2023). In spiking neural networks, the branching ratio can be measured with methods almost identical to those used in neuronal cultures (Cramer et al., 2020; Zeraati et al., 2024).
Why Does Brain Criticality Maximize Neural Computation?
Operating close to σ ≈ 1 provides four advantages that are central to both the critical brain hypothesis and the design of brain-inspired AI systems:
Maximal dynamic range. Shew et al. (2009) showed that the range of input intensities the cortex can discriminate is maximal when the excitation–inhibition balance places the network at criticality.Maximized information capacity. The entropy of avalanche patterns and the mutual information between input and output peak at σ ≈ 1 (Shew et al., 2011).Optimal fading memory. In the critical regime, the perturbation is sustained just long enough to influence processing without contaminating the distant future; it is the sweet spot between stability and temporal integration (Boedecker et al., 2012).Complexity as a unifying measure. Timme et al. (2016) demonstrated that neural complexity is maximized exactly at the critical point, linking criticality with formal theories of consciousness and processing.
Fig. 4. Four computational advantages of operating near the critical branching ratio (σ ≈ 1). At criticality, neural networks achieve maximal dynamic range, maximized information capacity, optimal fading memory, and maximum complexity — properties that are central to both the critical brain hypothesis and brain-inspired AI design.
The Brain Does Not Always Operate at σ = 1
This does not imply that the brain always operates at σ = 1. Evidence rather suggests a slightly subcritical and modulable regime: during demanding tasks the network approaches criticality, during deep sleep it moves away, and pathological states (epilepsy, deep anesthesia, certain psychiatric conditions) are associated with measurable deviations from this operational range (Meisel et al., 2017; Zimmern, 2020). The branching ratio is becoming a dynamic biomarker of the functional state of the nervous system.
Why We Use the Branching Ratio in Neuraxon: Bioinspired AI Design at the Edge of Chaos
Neuraxon is a bioinspired system that adopts dynamical principles of the cortex as design constraints. The branching ratio is one of the most important, and we use it for four reasons:
As a Real-Time Operational Invariant for Neural Network Stability
In deep spiking or recurrent architectures, the dual risk of activity collapse (silent network, vanishing gradients) and runaway explosion (saturation, exploding gradients) is structural. Monitoring σ in real time gives us a single diagnostic scalar, independent of the concrete architecture, that indicates whether the system is alive in the computational sense.
As a Bioinspired Self-Regulation Target Through Self-Organized Criticality
The network self-organizes toward criticality without the need for centralized fine-tuning, replicating the principle of self-organized criticality (Bornholdt & Röhl, 2003; Levina et al., 2007). This drastically reduces sensitivity to hyperparameters and endows the system with robustness against distribution shifts. As we explored in NIA Volume 7 on artificial life and digital ecosystems, this is exactly how emergent complexity arises from local rules without centralized control.
Fig. 5. Neuraxon 3D network during active simulation, showing cascading activity across ternary-state neurons. Brightly active nodes (pink) propagate signals through excitatory (green) and inhibitory (pink) connections while other neurons remain at rest (gray), illustrating a reverberating regime near the critical branching ratio (σ ≈ 1). This balanced state — neither silent nor explosive — is what Neuraxon self-organizes toward using bioinspired criticality principles. Explore the interactive demo athuggingface.co/spaces/DavidVivancos/Neuraxon. Source: Qubic Scientific Team.
As a Bridge Between Neuroscientific Observation and AI Design
The branching ratio is one of the very few magnitudes that is measured with the same formalism in electrophysiology, fMRI, and artificial networks. This allows for testing bidirectional hypotheses: if an intervention improves biological criticality, we can ask whether the same intervention — translated into the artificial architecture — improves the model's computation, and vice versa. This principle is central to the neuromodulation framework and the astrocytic gating mechanisms we have developed in previous volumes of this academy.
As a Functional, Not Aesthetic, Criterion for Brain-Inspired AI
Criticality is an operational constraint with empirical consequences. Operating near the reverberating regime improves — as measured in our internal evaluations and submitted publications — generalization capacity, stability under input perturbations, representational richness, and the temporal coherence of reasoning. These effects qualitatively match those reported in both the biological (Cocchi et al., 2017) and artificial (Cramer et al., 2020; Morales et al., 2023) literature.
The Branching Ratio: From Statistical Physics to Brain-Inspired AI Architecture
The branching ratio is one of those conceptual rara avis: simple enough to reduce to a single formula, deep enough to bridge statistical physics, neuroscience, AI, and systems design. For the biological brain, σ ≈ 1 seems to be the regime where the virtuous combination of sensitivity, memory, expressiveness, and robustness emerges. For artificial networks, the same frontier — rebranded as the edge of chaos — predicts maximal computational capacity.
And for Neuraxon, it is a guiding principle of bioinspired design: an auditable, self-regulating, and biologically meaningful metric that helps us keep the system alive, in the richest sense of the word.
References
Beggs, J. M., & Plenz, D. (2003). Neuronal avalanches in neocortical circuits. The Journal of Neuroscience, 23(35), 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003Bertschinger, N., & Natschläger, T. (2004). Real-time computation at the edge of chaos in recurrent neural networks. Neural Computation, 16(7), 1413–1436. https://doi.org/10.1162/089976604323057443Boedecker, J., Obst, O., Lizier, J. T., Mayer, N. M., & Asada, M. (2012). Information processing in echo state networks at the edge of chaos. Theory in Biosciences, 131(3), 205–213. https://doi.org/10.1007/s12064-011-0146-8Bornholdt, S., & Röhl, T. (2003). Self-organized critical neural networks. Physical Review E, 67(6), 066118. https://doi.org/10.1103/PhysRevE.67.066118Cocchi, L., Gollo, L. L., Zalesky, A., & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. https://doi.org/10.1016/j.pneurobio.2017.07.002Cramer, B., Stöckel, D., Kreft, M., Wibral, M., Schemmel, J., Meier, K., & Priesemann, V. (2020). Control of criticality and computation in spiking neuromorphic networks with plasticity. Nature Communications, 11, 2853. https://doi.org/10.1038/s41467-020-16548-3de Carvalho, J. X., & Prado, C. P. C. (2000). Self-organized criticality in the Olami-Feder-Christensen model. Physical Review Letters, 84(17), 4006–4009. https://doi.org/10.1103/PhysRevLett.84.4006Derrida, B., & Pomeau, Y. (1986). Random networks of automata: A simple annealed approximation. Europhysics Letters, 1(2), 45–49. https://doi.org/10.1209/0295-5075/1/2/001Hagemann, A., Wilting, J., Samimizad, B., Mormann, F., & Priesemann, V. (2021). Assessing criticality in pre-seizure single-neuron activity of human epileptic cortex. PLOS Computational Biology, 17(3), e1008773. https://doi.org/10.1371/journal.pcbi.1008773Haldeman, C., & Beggs, J. M. (2005). Critical branching captures activity in living neural networks and maximizes the number of metastable states. Physical Review Letters, 94(5), 058101. https://doi.org/10.1103/PhysRevLett.94.058101Hsu, D., Chen, W., Hsu, M., & Beggs, J. M. (2008). An open hypothesis: Is epilepsy learned, and can it be unlearned? Epilepsy & Behavior, 13(3), 511–522. https://doi.org/10.1016/j.yebeh.2008.05.007Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42(1–3), 12–37. https://doi.org/10.1016/0167-2789(90)90064-VLevina, A., Herrmann, J. M., & Geisel, T. (2007). Dynamical synapses causing self-organized criticality in neural networks. Nature Physics, 3(12), 857–860. https://doi.org/10.1038/nphys758Magnasco, M. O. (2022). Robustness and flexibility of neural function through dynamical criticality. Entropy, 24(5), 591. https://doi.org/10.3390/e24050591Meisel, C., Klaus, A., Vyazovskiy, V. V., & Plenz, D. (2017). The interplay between long- and short-range temporal correlations shapes cortex dynamics across vigilance states. The Journal of Neuroscience, 37(42), 10114–10124. https://doi.org/10.1523/JNEUROSCI.0448-17.2017Morales, G. B., di Santo, S., & Muñoz, M. A. (2023). Unveiling the intrinsic dynamics of biological and artificial neural networks: From criticality to optimal representations. Frontiers in Complex Systems, 1, 1276338. https://doi.org/10.3389/fcpxs.2023.1276338Plenz, D., Ribeiro, T. L., Miller, S. R., Kells, P. A., Vakili, A., & Capek, E. L. (2021). Self-organized criticality in the brain. Frontiers in Physics, 9, 639389. https://doi.org/10.3389/fphy.2021.639389Shew, W. L., Yang, H., Petermann, T., Roy, R., & Plenz, D. (2009). Neuronal avalanches imply maximum dynamic range in cortical networks at criticality. The Journal of Neuroscience, 29(49), 15595–15600. https://doi.org/10.1523/JNEUROSCI.3864-09.2009Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information capacity and transmission are maximized in balanced cortical networks with neuronal avalanches. The Journal of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011Spitzner, F. P., Dehning, J., Wilting, J., Hagemann, A., Neto, J. P., Zierenberg, J., & Priesemann, V. (2021). MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity. PLOS ONE, 16(4), e0249447. https://doi.org/10.1371/journal.pone.0249447Timme, N. M., Marshall, N. J., Bennett, N., Ripp, M., Lautzenhiser, E., & Beggs, J. M. (2016). Criticality maximizes complexity in neural tissue. Frontiers in Physiology, 7, 425. https://doi.org/10.3389/fphys.2016.00425Turrigiano, G. G. (2008). The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell, 135(3), 422–435. https://doi.org/10.1016/j.cell.2008.10.008Wang, J., Cao, R., Brunton, B. W., Smith, R. E. W., Buckner, R. L., & Liu, T. T. (2025). Genetic contributions to brain criticality and its relationship with human cognitive functions. Proceedings of the National Academy of Sciences, 122(26), e2417010122. https://doi.org/10.1073/pnas.2417010122Wilting, J., Dehning, J., Pinheiro Neto, J., Rudelt, L., Wibral, M., Zierenberg, J., & Priesemann, V. (2018). Operating in a reverberating regime enables rapid tuning of network states to task requirements. Frontiers in Systems Neuroscience, 12, 55. https://doi.org/10.3389/fnsys.2018.00055Wilting, J., & Priesemann, V. (2018). Inferring collective dynamical states from widely unobserved systems. Nature Communications, 9, 2325. https://doi.org/10.1038/s41467-018-04725-4Wilting, J., & Priesemann, V. (2019). 25 years of criticality in neuroscience — Established results, open controversies, novel concepts. Current Opinion in Neurobiology, 58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002Yu, C. (2022). Toward a unified analysis of the brain criticality hypothesis: Reviewing several available tools. Frontiers in Neural Circuits, 16, 911245. https://doi.org/10.3389/fncir.2022.911245Zeraati, R., Engel, T. A., & Levina, A. (2024). Estimating intrinsic timescales and criticality from neural recordings: Methods and pitfalls. Current Opinion in Neurobiology, 86, 102871. https://doi.org/10.1016/j.conb.2024.102871Zimmern, V. (2020). Why brain criticality is clinically relevant: A scoping review. Frontiers in Neural Circuits, 14, 54. https://doi.org/10.3389/fncir.2020.00054
Explore the Complete Neuraxon Intelligence Academy
This is Volume 8 of the #Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, #aigarth , and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Vol. 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Vol. 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Vol. 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Vol. 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Vol. 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Vol. 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Vol. 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's approach to emergent complexity and self-organized criticality in decentralized AI.
Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
What do a snow avalanche, a forest fire, an earthquake, and the spontaneous activity of the cerebral cortex have in common?
They all share a frontier between order and chaos, what is called a critical state. In the brain, that edge is measured by a simple parameter: the branching ratio (σ or m). It would be something like the average ratio of neuronal "offspring" that each "parent" neuron activates. When σ ≈ 1, activity neither dies out nor explodes; it reverberates.
Beggs and Plenz (2003) recorded the spontaneous activity of the cerebral cortex in rats and found that the activity formed cascade-like patterns, the so-called neuronal avalanches, with a branching ratio close to 1. The brain seemed to live at a critical point. In humans, the branching ratio σ once again appears close to unity (Wang et al., 2025; Plenz et al., 2021; Wilting & Priesemann, 2019).
At the critical point, systems simultaneously exhibit maximal sensitivity to perturbations (responsiveness), maximal dynamic capacity (number of accessible states), maximal information transmission, and maximal complexity (Timme et al., 2016; Shew et al., 2009, 2011).
What Is the Branching Ratio and How Is It Measured?
Conceptually, the branching ratio is trivial: if at instant t there are A(t) active neurons and at t+1 there are A(t+1), then:
σ = ⟨ A(t+1) / A(t) ⟩
Three regimes follow from this (de Carvalho & Prado, 2000; Haldeman & Beggs, 2005):
Subcritical (σ < 1): activity decays; the system "forgets" the perturbation quickly. It is stable but poor in memory and not very expressive.Supercritical (σ > 1): activity explodes into cascades. This is the signature of pathological regimes such as epileptic seizures (Hsu et al., 2008; Hagemann et al., 2021).Critical (σ ≈ 1): each spike, on average, generates another spike. Activity reverberates, neuronal avalanches obey power laws, and the system maintains a structured memory of the input.
The beauty of σ is that it is a single number that summarizes the global dynamical regime. But measuring it is less trivial. When applied to in vivo cortical recordings, the measurement reveals that the cortex does not operate exactly at σ = 1, but slightly below, in a regime that the authors call reverberating (Wilting et al., 2018). The difference is important: being exactly at σ = 1 would be like pedaling a bicycle balanced on a tightrope; being slightly below allows for rapid adjustment to task demands without the risk of runaway explosion.
Criticality in Artificial Neural Networks: From the Edge of Chaos to Reservoir Computing
Bertschinger and Natschläger (2004) showed that random recurrent threshold networks reach their maximal computational capacity on temporal processing tasks precisely at the order–chaos transition.
Boedecker et al. (2012) extended the analysis to echo state networks within the reservoir computing paradigm, confirming that information transfer capacity and active memory are maximized at the edge of chaos.
Fig. 3. A spiking neuromorphic network with synaptic plasticity self-organizes toward criticality under low external input, exhibiting power-law avalanche size distributions — the hallmark of the critical state in both biological and artificial neural networks. Under higher input, the network shifts to a subcritical regime with truncated distributions. Reproduced from Cramer et al. (2020), Nature Communications, 11, 2853. CC BY 4.0.
In the language of artificial neural networks, the measurement parameter is called the spectral radius. When it exceeds 1, trajectories diverge exponentially (chaos); when it is well below 1, the network collapses to the fixed point and loses memory. The spectral radius close to 1 is, in this context, the formal equivalent of the biological σ ≈ 1 (Magnasco, 2022; Morales et al., 2023). In spiking neural networks, the branching ratio can be measured with methods almost identical to those used in neuronal cultures (Cramer et al., 2020; Zeraati et al., 2024).
Why Does Brain Criticality Maximize Neural Computation?
Operating close to σ ≈ 1 provides four advantages that are central to both the critical brain hypothesis and the design of brain-inspired AI systems:
Maximal dynamic range. Shew et al. (2009) showed that the range of input intensities the cortex can discriminate is maximal when the excitation–inhibition balance places the network at criticality.Maximized information capacity. The entropy of avalanche patterns and the mutual information between input and output peak at σ ≈ 1 (Shew et al., 2011).Optimal fading memory. In the critical regime, the perturbation is sustained just long enough to influence processing without contaminating the distant future; it is the sweet spot between stability and temporal integration (Boedecker et al., 2012).Complexity as a unifying measure. Timme et al. (2016) demonstrated that neural complexity is maximized exactly at the critical point, linking criticality with formal theories of consciousness and processing.
Fig. 4. Four computational advantages of operating near the critical branching ratio (σ ≈ 1). At criticality, neural networks achieve maximal dynamic range, maximized information capacity, optimal fading memory, and maximum complexity — properties that are central to both the critical brain hypothesis and brain-inspired AI design.
The Brain Does Not Always Operate at σ = 1
This does not imply that the brain always operates at σ = 1. Evidence rather suggests a slightly subcritical and modulable regime: during demanding tasks the network approaches criticality, during deep sleep it moves away, and pathological states (epilepsy, deep anesthesia, certain psychiatric conditions) are associated with measurable deviations from this operational range (Meisel et al., 2017; Zimmern, 2020). The branching ratio is becoming a dynamic biomarker of the functional state of the nervous system.
Why We Use the Branching Ratio in Neuraxon: Bioinspired AI Design at the Edge of Chaos
Neuraxon is a bioinspired system that adopts dynamical principles of the cortex as design constraints. The branching ratio is one of the most important, and we use it for four reasons:
As a Real-Time Operational Invariant for Neural Network Stability
In deep spiking or recurrent architectures, the dual risk of activity collapse (silent network, vanishing gradients) and runaway explosion (saturation, exploding gradients) is structural. Monitoring σ in real time gives us a single diagnostic scalar, independent of the concrete architecture, that indicates whether the system is alive in the computational sense.
As a Bioinspired Self-Regulation Target Through Self-Organized Criticality
The network self-organizes toward criticality without the need for centralized fine-tuning, replicating the principle of self-organized criticality (Bornholdt & Röhl, 2003; Levina et al., 2007). This drastically reduces sensitivity to hyperparameters and endows the system with robustness against distribution shifts. As we explored in NIA Volume 7 on artificial life and digital ecosystems, this is exactly how emergent complexity arises from local rules without centralized control.
Fig. 5. Neuraxon 3D network during active simulation, showing cascading activity across ternary-state neurons. Brightly active nodes (pink) propagate signals through excitatory (green) and inhibitory (pink) connections while other neurons remain at rest (gray), illustrating a reverberating regime near the critical branching ratio (σ ≈ 1). This balanced state — neither silent nor explosive — is what Neuraxon self-organizes toward using bioinspired criticality principles. Explore the interactive demo athuggingface.co/spaces/DavidVivancos/Neuraxon. Source: Qubic Scientific Team.
As a Bridge Between Neuroscientific Observation and AI Design
The branching ratio is one of the very few magnitudes that is measured with the same formalism in electrophysiology, fMRI, and artificial networks. This allows for testing bidirectional hypotheses: if an intervention improves biological criticality, we can ask whether the same intervention — translated into the artificial architecture — improves the model's computation, and vice versa. This principle is central to the neuromodulation framework and the astrocytic gating mechanisms we have developed in previous volumes of this academy.
As a Functional, Not Aesthetic, Criterion for Brain-Inspired AI
Criticality is an operational constraint with empirical consequences. Operating near the reverberating regime improves — as measured in our internal evaluations and submitted publications — generalization capacity, stability under input perturbations, representational richness, and the temporal coherence of reasoning. These effects qualitatively match those reported in both the biological (Cocchi et al., 2017) and artificial (Cramer et al., 2020; Morales et al., 2023) literature.
The Branching Ratio: From Statistical Physics to Brain-Inspired AI Architecture
The branching ratio is one of those conceptual rara avis: simple enough to reduce to a single formula, deep enough to bridge statistical physics, neuroscience, AI, and systems design. For the biological brain, σ ≈ 1 seems to be the regime where the virtuous combination of sensitivity, memory, expressiveness, and robustness emerges. For artificial networks, the same frontier — rebranded as the edge of chaos — predicts maximal computational capacity.
And for Neuraxon, it is a guiding principle of bioinspired design: an auditable, self-regulating, and biologically meaningful metric that helps us keep the system alive, in the richest sense of the word.
References
Beggs, J. M., & Plenz, D. (2003). Neuronal avalanches in neocortical circuits. The Journal of Neuroscience, 23(35), 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003Bertschinger, N., & Natschläger, T. (2004). Real-time computation at the edge of chaos in recurrent neural networks. Neural Computation, 16(7), 1413–1436. https://doi.org/10.1162/089976604323057443Boedecker, J., Obst, O., Lizier, J. T., Mayer, N. M., & Asada, M. (2012). Information processing in echo state networks at the edge of chaos. Theory in Biosciences, 131(3), 205–213. https://doi.org/10.1007/s12064-011-0146-8Bornholdt, S., & Röhl, T. (2003). Self-organized critical neural networks. Physical Review E, 67(6), 066118. https://doi.org/10.1103/PhysRevE.67.066118Cocchi, L., Gollo, L. L., Zalesky, A., & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. https://doi.org/10.1016/j.pneurobio.2017.07.002Cramer, B., Stöckel, D., Kreft, M., Wibral, M., Schemmel, J., Meier, K., & Priesemann, V. (2020). Control of criticality and computation in spiking neuromorphic networks with plasticity. Nature Communications, 11, 2853. https://doi.org/10.1038/s41467-020-16548-3de Carvalho, J. X., & Prado, C. P. C. (2000). Self-organized criticality in the Olami-Feder-Christensen model. Physical Review Letters, 84(17), 4006–4009. https://doi.org/10.1103/PhysRevLett.84.4006Derrida, B., & Pomeau, Y. (1986). Random networks of automata: A simple annealed approximation. Europhysics Letters, 1(2), 45–49. https://doi.org/10.1209/0295-5075/1/2/001Hagemann, A., Wilting, J., Samimizad, B., Mormann, F., & Priesemann, V. (2021). Assessing criticality in pre-seizure single-neuron activity of human epileptic cortex. PLOS Computational Biology, 17(3), e1008773. https://doi.org/10.1371/journal.pcbi.1008773Haldeman, C., & Beggs, J. M. (2005). Critical branching captures activity in living neural networks and maximizes the number of metastable states. Physical Review Letters, 94(5), 058101. https://doi.org/10.1103/PhysRevLett.94.058101Hsu, D., Chen, W., Hsu, M., & Beggs, J. M. (2008). An open hypothesis: Is epilepsy learned, and can it be unlearned? Epilepsy & Behavior, 13(3), 511–522. https://doi.org/10.1016/j.yebeh.2008.05.007Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42(1–3), 12–37. https://doi.org/10.1016/0167-2789(90)90064-VLevina, A., Herrmann, J. M., & Geisel, T. (2007). Dynamical synapses causing self-organized criticality in neural networks. Nature Physics, 3(12), 857–860. https://doi.org/10.1038/nphys758Magnasco, M. O. (2022). Robustness and flexibility of neural function through dynamical criticality. Entropy, 24(5), 591. https://doi.org/10.3390/e24050591Meisel, C., Klaus, A., Vyazovskiy, V. V., & Plenz, D. (2017). The interplay between long- and short-range temporal correlations shapes cortex dynamics across vigilance states. The Journal of Neuroscience, 37(42), 10114–10124. https://doi.org/10.1523/JNEUROSCI.0448-17.2017Morales, G. B., di Santo, S., & Muñoz, M. A. (2023). Unveiling the intrinsic dynamics of biological and artificial neural networks: From criticality to optimal representations. Frontiers in Complex Systems, 1, 1276338. https://doi.org/10.3389/fcpxs.2023.1276338Plenz, D., Ribeiro, T. L., Miller, S. R., Kells, P. A., Vakili, A., & Capek, E. L. (2021). Self-organized criticality in the brain. Frontiers in Physics, 9, 639389. https://doi.org/10.3389/fphy.2021.639389Shew, W. L., Yang, H., Petermann, T., Roy, R., & Plenz, D. (2009). Neuronal avalanches imply maximum dynamic range in cortical networks at criticality. The Journal of Neuroscience, 29(49), 15595–15600. https://doi.org/10.1523/JNEUROSCI.3864-09.2009Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information capacity and transmission are maximized in balanced cortical networks with neuronal avalanches. The Journal of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011Spitzner, F. P., Dehning, J., Wilting, J., Hagemann, A., Neto, J. P., Zierenberg, J., & Priesemann, V. (2021). MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity. PLOS ONE, 16(4), e0249447. https://doi.org/10.1371/journal.pone.0249447Timme, N. M., Marshall, N. J., Bennett, N., Ripp, M., Lautzenhiser, E., & Beggs, J. M. (2016). Criticality maximizes complexity in neural tissue. Frontiers in Physiology, 7, 425. https://doi.org/10.3389/fphys.2016.00425Turrigiano, G. G. (2008). The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell, 135(3), 422–435. https://doi.org/10.1016/j.cell.2008.10.008Wang, J., Cao, R., Brunton, B. W., Smith, R. E. W., Buckner, R. L., & Liu, T. T. (2025). Genetic contributions to brain criticality and its relationship with human cognitive functions. Proceedings of the National Academy of Sciences, 122(26), e2417010122. https://doi.org/10.1073/pnas.2417010122Wilting, J., Dehning, J., Pinheiro Neto, J., Rudelt, L., Wibral, M., Zierenberg, J., & Priesemann, V. (2018). Operating in a reverberating regime enables rapid tuning of network states to task requirements. Frontiers in Systems Neuroscience, 12, 55. https://doi.org/10.3389/fnsys.2018.00055Wilting, J., & Priesemann, V. (2018). Inferring collective dynamical states from widely unobserved systems. Nature Communications, 9, 2325. https://doi.org/10.1038/s41467-018-04725-4Wilting, J., & Priesemann, V. (2019). 25 years of criticality in neuroscience — Established results, open controversies, novel concepts. Current Opinion in Neurobiology, 58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002Yu, C. (2022). Toward a unified analysis of the brain criticality hypothesis: Reviewing several available tools. Frontiers in Neural Circuits, 16, 911245. https://doi.org/10.3389/fncir.2022.911245Zeraati, R., Engel, T. A., & Levina, A. (2024). Estimating intrinsic timescales and criticality from neural recordings: Methods and pitfalls. Current Opinion in Neurobiology, 86, 102871. https://doi.org/10.1016/j.conb.2024.102871Zimmern, V. (2020). Why brain criticality is clinically relevant: A scoping review. Frontiers in Neural Circuits, 14, 54. https://doi.org/10.3389/fncir.2020.00054
Explore the Complete Neuraxon Intelligence Academy
This is Volume 8 of the #Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, #aigarth , and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Vol. 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Vol. 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Vol. 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Vol. 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Vol. 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Vol. 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Vol. 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's approach to emergent complexity and self-organized criticality in decentralized AI.
Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
130
·
--
Článok
Neuraxon: Implementing Brain Criticality in Artificial NetworksWritten by Qubic Scientific TeamBranching ratio and criticality in biological networks, in artificial networks, and as a bioinspired principle in Neuraxon
What do a snow avalanche, a forest fire, an earthquake, and the spontaneous activity of the cerebral cortex have in common?
They all share a frontier between order and chaos, what is called a critical state. In the brain, that edge is measured by a simple parameter: the branching ratio (σ or m). It would be something like the average ratio of neuronal "offspring" that each "parent" neuron activates. When σ ≈ 1, activity neither dies out nor explodes; it reverberates.
Beggs and Plenz (2003) recorded the spontaneous activity of the cerebral cortex in rats and found that the activity formed cascade-like patterns, the so-called neuronal avalanches, with a branching ratio close to 1. The brain seemed to live at a critical point. In humans, the branching ratio σ once again appears close to unity (Wang et al., 2025; Plenz et al., 2021; Wilting & Priesemann, 2019).
At the critical point, systems simultaneously exhibit maximal sensitivity to perturbations (responsiveness), maximal dynamic capacity (number of accessible states), maximal information transmission, and maximal complexity (Timme et al., 2016; Shew et al., 2009, 2011).
What Is the Branching Ratio and How Is It Measured?
Conceptually, the branching ratio is trivial: if at instant t there are A(t) active neurons and at t+1 there are A(t+1), then:
σ = ⟨ A(t+1) / A(t) ⟩
Three regimes follow from this (de Carvalho & Prado, 2000; Haldeman & Beggs, 2005):
Subcritical (σ < 1): activity decays; the system "forgets" the perturbation quickly. It is stable but poor in memory and not very expressive.Supercritical (σ > 1): activity explodes into cascades. This is the signature of pathological regimes such as epileptic seizures (Hsu et al., 2008; Hagemann et al., 2021).Critical (σ ≈ 1): each spike, on average, generates another spike. Activity reverberates, neuronal avalanches obey power laws, and the system maintains a structured memory of the input.
The beauty of σ is that it is a single number that summarizes the global dynamical regime. But measuring it is less trivial. When applied to in vivo cortical recordings, the measurement reveals that the cortex does not operate exactly at σ = 1, but slightly below, in a regime that the authors call reverberating (Wilting et al., 2018). The difference is important: being exactly at σ = 1 would be like pedaling a bicycle balanced on a tightrope; being slightly below allows for rapid adjustment to task demands without the risk of runaway explosion.
Criticality in Artificial Neural Networks: From the Edge of Chaos to Reservoir Computing
Bertschinger and Natschläger (2004) showed that random recurrent threshold networks reach their maximal computational capacity on temporal processing tasks precisely at the order–chaos transition.
Boedecker et al. (2012) extended the analysis to echo state networks within the reservoir computing paradigm, confirming that information transfer capacity and active memory are maximized at the edge of chaos.
Fig. 3. A spiking neuromorphic network with synaptic plasticity self-organizes toward criticality under low external input, exhibiting power-law avalanche size distributions — the hallmark of the critical state in both biological and artificial neural networks. Under higher input, the network shifts to a subcritical regime with truncated distributions. Reproduced from Cramer et al. (2020), Nature Communications, 11, 2853. CC BY 4.0.
In the language of artificial neural networks, the measurement parameter is called the spectral radius. When it exceeds 1, trajectories diverge exponentially (chaos); when it is well below 1, the network collapses to the fixed point and loses memory. The spectral radius close to 1 is, in this context, the formal equivalent of the biological σ ≈ 1 (Magnasco, 2022; Morales et al., 2023). In spiking neural networks, the branching ratio can be measured with methods almost identical to those used in neuronal cultures (Cramer et al., 2020; Zeraati et al., 2024).
Why Does Brain Criticality Maximize Neural Computation?
Operating close to σ ≈ 1 provides four advantages that are central to both the critical brain hypothesis and the design of brain-inspired AI systems:
Maximal dynamic range. Shew et al. (2009) showed that the range of input intensities the cortex can discriminate is maximal when the excitation–inhibition balance places the network at criticality.Maximized information capacity. The entropy of avalanche patterns and the mutual information between input and output peak at σ ≈ 1 (Shew et al., 2011).Optimal fading memory. In the critical regime, the perturbation is sustained just long enough to influence processing without contaminating the distant future; it is the sweet spot between stability and temporal integration (Boedecker et al., 2012).Complexity as a unifying measure. Timme et al. (2016) demonstrated that neural complexity is maximized exactly at the critical point, linking criticality with formal theories of consciousness and processing.
Fig. 4. Four computational advantages of operating near the critical branching ratio (σ ≈ 1). At criticality, neural networks achieve maximal dynamic range, maximized information capacity, optimal fading memory, and maximum complexity — properties that are central to both the critical brain hypothesis and brain-inspired AI design.
The Brain Does Not Always Operate at σ = 1
This does not imply that the brain always operates at σ = 1. Evidence rather suggests a slightly subcritical and modulable regime: during demanding tasks the network approaches criticality, during deep sleep it moves away, and pathological states (epilepsy, deep anesthesia, certain psychiatric conditions) are associated with measurable deviations from this operational range (Meisel et al., 2017; Zimmern, 2020). The branching ratio is becoming a dynamic biomarker of the functional state of the nervous system.
Why We Use the Branching Ratio in Neuraxon: Bioinspired AI Design at the Edge of Chaos
Neuraxon is a bioinspired system that adopts dynamical principles of the cortex as design constraints. The branching ratio is one of the most important, and we use it for four reasons:
As a Real-Time Operational Invariant for Neural Network Stability
In deep spiking or recurrent architectures, the dual risk of activity collapse (silent network, vanishing gradients) and runaway explosion (saturation, exploding gradients) is structural. Monitoring σ in real time gives us a single diagnostic scalar, independent of the concrete architecture, that indicates whether the system is alive in the computational sense.
As a Bioinspired Self-Regulation Target Through Self-Organized Criticality
The network self-organizes toward criticality without the need for centralized fine-tuning, replicating the principle of self-organized criticality (Bornholdt & Röhl, 2003; Levina et al., 2007). This drastically reduces sensitivity to hyperparameters and endows the system with robustness against distribution shifts. As we explored in NIA Volume 7 on artificial life and digital ecosystems, this is exactly how emergent complexity arises from local rules without centralized control.
Fig. 5. Neuraxon 3D network during active simulation, showing cascading activity across ternary-state neurons. Brightly active nodes (pink) propagate signals through excitatory (green) and inhibitory (pink) connections while other neurons remain at rest (gray), illustrating a reverberating regime near the critical branching ratio (σ ≈ 1). This balanced state — neither silent nor explosive — is what Neuraxon self-organizes toward using bioinspired criticality principles. Explore the interactive demo athuggingface.co/spaces/DavidVivancos/Neuraxon. Source: Qubic Scientific Team.
As a Bridge Between Neuroscientific Observation and AI Design
The branching ratio is one of the very few magnitudes that is measured with the same formalism in electrophysiology, fMRI, and artificial networks. This allows for testing bidirectional hypotheses: if an intervention improves biological criticality, we can ask whether the same intervention — translated into the artificial architecture — improves the model's computation, and vice versa. This principle is central to the neuromodulation framework and the astrocytic gating mechanisms we have developed in previous volumes of this academy.
As a Functional, Not Aesthetic, Criterion for Brain-Inspired AI
Criticality is an operational constraint with empirical consequences. Operating near the reverberating regime improves — as measured in our internal evaluations and submitted publications — generalization capacity, stability under input perturbations, representational richness, and the temporal coherence of reasoning. These effects qualitatively match those reported in both the biological (Cocchi et al., 2017) and artificial (Cramer et al., 2020; Morales et al., 2023) literature.
The Branching Ratio: From Statistical Physics to Brain-Inspired AI Architecture
The branching ratio is one of those conceptual rara avis: simple enough to reduce to a single formula, deep enough to bridge statistical physics, neuroscience, AI, and systems design. For the biological brain, σ ≈ 1 seems to be the regime where the virtuous combination of sensitivity, memory, expressiveness, and robustness emerges. For artificial networks, the same frontier — rebranded as the edge of chaos — predicts maximal computational capacity.
And for Neuraxon, it is a guiding principle of bioinspired design: an auditable, self-regulating, and biologically meaningful metric that helps us keep the system alive, in the richest sense of the word.
References
Beggs, J. M., & Plenz, D. (2003). Neuronal avalanches in neocortical circuits. The Journal of Neuroscience, 23(35), 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003Bertschinger, N., & Natschläger, T. (2004). Real-time computation at the edge of chaos in recurrent neural networks. Neural Computation, 16(7), 1413–1436. https://doi.org/10.1162/089976604323057443Boedecker, J., Obst, O., Lizier, J. T., Mayer, N. M., & Asada, M. (2012). Information processing in echo state networks at the edge of chaos. Theory in Biosciences, 131(3), 205–213. https://doi.org/10.1007/s12064-011-0146-8Bornholdt, S., & Röhl, T. (2003). Self-organized critical neural networks. Physical Review E, 67(6), 066118. https://doi.org/10.1103/PhysRevE.67.066118Cocchi, L., Gollo, L. L., Zalesky, A., & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. https://doi.org/10.1016/j.pneurobio.2017.07.002Cramer, B., Stöckel, D., Kreft, M., Wibral, M., Schemmel, J., Meier, K., & Priesemann, V. (2020). Control of criticality and computation in spiking neuromorphic networks with plasticity. Nature Communications, 11, 2853. https://doi.org/10.1038/s41467-020-16548-3de Carvalho, J. X., & Prado, C. P. C. (2000). Self-organized criticality in the Olami-Feder-Christensen model. Physical Review Letters, 84(17), 4006–4009. https://doi.org/10.1103/PhysRevLett.84.4006Derrida, B., & Pomeau, Y. (1986). Random networks of automata: A simple annealed approximation. Europhysics Letters, 1(2), 45–49. https://doi.org/10.1209/0295-5075/1/2/001Hagemann, A., Wilting, J., Samimizad, B., Mormann, F., & Priesemann, V. (2021). Assessing criticality in pre-seizure single-neuron activity of human epileptic cortex. PLOS Computational Biology, 17(3), e1008773. https://doi.org/10.1371/journal.pcbi.1008773Haldeman, C., & Beggs, J. M. (2005). Critical branching captures activity in living neural networks and maximizes the number of metastable states. Physical Review Letters, 94(5), 058101. https://doi.org/10.1103/PhysRevLett.94.058101Hsu, D., Chen, W., Hsu, M., & Beggs, J. M. (2008). An open hypothesis: Is epilepsy learned, and can it be unlearned? Epilepsy & Behavior, 13(3), 511–522. https://doi.org/10.1016/j.yebeh.2008.05.007Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42(1–3), 12–37. https://doi.org/10.1016/0167-2789(90)90064-VLevina, A., Herrmann, J. M., & Geisel, T. (2007). Dynamical synapses causing self-organized criticality in neural networks. Nature Physics, 3(12), 857–860. https://doi.org/10.1038/nphys758Magnasco, M. O. (2022). Robustness and flexibility of neural function through dynamical criticality. Entropy, 24(5), 591. https://doi.org/10.3390/e24050591Meisel, C., Klaus, A., Vyazovskiy, V. V., & Plenz, D. (2017). The interplay between long- and short-range temporal correlations shapes cortex dynamics across vigilance states. The Journal of Neuroscience, 37(42), 10114–10124. https://doi.org/10.1523/JNEUROSCI.0448-17.2017Morales, G. B., di Santo, S., & Muñoz, M. A. (2023). Unveiling the intrinsic dynamics of biological and artificial neural networks: From criticality to optimal representations. Frontiers in Complex Systems, 1, 1276338. https://doi.org/10.3389/fcpxs.2023.1276338Plenz, D., Ribeiro, T. L., Miller, S. R., Kells, P. A., Vakili, A., & Capek, E. L. (2021). Self-organized criticality in the brain. Frontiers in Physics, 9, 639389. https://doi.org/10.3389/fphy.2021.639389Shew, W. L., Yang, H., Petermann, T., Roy, R., & Plenz, D. (2009). Neuronal avalanches imply maximum dynamic range in cortical networks at criticality. The Journal of Neuroscience, 29(49), 15595–15600. https://doi.org/10.1523/JNEUROSCI.3864-09.2009Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information capacity and transmission are maximized in balanced cortical networks with neuronal avalanches. The Journal of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011Spitzner, F. P., Dehning, J., Wilting, J., Hagemann, A., Neto, J. P., Zierenberg, J., & Priesemann, V. (2021). MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity. PLOS ONE, 16(4), e0249447. https://doi.org/10.1371/journal.pone.0249447Timme, N. M., Marshall, N. J., Bennett, N., Ripp, M., Lautzenhiser, E., & Beggs, J. M. (2016). Criticality maximizes complexity in neural tissue. Frontiers in Physiology, 7, 425. https://doi.org/10.3389/fphys.2016.00425Turrigiano, G. G. (2008). The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell, 135(3), 422–435. https://doi.org/10.1016/j.cell.2008.10.008Wang, J., Cao, R., Brunton, B. W., Smith, R. E. W., Buckner, R. L., & Liu, T. T. (2025). Genetic contributions to brain criticality and its relationship with human cognitive functions. Proceedings of the National Academy of Sciences, 122(26), e2417010122. https://doi.org/10.1073/pnas.2417010122Wilting, J., Dehning, J., Pinheiro Neto, J., Rudelt, L., Wibral, M., Zierenberg, J., & Priesemann, V. (2018). Operating in a reverberating regime enables rapid tuning of network states to task requirements. Frontiers in Systems Neuroscience, 12, 55. https://doi.org/10.3389/fnsys.2018.00055Wilting, J., & Priesemann, V. (2018). Inferring collective dynamical states from widely unobserved systems. Nature Communications, 9, 2325. https://doi.org/10.1038/s41467-018-04725-4Wilting, J., & Priesemann, V. (2019). 25 years of criticality in neuroscience — Established results, open controversies, novel concepts. Current Opinion in Neurobiology, 58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002Yu, C. (2022). Toward a unified analysis of the brain criticality hypothesis: Reviewing several available tools. Frontiers in Neural Circuits, 16, 911245. https://doi.org/10.3389/fncir.2022.911245Zeraati, R., Engel, T. A., & Levina, A. (2024). Estimating intrinsic timescales and criticality from neural recordings: Methods and pitfalls. Current Opinion in Neurobiology, 86, 102871. https://doi.org/10.1016/j.conb.2024.102871Zimmern, V. (2020). Why brain criticality is clinically relevant: A scoping review. Frontiers in Neural Circuits, 14, 54. https://doi.org/10.3389/fncir.2020.00054
Explore the Complete Neuraxon Intelligence Academy
This is Volume 8 of the #Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, #aigarth , and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
[NIA Vol. 1](https://www.binance.com/en/square/post/295315343732018): Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.[NIA Vol. 2](https://www.binance.com/en/square/post/295304276561778): Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.[NIA Vol. 3](https://www.binance.com/en/square/post/295306656801506): Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.[NIA Vol. 4](https://www.binance.com/en/square/post/295302152913618): Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.[NIA Vol. 5](https://www.binance.com/en/square/post/302913958960674): Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.[NIA Vol. 6](https://www.binance.com/en/square/post/310198879866145): Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.[NIA Vol. 7](https://www.binance.com/en/square/post/321350661453970): Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's approach to emergent complexity and self-organized criticality in decentralized AI.
Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
Neuraxon: Implementing Brain Criticality in Artificial Networks
Written by Qubic Scientific TeamBranching ratio and criticality in biological networks, in artificial networks, and as a bioinspired principle in Neuraxon
What do a snow avalanche, a forest fire, an earthquake, and the spontaneous activity of the cerebral cortex have in common?
They all share a frontier between order and chaos, what is called a critical state. In the brain, that edge is measured by a simple parameter: the branching ratio (σ or m). It would be something like the average ratio of neuronal "offspring" that each "parent" neuron activates. When σ ≈ 1, activity neither dies out nor explodes; it reverberates.
Beggs and Plenz (2003) recorded the spontaneous activity of the cerebral cortex in rats and found that the activity formed cascade-like patterns, the so-called neuronal avalanches, with a branching ratio close to 1. The brain seemed to live at a critical point. In humans, the branching ratio σ once again appears close to unity (Wang et al., 2025; Plenz et al., 2021; Wilting & Priesemann, 2019).
At the critical point, systems simultaneously exhibit maximal sensitivity to perturbations (responsiveness), maximal dynamic capacity (number of accessible states), maximal information transmission, and maximal complexity (Timme et al., 2016; Shew et al., 2009, 2011).
What Is the Branching Ratio and How Is It Measured?
Conceptually, the branching ratio is trivial: if at instant t there are A(t) active neurons and at t+1 there are A(t+1), then:
σ = ⟨ A(t+1) / A(t) ⟩
Three regimes follow from this (de Carvalho & Prado, 2000; Haldeman & Beggs, 2005):
Subcritical (σ < 1): activity decays; the system "forgets" the perturbation quickly. It is stable but poor in memory and not very expressive.Supercritical (σ > 1): activity explodes into cascades. This is the signature of pathological regimes such as epileptic seizures (Hsu et al., 2008; Hagemann et al., 2021).Critical (σ ≈ 1): each spike, on average, generates another spike. Activity reverberates, neuronal avalanches obey power laws, and the system maintains a structured memory of the input.
The beauty of σ is that it is a single number that summarizes the global dynamical regime. But measuring it is less trivial. When applied to in vivo cortical recordings, the measurement reveals that the cortex does not operate exactly at σ = 1, but slightly below, in a regime that the authors call reverberating (Wilting et al., 2018). The difference is important: being exactly at σ = 1 would be like pedaling a bicycle balanced on a tightrope; being slightly below allows for rapid adjustment to task demands without the risk of runaway explosion.
Criticality in Artificial Neural Networks: From the Edge of Chaos to Reservoir Computing
Bertschinger and Natschläger (2004) showed that random recurrent threshold networks reach their maximal computational capacity on temporal processing tasks precisely at the order–chaos transition.
Boedecker et al. (2012) extended the analysis to echo state networks within the reservoir computing paradigm, confirming that information transfer capacity and active memory are maximized at the edge of chaos.
Fig. 3. A spiking neuromorphic network with synaptic plasticity self-organizes toward criticality under low external input, exhibiting power-law avalanche size distributions — the hallmark of the critical state in both biological and artificial neural networks. Under higher input, the network shifts to a subcritical regime with truncated distributions. Reproduced from Cramer et al. (2020), Nature Communications, 11, 2853. CC BY 4.0.
In the language of artificial neural networks, the measurement parameter is called the spectral radius. When it exceeds 1, trajectories diverge exponentially (chaos); when it is well below 1, the network collapses to the fixed point and loses memory. The spectral radius close to 1 is, in this context, the formal equivalent of the biological σ ≈ 1 (Magnasco, 2022; Morales et al., 2023). In spiking neural networks, the branching ratio can be measured with methods almost identical to those used in neuronal cultures (Cramer et al., 2020; Zeraati et al., 2024).
Why Does Brain Criticality Maximize Neural Computation?
Operating close to σ ≈ 1 provides four advantages that are central to both the critical brain hypothesis and the design of brain-inspired AI systems:
Maximal dynamic range. Shew et al. (2009) showed that the range of input intensities the cortex can discriminate is maximal when the excitation–inhibition balance places the network at criticality.Maximized information capacity. The entropy of avalanche patterns and the mutual information between input and output peak at σ ≈ 1 (Shew et al., 2011).Optimal fading memory. In the critical regime, the perturbation is sustained just long enough to influence processing without contaminating the distant future; it is the sweet spot between stability and temporal integration (Boedecker et al., 2012).Complexity as a unifying measure. Timme et al. (2016) demonstrated that neural complexity is maximized exactly at the critical point, linking criticality with formal theories of consciousness and processing.
Fig. 4. Four computational advantages of operating near the critical branching ratio (σ ≈ 1). At criticality, neural networks achieve maximal dynamic range, maximized information capacity, optimal fading memory, and maximum complexity — properties that are central to both the critical brain hypothesis and brain-inspired AI design.
The Brain Does Not Always Operate at σ = 1
This does not imply that the brain always operates at σ = 1. Evidence rather suggests a slightly subcritical and modulable regime: during demanding tasks the network approaches criticality, during deep sleep it moves away, and pathological states (epilepsy, deep anesthesia, certain psychiatric conditions) are associated with measurable deviations from this operational range (Meisel et al., 2017; Zimmern, 2020). The branching ratio is becoming a dynamic biomarker of the functional state of the nervous system.
Why We Use the Branching Ratio in Neuraxon: Bioinspired AI Design at the Edge of Chaos
Neuraxon is a bioinspired system that adopts dynamical principles of the cortex as design constraints. The branching ratio is one of the most important, and we use it for four reasons:
As a Real-Time Operational Invariant for Neural Network Stability
In deep spiking or recurrent architectures, the dual risk of activity collapse (silent network, vanishing gradients) and runaway explosion (saturation, exploding gradients) is structural. Monitoring σ in real time gives us a single diagnostic scalar, independent of the concrete architecture, that indicates whether the system is alive in the computational sense.
As a Bioinspired Self-Regulation Target Through Self-Organized Criticality
The network self-organizes toward criticality without the need for centralized fine-tuning, replicating the principle of self-organized criticality (Bornholdt & Röhl, 2003; Levina et al., 2007). This drastically reduces sensitivity to hyperparameters and endows the system with robustness against distribution shifts. As we explored in NIA Volume 7 on artificial life and digital ecosystems, this is exactly how emergent complexity arises from local rules without centralized control.
Fig. 5. Neuraxon 3D network during active simulation, showing cascading activity across ternary-state neurons. Brightly active nodes (pink) propagate signals through excitatory (green) and inhibitory (pink) connections while other neurons remain at rest (gray), illustrating a reverberating regime near the critical branching ratio (σ ≈ 1). This balanced state — neither silent nor explosive — is what Neuraxon self-organizes toward using bioinspired criticality principles. Explore the interactive demo athuggingface.co/spaces/DavidVivancos/Neuraxon. Source: Qubic Scientific Team.
As a Bridge Between Neuroscientific Observation and AI Design
The branching ratio is one of the very few magnitudes that is measured with the same formalism in electrophysiology, fMRI, and artificial networks. This allows for testing bidirectional hypotheses: if an intervention improves biological criticality, we can ask whether the same intervention — translated into the artificial architecture — improves the model's computation, and vice versa. This principle is central to the neuromodulation framework and the astrocytic gating mechanisms we have developed in previous volumes of this academy.
As a Functional, Not Aesthetic, Criterion for Brain-Inspired AI
Criticality is an operational constraint with empirical consequences. Operating near the reverberating regime improves — as measured in our internal evaluations and submitted publications — generalization capacity, stability under input perturbations, representational richness, and the temporal coherence of reasoning. These effects qualitatively match those reported in both the biological (Cocchi et al., 2017) and artificial (Cramer et al., 2020; Morales et al., 2023) literature.
The Branching Ratio: From Statistical Physics to Brain-Inspired AI Architecture
The branching ratio is one of those conceptual rara avis: simple enough to reduce to a single formula, deep enough to bridge statistical physics, neuroscience, AI, and systems design. For the biological brain, σ ≈ 1 seems to be the regime where the virtuous combination of sensitivity, memory, expressiveness, and robustness emerges. For artificial networks, the same frontier — rebranded as the edge of chaos — predicts maximal computational capacity.
And for Neuraxon, it is a guiding principle of bioinspired design: an auditable, self-regulating, and biologically meaningful metric that helps us keep the system alive, in the richest sense of the word.
References
Beggs, J. M., & Plenz, D. (2003). Neuronal avalanches in neocortical circuits. The Journal of Neuroscience, 23(35), 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003Bertschinger, N., & Natschläger, T. (2004). Real-time computation at the edge of chaos in recurrent neural networks. Neural Computation, 16(7), 1413–1436. https://doi.org/10.1162/089976604323057443Boedecker, J., Obst, O., Lizier, J. T., Mayer, N. M., & Asada, M. (2012). Information processing in echo state networks at the edge of chaos. Theory in Biosciences, 131(3), 205–213. https://doi.org/10.1007/s12064-011-0146-8Bornholdt, S., & Röhl, T. (2003). Self-organized critical neural networks. Physical Review E, 67(6), 066118. https://doi.org/10.1103/PhysRevE.67.066118Cocchi, L., Gollo, L. L., Zalesky, A., & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. https://doi.org/10.1016/j.pneurobio.2017.07.002Cramer, B., Stöckel, D., Kreft, M., Wibral, M., Schemmel, J., Meier, K., & Priesemann, V. (2020). Control of criticality and computation in spiking neuromorphic networks with plasticity. Nature Communications, 11, 2853. https://doi.org/10.1038/s41467-020-16548-3de Carvalho, J. X., & Prado, C. P. C. (2000). Self-organized criticality in the Olami-Feder-Christensen model. Physical Review Letters, 84(17), 4006–4009. https://doi.org/10.1103/PhysRevLett.84.4006Derrida, B., & Pomeau, Y. (1986). Random networks of automata: A simple annealed approximation. Europhysics Letters, 1(2), 45–49. https://doi.org/10.1209/0295-5075/1/2/001Hagemann, A., Wilting, J., Samimizad, B., Mormann, F., & Priesemann, V. (2021). Assessing criticality in pre-seizure single-neuron activity of human epileptic cortex. PLOS Computational Biology, 17(3), e1008773. https://doi.org/10.1371/journal.pcbi.1008773Haldeman, C., & Beggs, J. M. (2005). Critical branching captures activity in living neural networks and maximizes the number of metastable states. Physical Review Letters, 94(5), 058101. https://doi.org/10.1103/PhysRevLett.94.058101Hsu, D., Chen, W., Hsu, M., & Beggs, J. M. (2008). An open hypothesis: Is epilepsy learned, and can it be unlearned? Epilepsy & Behavior, 13(3), 511–522. https://doi.org/10.1016/j.yebeh.2008.05.007Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42(1–3), 12–37. https://doi.org/10.1016/0167-2789(90)90064-VLevina, A., Herrmann, J. M., & Geisel, T. (2007). Dynamical synapses causing self-organized criticality in neural networks. Nature Physics, 3(12), 857–860. https://doi.org/10.1038/nphys758Magnasco, M. O. (2022). Robustness and flexibility of neural function through dynamical criticality. Entropy, 24(5), 591. https://doi.org/10.3390/e24050591Meisel, C., Klaus, A., Vyazovskiy, V. V., & Plenz, D. (2017). The interplay between long- and short-range temporal correlations shapes cortex dynamics across vigilance states. The Journal of Neuroscience, 37(42), 10114–10124. https://doi.org/10.1523/JNEUROSCI.0448-17.2017Morales, G. B., di Santo, S., & Muñoz, M. A. (2023). Unveiling the intrinsic dynamics of biological and artificial neural networks: From criticality to optimal representations. Frontiers in Complex Systems, 1, 1276338. https://doi.org/10.3389/fcpxs.2023.1276338Plenz, D., Ribeiro, T. L., Miller, S. R., Kells, P. A., Vakili, A., & Capek, E. L. (2021). Self-organized criticality in the brain. Frontiers in Physics, 9, 639389. https://doi.org/10.3389/fphy.2021.639389Shew, W. L., Yang, H., Petermann, T., Roy, R., & Plenz, D. (2009). Neuronal avalanches imply maximum dynamic range in cortical networks at criticality. The Journal of Neuroscience, 29(49), 15595–15600. https://doi.org/10.1523/JNEUROSCI.3864-09.2009Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information capacity and transmission are maximized in balanced cortical networks with neuronal avalanches. The Journal of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011Spitzner, F. P., Dehning, J., Wilting, J., Hagemann, A., Neto, J. P., Zierenberg, J., & Priesemann, V. (2021). MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity. PLOS ONE, 16(4), e0249447. https://doi.org/10.1371/journal.pone.0249447Timme, N. M., Marshall, N. J., Bennett, N., Ripp, M., Lautzenhiser, E., & Beggs, J. M. (2016). Criticality maximizes complexity in neural tissue. Frontiers in Physiology, 7, 425. https://doi.org/10.3389/fphys.2016.00425Turrigiano, G. G. (2008). The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell, 135(3), 422–435. https://doi.org/10.1016/j.cell.2008.10.008Wang, J., Cao, R., Brunton, B. W., Smith, R. E. W., Buckner, R. L., & Liu, T. T. (2025). Genetic contributions to brain criticality and its relationship with human cognitive functions. Proceedings of the National Academy of Sciences, 122(26), e2417010122. https://doi.org/10.1073/pnas.2417010122Wilting, J., Dehning, J., Pinheiro Neto, J., Rudelt, L., Wibral, M., Zierenberg, J., & Priesemann, V. (2018). Operating in a reverberating regime enables rapid tuning of network states to task requirements. Frontiers in Systems Neuroscience, 12, 55. https://doi.org/10.3389/fnsys.2018.00055Wilting, J., & Priesemann, V. (2018). Inferring collective dynamical states from widely unobserved systems. Nature Communications, 9, 2325. https://doi.org/10.1038/s41467-018-04725-4Wilting, J., & Priesemann, V. (2019). 25 years of criticality in neuroscience — Established results, open controversies, novel concepts. Current Opinion in Neurobiology, 58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002Yu, C. (2022). Toward a unified analysis of the brain criticality hypothesis: Reviewing several available tools. Frontiers in Neural Circuits, 16, 911245. https://doi.org/10.3389/fncir.2022.911245Zeraati, R., Engel, T. A., & Levina, A. (2024). Estimating intrinsic timescales and criticality from neural recordings: Methods and pitfalls. Current Opinion in Neurobiology, 86, 102871. https://doi.org/10.1016/j.conb.2024.102871Zimmern, V. (2020). Why brain criticality is clinically relevant: A scoping review. Frontiers in Neural Circuits, 14, 54. https://doi.org/10.3389/fncir.2020.00054
Explore the Complete Neuraxon Intelligence Academy
This is Volume 8 of the #Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, #aigarth , and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Vol. 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Vol. 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Vol. 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Vol. 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Vol. 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Vol. 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Vol. 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's approach to emergent complexity and self-organized criticality in decentralized AI.
Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
What do a snow avalanche, a forest fire, an earthquake, and the spontaneous activity of the cerebral cortex have in common?
They all share a frontier between order and chaos, what is called a critical state. In the brain, that edge is measured by a simple parameter: the branching ratio (σ or m). It would be something like the average ratio of neuronal "offspring" that each "parent" neuron activates. When σ ≈ 1, activity neither dies out nor explodes; it reverberates.
Beggs and Plenz (2003) recorded the spontaneous activity of the cerebral cortex in rats and found that the activity formed cascade-like patterns, the so-called neuronal avalanches, with a branching ratio close to 1. The brain seemed to live at a critical point. In humans, the branching ratio σ once again appears close to unity (Wang et al., 2025; Plenz et al., 2021; Wilting & Priesemann, 2019).
At the critical point, systems simultaneously exhibit maximal sensitivity to perturbations (responsiveness), maximal dynamic capacity (number of accessible states), maximal information transmission, and maximal complexity (Timme et al., 2016; Shew et al., 2009, 2011).
What Is the Branching Ratio and How Is It Measured?
Conceptually, the branching ratio is trivial: if at instant t there are A(t) active neurons and at t+1 there are A(t+1), then:
σ = ⟨ A(t+1) / A(t) ⟩
Three regimes follow from this (de Carvalho & Prado, 2000; Haldeman & Beggs, 2005):
Subcritical (σ < 1): activity decays; the system "forgets" the perturbation quickly. It is stable but poor in memory and not very expressive.Supercritical (σ > 1): activity explodes into cascades. This is the signature of pathological regimes such as epileptic seizures (Hsu et al., 2008; Hagemann et al., 2021).Critical (σ ≈ 1): each spike, on average, generates another spike. Activity reverberates, neuronal avalanches obey power laws, and the system maintains a structured memory of the input.
The beauty of σ is that it is a single number that summarizes the global dynamical regime. But measuring it is less trivial. When applied to in vivo cortical recordings, the measurement reveals that the cortex does not operate exactly at σ = 1, but slightly below, in a regime that the authors call reverberating (Wilting et al., 2018). The difference is important: being exactly at σ = 1 would be like pedaling a bicycle balanced on a tightrope; being slightly below allows for rapid adjustment to task demands without the risk of runaway explosion.
Criticality in Artificial Neural Networks: From the Edge of Chaos to Reservoir Computing
Bertschinger and Natschläger (2004) showed that random recurrent threshold networks reach their maximal computational capacity on temporal processing tasks precisely at the order–chaos transition.
Boedecker et al. (2012) extended the analysis to echo state networks within the reservoir computing paradigm, confirming that information transfer capacity and active memory are maximized at the edge of chaos.
Fig. 3. A spiking neuromorphic network with synaptic plasticity self-organizes toward criticality under low external input, exhibiting power-law avalanche size distributions — the hallmark of the critical state in both biological and artificial neural networks. Under higher input, the network shifts to a subcritical regime with truncated distributions. Reproduced from Cramer et al. (2020), Nature Communications, 11, 2853. CC BY 4.0.
In the language of artificial neural networks, the measurement parameter is called the spectral radius. When it exceeds 1, trajectories diverge exponentially (chaos); when it is well below 1, the network collapses to the fixed point and loses memory. The spectral radius close to 1 is, in this context, the formal equivalent of the biological σ ≈ 1 (Magnasco, 2022; Morales et al., 2023). In spiking neural networks, the branching ratio can be measured with methods almost identical to those used in neuronal cultures (Cramer et al., 2020; Zeraati et al., 2024).
Why Does Brain Criticality Maximize Neural Computation?
Operating close to σ ≈ 1 provides four advantages that are central to both the critical brain hypothesis and the design of brain-inspired AI systems:
Maximal dynamic range. Shew et al. (2009) showed that the range of input intensities the cortex can discriminate is maximal when the excitation–inhibition balance places the network at criticality.Maximized information capacity. The entropy of avalanche patterns and the mutual information between input and output peak at σ ≈ 1 (Shew et al., 2011).Optimal fading memory. In the critical regime, the perturbation is sustained just long enough to influence processing without contaminating the distant future; it is the sweet spot between stability and temporal integration (Boedecker et al., 2012).Complexity as a unifying measure. Timme et al. (2016) demonstrated that neural complexity is maximized exactly at the critical point, linking criticality with formal theories of consciousness and processing.
Fig. 4. Four computational advantages of operating near the critical branching ratio (σ ≈ 1). At criticality, neural networks achieve maximal dynamic range, maximized information capacity, optimal fading memory, and maximum complexity — properties that are central to both the critical brain hypothesis and brain-inspired AI design.
The Brain Does Not Always Operate at σ = 1
This does not imply that the brain always operates at σ = 1. Evidence rather suggests a slightly subcritical and modulable regime: during demanding tasks the network approaches criticality, during deep sleep it moves away, and pathological states (epilepsy, deep anesthesia, certain psychiatric conditions) are associated with measurable deviations from this operational range (Meisel et al., 2017; Zimmern, 2020). The branching ratio is becoming a dynamic biomarker of the functional state of the nervous system.
Why We Use the Branching Ratio in Neuraxon: Bioinspired AI Design at the Edge of Chaos
Neuraxon is a bioinspired system that adopts dynamical principles of the cortex as design constraints. The branching ratio is one of the most important, and we use it for four reasons:
As a Real-Time Operational Invariant for Neural Network Stability
In deep spiking or recurrent architectures, the dual risk of activity collapse (silent network, vanishing gradients) and runaway explosion (saturation, exploding gradients) is structural. Monitoring σ in real time gives us a single diagnostic scalar, independent of the concrete architecture, that indicates whether the system is alive in the computational sense.
As a Bioinspired Self-Regulation Target Through Self-Organized Criticality
The network self-organizes toward criticality without the need for centralized fine-tuning, replicating the principle of self-organized criticality (Bornholdt & Röhl, 2003; Levina et al., 2007). This drastically reduces sensitivity to hyperparameters and endows the system with robustness against distribution shifts. As we explored in NIA Volume 7 on artificial life and digital ecosystems, this is exactly how emergent complexity arises from local rules without centralized control.
Fig. 5. Neuraxon 3D network during active simulation, showing cascading activity across ternary-state neurons. Brightly active nodes (pink) propagate signals through excitatory (green) and inhibitory (pink) connections while other neurons remain at rest (gray), illustrating a reverberating regime near the critical branching ratio (σ ≈ 1). This balanced state — neither silent nor explosive — is what Neuraxon self-organizes toward using bioinspired criticality principles. Explore the interactive demo athuggingface.co/spaces/DavidVivancos/Neuraxon. Source: Qubic Scientific Team.
As a Bridge Between Neuroscientific Observation and AI Design
The branching ratio is one of the very few magnitudes that is measured with the same formalism in electrophysiology, fMRI, and artificial networks. This allows for testing bidirectional hypotheses: if an intervention improves biological criticality, we can ask whether the same intervention — translated into the artificial architecture — improves the model's computation, and vice versa. This principle is central to the neuromodulation framework and the astrocytic gating mechanisms we have developed in previous volumes of this academy.
As a Functional, Not Aesthetic, Criterion for Brain-Inspired AI
Criticality is an operational constraint with empirical consequences. Operating near the reverberating regime improves — as measured in our internal evaluations and submitted publications — generalization capacity, stability under input perturbations, representational richness, and the temporal coherence of reasoning. These effects qualitatively match those reported in both the biological (Cocchi et al., 2017) and artificial (Cramer et al., 2020; Morales et al., 2023) literature.
The Branching Ratio: From Statistical Physics to Brain-Inspired AI Architecture
The branching ratio is one of those conceptual rara avis: simple enough to reduce to a single formula, deep enough to bridge statistical physics, neuroscience, AI, and systems design. For the biological brain, σ ≈ 1 seems to be the regime where the virtuous combination of sensitivity, memory, expressiveness, and robustness emerges. For artificial networks, the same frontier — rebranded as the edge of chaos — predicts maximal computational capacity.
And for Neuraxon, it is a guiding principle of bioinspired design: an auditable, self-regulating, and biologically meaningful metric that helps us keep the system alive, in the richest sense of the word.
References
Beggs, J. M., & Plenz, D. (2003). Neuronal avalanches in neocortical circuits. The Journal of Neuroscience, 23(35), 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003Bertschinger, N., & Natschläger, T. (2004). Real-time computation at the edge of chaos in recurrent neural networks. Neural Computation, 16(7), 1413–1436. https://doi.org/10.1162/089976604323057443Boedecker, J., Obst, O., Lizier, J. T., Mayer, N. M., & Asada, M. (2012). Information processing in echo state networks at the edge of chaos. Theory in Biosciences, 131(3), 205–213. https://doi.org/10.1007/s12064-011-0146-8Bornholdt, S., & Röhl, T. (2003). Self-organized critical neural networks. Physical Review E, 67(6), 066118. https://doi.org/10.1103/PhysRevE.67.066118Cocchi, L., Gollo, L. L., Zalesky, A., & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. https://doi.org/10.1016/j.pneurobio.2017.07.002Cramer, B., Stöckel, D., Kreft, M., Wibral, M., Schemmel, J., Meier, K., & Priesemann, V. (2020). Control of criticality and computation in spiking neuromorphic networks with plasticity. Nature Communications, 11, 2853. https://doi.org/10.1038/s41467-020-16548-3de Carvalho, J. X., & Prado, C. P. C. (2000). Self-organized criticality in the Olami-Feder-Christensen model. Physical Review Letters, 84(17), 4006–4009. https://doi.org/10.1103/PhysRevLett.84.4006Derrida, B., & Pomeau, Y. (1986). Random networks of automata: A simple annealed approximation. Europhysics Letters, 1(2), 45–49. https://doi.org/10.1209/0295-5075/1/2/001Hagemann, A., Wilting, J., Samimizad, B., Mormann, F., & Priesemann, V. (2021). Assessing criticality in pre-seizure single-neuron activity of human epileptic cortex. PLOS Computational Biology, 17(3), e1008773. https://doi.org/10.1371/journal.pcbi.1008773Haldeman, C., & Beggs, J. M. (2005). Critical branching captures activity in living neural networks and maximizes the number of metastable states. Physical Review Letters, 94(5), 058101. https://doi.org/10.1103/PhysRevLett.94.058101Hsu, D., Chen, W., Hsu, M., & Beggs, J. M. (2008). An open hypothesis: Is epilepsy learned, and can it be unlearned? Epilepsy & Behavior, 13(3), 511–522. https://doi.org/10.1016/j.yebeh.2008.05.007Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42(1–3), 12–37. https://doi.org/10.1016/0167-2789(90)90064-VLevina, A., Herrmann, J. M., & Geisel, T. (2007). Dynamical synapses causing self-organized criticality in neural networks. Nature Physics, 3(12), 857–860. https://doi.org/10.1038/nphys758Magnasco, M. O. (2022). Robustness and flexibility of neural function through dynamical criticality. Entropy, 24(5), 591. https://doi.org/10.3390/e24050591Meisel, C., Klaus, A., Vyazovskiy, V. V., & Plenz, D. (2017). The interplay between long- and short-range temporal correlations shapes cortex dynamics across vigilance states. The Journal of Neuroscience, 37(42), 10114–10124. https://doi.org/10.1523/JNEUROSCI.0448-17.2017Morales, G. B., di Santo, S., & Muñoz, M. A. (2023). Unveiling the intrinsic dynamics of biological and artificial neural networks: From criticality to optimal representations. Frontiers in Complex Systems, 1, 1276338. https://doi.org/10.3389/fcpxs.2023.1276338Plenz, D., Ribeiro, T. L., Miller, S. R., Kells, P. A., Vakili, A., & Capek, E. L. (2021). Self-organized criticality in the brain. Frontiers in Physics, 9, 639389. https://doi.org/10.3389/fphy.2021.639389Shew, W. L., Yang, H., Petermann, T., Roy, R., & Plenz, D. (2009). Neuronal avalanches imply maximum dynamic range in cortical networks at criticality. The Journal of Neuroscience, 29(49), 15595–15600. https://doi.org/10.1523/JNEUROSCI.3864-09.2009Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information capacity and transmission are maximized in balanced cortical networks with neuronal avalanches. The Journal of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011Spitzner, F. P., Dehning, J., Wilting, J., Hagemann, A., Neto, J. P., Zierenberg, J., & Priesemann, V. (2021). MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity. PLOS ONE, 16(4), e0249447. https://doi.org/10.1371/journal.pone.0249447Timme, N. M., Marshall, N. J., Bennett, N., Ripp, M., Lautzenhiser, E., & Beggs, J. M. (2016). Criticality maximizes complexity in neural tissue. Frontiers in Physiology, 7, 425. https://doi.org/10.3389/fphys.2016.00425Turrigiano, G. G. (2008). The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell, 135(3), 422–435. https://doi.org/10.1016/j.cell.2008.10.008Wang, J., Cao, R., Brunton, B. W., Smith, R. E. W., Buckner, R. L., & Liu, T. T. (2025). Genetic contributions to brain criticality and its relationship with human cognitive functions. Proceedings of the National Academy of Sciences, 122(26), e2417010122. https://doi.org/10.1073/pnas.2417010122Wilting, J., Dehning, J., Pinheiro Neto, J., Rudelt, L., Wibral, M., Zierenberg, J., & Priesemann, V. (2018). Operating in a reverberating regime enables rapid tuning of network states to task requirements. Frontiers in Systems Neuroscience, 12, 55. https://doi.org/10.3389/fnsys.2018.00055Wilting, J., & Priesemann, V. (2018). Inferring collective dynamical states from widely unobserved systems. Nature Communications, 9, 2325. https://doi.org/10.1038/s41467-018-04725-4Wilting, J., & Priesemann, V. (2019). 25 years of criticality in neuroscience — Established results, open controversies, novel concepts. Current Opinion in Neurobiology, 58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002Yu, C. (2022). Toward a unified analysis of the brain criticality hypothesis: Reviewing several available tools. Frontiers in Neural Circuits, 16, 911245. https://doi.org/10.3389/fncir.2022.911245Zeraati, R., Engel, T. A., & Levina, A. (2024). Estimating intrinsic timescales and criticality from neural recordings: Methods and pitfalls. Current Opinion in Neurobiology, 86, 102871. https://doi.org/10.1016/j.conb.2024.102871Zimmern, V. (2020). Why brain criticality is clinically relevant: A scoping review. Frontiers in Neural Circuits, 14, 54. https://doi.org/10.3389/fncir.2020.00054
Explore the Complete Neuraxon Intelligence Academy
This is Volume 8 of the #Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind Neuraxon, #aigarth , and Qubic's approach to brain-inspired, #decentralized artificial intelligence:
NIA Vol. 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Vol. 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Vol. 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain's chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon's architecture.NIA Vol. 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon's third-path approach.NIA Vol. 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Vol. 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.NIA Vol. 7: Conway's Game of Life, Artificial Life, and Digital Ecosystems — The science behind Qubic, Aigarth, and Neuraxon's approach to emergent complexity and self-organized criticality in decentralized AI.
Qubic is a decentralized, open-source network for experimental technology. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
1
128
·
--
Článok
Digital Ecosystems, Conway’s Game of Life, and Why Emergent Complexity Matters for Decentralized AINeuraxon Intelligence Academy — Volume 7
By the Qubic Scientific Team
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
What no one expected was what emerged from those four lines of rules. Stable structures. Oscillators that pulse forever and gliders that travel across the grid. Cannons that fire gliders periodically. Constructions were complex enough that, eventually, someone would build a Turing machine inside the Game of Life. Inside Conway’s grid you can, in principle, run any computation that exists.
of Life to Artificial Life (Alife)
In the eighties, Christopher Langton and a group of researchers turned this idea into a discipline of its own: Artificial Life, or Alife. The proposal was simple. Biology has historically studied life as we know it, the carbon-based one, the one that emerged on this particular planet. But life is, perhaps, a more general phenomenon. If we can build artificial systems that show the properties we associate with the living, self-organisation, adaptation, evolution, reproduction, response to the environment, then we are studying life as it could be, not just as it happens to be.
Alife is not a search for digital pets. It is a science of fundamental dynamics. Its experimental tools are simulators where simple agents follow local rules, and where the researcher watches what emerges at the global scale.
Several findings have stayed as cornerstones. The first, already implicit in Conway, is that simple local rules can generate global complexity without anyone designing it. The second came from Langton himself: there is a critical regime, called the edge of chaos, where systems are neither rigidly ordered nor fully chaotic, and where almost everything interesting happens. Computation, learning, adaptation, all flourish in that thin band. Below it, the system freezes. Above it, it dissolves into noise.
A third finding, less famous but more uncomfortable, is that properties we usually associate with intention, like cooperation, specialisation, division of labour, can emerge in systems that have not been programmed to cooperate. They emerge as consequences of the dynamics, not as goals. This one is hard to digest for the self proclaimed superior species, because our intuition tells us that if we want X, we have to optimise for X. Alife shows, again and again, that this is not always true.
What Are Digital Ecosystems? From Cellular Automata to Multi-Agent Neural Systems
A digital ecosystem is the natural evolution of these artificial life ideas. Instead of a single rule shared by all cells, you have several agents, each with their own rules, sharing a common environment, competing or cooperating for resources, reproducing, and dying. The substrate may be a 2D grid as in Conway, a continuous fluid as in Lenia, a richer world with terrain and food as in Biomaker CA. The details vary. The principle does not.
What makes a digital ecosystem interesting is not the underlying technology, but what it lets you observe. Population dynamics. Boundaries that form between species. Niches that open and close. Strategies that appear, dominate for a while, are displaced, and come back. Cycles that look like those of real ecosystems, sometimes surprisingly so. And the question that runs underneath all of it: when can we say that something has emerged, that the system has discovered something we did not put into it.
The Digital Ecosystems interactive platform by Sakana AI, showing real-time parameter sliders, population timeline, checkpoint tray, and simulation canvas. Users can steer the ecosystem and branch into alternative futures from any saved state.
There is recent work worth looking at. The team at Sakana AI, for instance, has just released Digital Ecosystems, an interactive platform where five neural cellular automata species compete on a shared grid in real time and where you can move the parameters with sliders, save states, and explore divergent futures from a single checkpoint. It is the latest and most accessible link in a chain that goes back to Conway, and it is worth playing with for an afternoon, just to feel how these dynamics behave when you can actually touch them.
Why Artificial Life and Emergent Complexity Matter for Qubic, Aigarth, and Neuraxon
The temptation, when reading about Conway, Langton, Lenia, or Sakana, is to file all this away as elegant intellectual entertainment. It is not. It is the conceptual scaffolding our project stands on.
Qubic: Self-Organising Decentralized Infrastructure
Qubic is, at the infrastructure level, a decentralised network of thousands of nodes competing and cooperating to validate computations and earn rewards. Without the right local rules, that network either centralises or falls apart. With the right rules, it self-organises into a stable, productive ecosystem. The validity of Qubic’s design rests on principles that come, in part, from artificial life research: how do you reach global stability without a central authority, and how do you make competition produce something useful for everyone.
Aigarth: Evolutionary AI at the Edge of Chaos
Aigarth goes further. It is not just a network, it is an evolving tissue. Networks of artificial neurons that mutate, prune, generate offspring, reorganise their topology under adaptive pressure. There are local rules, fitness criteria, or evolutionary dynamics. This is artificial life applied to AI architectures. And as with everything in Alife, what emerges depends on the regime the system operates in. Too rigid, no exploration. Too chaotic, no stability. The edge of chaos is, here too, where the interesting things happen.
Neuraxon: Trinary States and Self-Organized Criticality in Brain-Inspired AI
Neuraxon, the basic unit Aigarth is built on, was designed with this in mind. The trinary state (-1, 0, +1) is not a quantisation trick to save bits, even though it does also cut compute cost. It is a structural decision. The neutral state is a buffer that allows smooth transitions, that prevents the system from oscillating violently between extremes, and gives time for slow synapses and neuromodulators to act. As we have discussed in earlier volumes of the Neuraxon Intelligence Academy, this is what lets the system navigate the edge of chaos without collapsing.
In our experiments with NxonLife, the simulator we built to watch Neuraxon networks evolve in Game-of-Life-inspired environments, we have measured exactly the properties Alife predicts. A branching ratio close to 1, the classical signature of self-organised criticality. Long-range temporal correlations following 1/f dynamics. Activity that sustains itself for thousands of ticks without external resets, without imposed normalisation, without anyone telling the system what to do. The networks find that regime by themselves, because the architecture has been built for it to be possible.
From Artificial Life Simulations to Decentralized AI Infrastructure: An Old Idea, a New Substrate
Growth-gate steepness sweep in Sakana AI's Digital Ecosystems. Lowering the gate steepness pushes species from rigid territorial boundaries into an excitable edge-of-chaos regime where emergent complexity and cooperation arise. Source: Sakana AI (2026)
What Conway showed in 1970, Langton in 1990, the Lenia team more recently, and Sakana AI a few weeks ago, is that complexity emerges from local rules and well-chosen parameters. What we are doing with Qubic, Aigarth and Neuraxon is taking that insight to its logical conclusion: not just observing simulated ecosystems, but building real distributed infrastructure on its principles.
The basic intuition does not change. Live systems live in time. They organise themselves between order and chaos. They cooperate without anyone instructing them to. They emerge, they do not design themselves.
Conway’s Game of Life was a postcard. Artificial life is a discipline. Digital ecosystems are a tool. Qubic, Aigarth and Neuraxon are an attempt to take all of this from the simulator and turn it into a working network. The ideas have been there for fifty years. The substrate to make them productive at scale is what we are building now.
References
Conway, J. H. (in Gardner, M.) (1970). Mathematical games: The fantastic combinations of John Conway’s new solitaire game “Life”. Scientific American, 223, 120–123. [Link]Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42, 12–37. [Link]Bedau, M. A. (2003). Artificial life: organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. [Link]Chan, B. W.-C. (2019). Lenia: Biology of artificial life. Complex Systems, 28(3), 251–286. [Link]Mordvintsev, A., Randazzo, E., Niklasson, E., & Levin, M. (2020). Growing neural cellular automata. Distill, 5(2), e23. [Link]Darlow, L. (2026). Digital Ecosystems: Interactive Multi-Agent Neural Cellular Automata. Sakana AI. [Link]Vivancos, D., & Sanchez, J. (2025). From Perceptrons to Neuraxons: A new neural growth and computation blueprint. Qubic Science. [Link]Vivancos, D., & Sanchez, J. (2025). Time-embedded trinary state dynamics learning architecture. Preprint. [Link]
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 7 of the Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind #Neuraxon , #aigarth , and Qubic’s approach to brain-inspired, #decentralized artificial intelligence:
[NIA Volume 1](https://www.binance.com/en/square/post/295315343732018): Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.[NIA Volume 2](https://www.binance.com/en/square/post/295304276561778): Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.[NIA Volume 3](https://www.binance.com/en/square/post/295306656801506): Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain’s chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon’s architecture.[NIA Volume 4](https://www.binance.com/en/square/post/295302152913618): Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon’s third-path approach.[NIA Volume 5](https://www.binance.com/en/square/post/302913958960674): Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.[NIA Volume 6](https://www.binance.com/en/square/post/310198879866145): Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
Digital Ecosystems, Conway’s Game of Life, and Why Emergent Complexity Matters for Decentralized AI
Neuraxon Intelligence Academy — Volume 7
By the Qubic Scientific Team
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
What no one expected was what emerged from those four lines of rules. Stable structures. Oscillators that pulse forever and gliders that travel across the grid. Cannons that fire gliders periodically. Constructions were complex enough that, eventually, someone would build a Turing machine inside the Game of Life. Inside Conway’s grid you can, in principle, run any computation that exists.
of Life to Artificial Life (Alife)
In the eighties, Christopher Langton and a group of researchers turned this idea into a discipline of its own: Artificial Life, or Alife. The proposal was simple. Biology has historically studied life as we know it, the carbon-based one, the one that emerged on this particular planet. But life is, perhaps, a more general phenomenon. If we can build artificial systems that show the properties we associate with the living, self-organisation, adaptation, evolution, reproduction, response to the environment, then we are studying life as it could be, not just as it happens to be.
Alife is not a search for digital pets. It is a science of fundamental dynamics. Its experimental tools are simulators where simple agents follow local rules, and where the researcher watches what emerges at the global scale.
Several findings have stayed as cornerstones. The first, already implicit in Conway, is that simple local rules can generate global complexity without anyone designing it. The second came from Langton himself: there is a critical regime, called the edge of chaos, where systems are neither rigidly ordered nor fully chaotic, and where almost everything interesting happens. Computation, learning, adaptation, all flourish in that thin band. Below it, the system freezes. Above it, it dissolves into noise.
A third finding, less famous but more uncomfortable, is that properties we usually associate with intention, like cooperation, specialisation, division of labour, can emerge in systems that have not been programmed to cooperate. They emerge as consequences of the dynamics, not as goals. This one is hard to digest for the self proclaimed superior species, because our intuition tells us that if we want X, we have to optimise for X. Alife shows, again and again, that this is not always true.
What Are Digital Ecosystems? From Cellular Automata to Multi-Agent Neural Systems
A digital ecosystem is the natural evolution of these artificial life ideas. Instead of a single rule shared by all cells, you have several agents, each with their own rules, sharing a common environment, competing or cooperating for resources, reproducing, and dying. The substrate may be a 2D grid as in Conway, a continuous fluid as in Lenia, a richer world with terrain and food as in Biomaker CA. The details vary. The principle does not.
What makes a digital ecosystem interesting is not the underlying technology, but what it lets you observe. Population dynamics. Boundaries that form between species. Niches that open and close. Strategies that appear, dominate for a while, are displaced, and come back. Cycles that look like those of real ecosystems, sometimes surprisingly so. And the question that runs underneath all of it: when can we say that something has emerged, that the system has discovered something we did not put into it.
The Digital Ecosystems interactive platform by Sakana AI, showing real-time parameter sliders, population timeline, checkpoint tray, and simulation canvas. Users can steer the ecosystem and branch into alternative futures from any saved state.
There is recent work worth looking at. The team at Sakana AI, for instance, has just released Digital Ecosystems, an interactive platform where five neural cellular automata species compete on a shared grid in real time and where you can move the parameters with sliders, save states, and explore divergent futures from a single checkpoint. It is the latest and most accessible link in a chain that goes back to Conway, and it is worth playing with for an afternoon, just to feel how these dynamics behave when you can actually touch them.
Why Artificial Life and Emergent Complexity Matter for Qubic, Aigarth, and Neuraxon
The temptation, when reading about Conway, Langton, Lenia, or Sakana, is to file all this away as elegant intellectual entertainment. It is not. It is the conceptual scaffolding our project stands on.
Qubic: Self-Organising Decentralized Infrastructure
Qubic is, at the infrastructure level, a decentralised network of thousands of nodes competing and cooperating to validate computations and earn rewards. Without the right local rules, that network either centralises or falls apart. With the right rules, it self-organises into a stable, productive ecosystem. The validity of Qubic’s design rests on principles that come, in part, from artificial life research: how do you reach global stability without a central authority, and how do you make competition produce something useful for everyone.
Aigarth: Evolutionary AI at the Edge of Chaos
Aigarth goes further. It is not just a network, it is an evolving tissue. Networks of artificial neurons that mutate, prune, generate offspring, reorganise their topology under adaptive pressure. There are local rules, fitness criteria, or evolutionary dynamics. This is artificial life applied to AI architectures. And as with everything in Alife, what emerges depends on the regime the system operates in. Too rigid, no exploration. Too chaotic, no stability. The edge of chaos is, here too, where the interesting things happen.
Neuraxon: Trinary States and Self-Organized Criticality in Brain-Inspired AI
Neuraxon, the basic unit Aigarth is built on, was designed with this in mind. The trinary state (-1, 0, +1) is not a quantisation trick to save bits, even though it does also cut compute cost. It is a structural decision. The neutral state is a buffer that allows smooth transitions, that prevents the system from oscillating violently between extremes, and gives time for slow synapses and neuromodulators to act. As we have discussed in earlier volumes of the Neuraxon Intelligence Academy, this is what lets the system navigate the edge of chaos without collapsing.
In our experiments with NxonLife, the simulator we built to watch Neuraxon networks evolve in Game-of-Life-inspired environments, we have measured exactly the properties Alife predicts. A branching ratio close to 1, the classical signature of self-organised criticality. Long-range temporal correlations following 1/f dynamics. Activity that sustains itself for thousands of ticks without external resets, without imposed normalisation, without anyone telling the system what to do. The networks find that regime by themselves, because the architecture has been built for it to be possible.
From Artificial Life Simulations to Decentralized AI Infrastructure: An Old Idea, a New Substrate
Growth-gate steepness sweep in Sakana AI's Digital Ecosystems. Lowering the gate steepness pushes species from rigid territorial boundaries into an excitable edge-of-chaos regime where emergent complexity and cooperation arise. Source: Sakana AI (2026)
What Conway showed in 1970, Langton in 1990, the Lenia team more recently, and Sakana AI a few weeks ago, is that complexity emerges from local rules and well-chosen parameters. What we are doing with Qubic, Aigarth and Neuraxon is taking that insight to its logical conclusion: not just observing simulated ecosystems, but building real distributed infrastructure on its principles.
The basic intuition does not change. Live systems live in time. They organise themselves between order and chaos. They cooperate without anyone instructing them to. They emerge, they do not design themselves.
Conway’s Game of Life was a postcard. Artificial life is a discipline. Digital ecosystems are a tool. Qubic, Aigarth and Neuraxon are an attempt to take all of this from the simulator and turn it into a working network. The ideas have been there for fifty years. The substrate to make them productive at scale is what we are building now.
References
Conway, J. H. (in Gardner, M.) (1970). Mathematical games: The fantastic combinations of John Conway’s new solitaire game “Life”. Scientific American, 223, 120–123. [Link]Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42, 12–37. [Link]Bedau, M. A. (2003). Artificial life: organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. [Link]Chan, B. W.-C. (2019). Lenia: Biology of artificial life. Complex Systems, 28(3), 251–286. [Link]Mordvintsev, A., Randazzo, E., Niklasson, E., & Levin, M. (2020). Growing neural cellular automata. Distill, 5(2), e23. [Link]Darlow, L. (2026). Digital Ecosystems: Interactive Multi-Agent Neural Cellular Automata. Sakana AI. [Link]Vivancos, D., & Sanchez, J. (2025). From Perceptrons to Neuraxons: A new neural growth and computation blueprint. Qubic Science. [Link]Vivancos, D., & Sanchez, J. (2025). Time-embedded trinary state dynamics learning architecture. Preprint. [Link]
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 7 of the Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind #Neuraxon , #aigarth , and Qubic’s approach to brain-inspired, #decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain’s chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon’s architecture.NIA Volume 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon’s third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
By the Qubic Scientific Team
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
In 1970, Martin Gardner published in Scientific American a recreational game invented by John Conway: the Game of Life. The rules fit on a postcard. A two-dimensional grid of cells in which each cell was alive or dead. At every step, a living cell stayed alive if it had two or three living neighbours, otherwise it died. A dead cell with exactly three living neighbours was born. Nothing else, as simple as that.
What no one expected was what emerged from those four lines of rules. Stable structures. Oscillators that pulse forever and gliders that travel across the grid. Cannons that fire gliders periodically. Constructions were complex enough that, eventually, someone would build a Turing machine inside the Game of Life. Inside Conway’s grid you can, in principle, run any computation that exists.
of Life to Artificial Life (Alife)
In the eighties, Christopher Langton and a group of researchers turned this idea into a discipline of its own: Artificial Life, or Alife. The proposal was simple. Biology has historically studied life as we know it, the carbon-based one, the one that emerged on this particular planet. But life is, perhaps, a more general phenomenon. If we can build artificial systems that show the properties we associate with the living, self-organisation, adaptation, evolution, reproduction, response to the environment, then we are studying life as it could be, not just as it happens to be.
Alife is not a search for digital pets. It is a science of fundamental dynamics. Its experimental tools are simulators where simple agents follow local rules, and where the researcher watches what emerges at the global scale.
Several findings have stayed as cornerstones. The first, already implicit in Conway, is that simple local rules can generate global complexity without anyone designing it. The second came from Langton himself: there is a critical regime, called the edge of chaos, where systems are neither rigidly ordered nor fully chaotic, and where almost everything interesting happens. Computation, learning, adaptation, all flourish in that thin band. Below it, the system freezes. Above it, it dissolves into noise.
A third finding, less famous but more uncomfortable, is that properties we usually associate with intention, like cooperation, specialisation, division of labour, can emerge in systems that have not been programmed to cooperate. They emerge as consequences of the dynamics, not as goals. This one is hard to digest for the self proclaimed superior species, because our intuition tells us that if we want X, we have to optimise for X. Alife shows, again and again, that this is not always true.
What Are Digital Ecosystems? From Cellular Automata to Multi-Agent Neural Systems
A digital ecosystem is the natural evolution of these artificial life ideas. Instead of a single rule shared by all cells, you have several agents, each with their own rules, sharing a common environment, competing or cooperating for resources, reproducing, and dying. The substrate may be a 2D grid as in Conway, a continuous fluid as in Lenia, a richer world with terrain and food as in Biomaker CA. The details vary. The principle does not.
What makes a digital ecosystem interesting is not the underlying technology, but what it lets you observe. Population dynamics. Boundaries that form between species. Niches that open and close. Strategies that appear, dominate for a while, are displaced, and come back. Cycles that look like those of real ecosystems, sometimes surprisingly so. And the question that runs underneath all of it: when can we say that something has emerged, that the system has discovered something we did not put into it.
The Digital Ecosystems interactive platform by Sakana AI, showing real-time parameter sliders, population timeline, checkpoint tray, and simulation canvas. Users can steer the ecosystem and branch into alternative futures from any saved state.
There is recent work worth looking at. The team at Sakana AI, for instance, has just released Digital Ecosystems, an interactive platform where five neural cellular automata species compete on a shared grid in real time and where you can move the parameters with sliders, save states, and explore divergent futures from a single checkpoint. It is the latest and most accessible link in a chain that goes back to Conway, and it is worth playing with for an afternoon, just to feel how these dynamics behave when you can actually touch them.
Why Artificial Life and Emergent Complexity Matter for Qubic, Aigarth, and Neuraxon
The temptation, when reading about Conway, Langton, Lenia, or Sakana, is to file all this away as elegant intellectual entertainment. It is not. It is the conceptual scaffolding our project stands on.
Qubic: Self-Organising Decentralized Infrastructure
Qubic is, at the infrastructure level, a decentralised network of thousands of nodes competing and cooperating to validate computations and earn rewards. Without the right local rules, that network either centralises or falls apart. With the right rules, it self-organises into a stable, productive ecosystem. The validity of Qubic’s design rests on principles that come, in part, from artificial life research: how do you reach global stability without a central authority, and how do you make competition produce something useful for everyone.
Aigarth: Evolutionary AI at the Edge of Chaos
Aigarth goes further. It is not just a network, it is an evolving tissue. Networks of artificial neurons that mutate, prune, generate offspring, reorganise their topology under adaptive pressure. There are local rules, fitness criteria, or evolutionary dynamics. This is artificial life applied to AI architectures. And as with everything in Alife, what emerges depends on the regime the system operates in. Too rigid, no exploration. Too chaotic, no stability. The edge of chaos is, here too, where the interesting things happen.
Neuraxon: Trinary States and Self-Organized Criticality in Brain-Inspired AI
Neuraxon, the basic unit Aigarth is built on, was designed with this in mind. The trinary state (-1, 0, +1) is not a quantisation trick to save bits, even though it does also cut compute cost. It is a structural decision. The neutral state is a buffer that allows smooth transitions, that prevents the system from oscillating violently between extremes, and gives time for slow synapses and neuromodulators to act. As we have discussed in earlier volumes of the Neuraxon Intelligence Academy, this is what lets the system navigate the edge of chaos without collapsing.
In our experiments with NxonLife, the simulator we built to watch Neuraxon networks evolve in Game-of-Life-inspired environments, we have measured exactly the properties Alife predicts. A branching ratio close to 1, the classical signature of self-organised criticality. Long-range temporal correlations following 1/f dynamics. Activity that sustains itself for thousands of ticks without external resets, without imposed normalisation, without anyone telling the system what to do. The networks find that regime by themselves, because the architecture has been built for it to be possible.
From Artificial Life Simulations to Decentralized AI Infrastructure: An Old Idea, a New Substrate
Growth-gate steepness sweep in Sakana AI's Digital Ecosystems. Lowering the gate steepness pushes species from rigid territorial boundaries into an excitable edge-of-chaos regime where emergent complexity and cooperation arise. Source: Sakana AI (2026)
What Conway showed in 1970, Langton in 1990, the Lenia team more recently, and Sakana AI a few weeks ago, is that complexity emerges from local rules and well-chosen parameters. What we are doing with Qubic, Aigarth and Neuraxon is taking that insight to its logical conclusion: not just observing simulated ecosystems, but building real distributed infrastructure on its principles.
The basic intuition does not change. Live systems live in time. They organise themselves between order and chaos. They cooperate without anyone instructing them to. They emerge, they do not design themselves.
Conway’s Game of Life was a postcard. Artificial life is a discipline. Digital ecosystems are a tool. Qubic, Aigarth and Neuraxon are an attempt to take all of this from the simulator and turn it into a working network. The ideas have been there for fifty years. The substrate to make them productive at scale is what we are building now.
References
Conway, J. H. (in Gardner, M.) (1970). Mathematical games: The fantastic combinations of John Conway’s new solitaire game “Life”. Scientific American, 223, 120–123. [Link]Langton, C. G. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42, 12–37. [Link]Bedau, M. A. (2003). Artificial life: organization, adaptation and complexity from the bottom up. Trends in Cognitive Sciences, 7(11), 505–512. [Link]Chan, B. W.-C. (2019). Lenia: Biology of artificial life. Complex Systems, 28(3), 251–286. [Link]Mordvintsev, A., Randazzo, E., Niklasson, E., & Levin, M. (2020). Growing neural cellular automata. Distill, 5(2), e23. [Link]Darlow, L. (2026). Digital Ecosystems: Interactive Multi-Agent Neural Cellular Automata. Sakana AI. [Link]Vivancos, D., & Sanchez, J. (2025). From Perceptrons to Neuraxons: A new neural growth and computation blueprint. Qubic Science. [Link]Vivancos, D., & Sanchez, J. (2025). Time-embedded trinary state dynamics learning architecture. Preprint. [Link]
Explore the Complete Neuraxon Intelligence Academy Series
This is Volume 7 of the Neuraxon Intelligence #academy by the #Qubic Scientific Team. If you are just joining us, explore the complete series to build a full understanding of the science behind #Neuraxon , #aigarth , and Qubic’s approach to brain-inspired, #decentralized artificial intelligence:
NIA Volume 1: Why Intelligence Is Not Computed in Steps, but in Time — Explores why biological intelligence operates in continuous time rather than discrete computational steps like traditional LLMs.NIA Volume 2: Ternary Dynamics as a Model of Living Intelligence — Explains ternary dynamics and why three-state logic (excitatory, neutral, inhibitory) matters for modeling living systems.NIA Volume 3: Neuromodulation and Brain-Inspired AI — Covers neuromodulation and how the brain’s chemical signaling (dopamine, serotonin, acetylcholine, norepinephrine) inspires Neuraxon’s architecture.NIA Volume 4: Neural Networks in AI and Neuroscience — A deep comparison of biological neural networks, artificial neural networks, and Neuraxon’s third-path approach.NIA Volume 5: Astrocytes and Brain-Inspired AI — How astrocytic gating transforms neural network plasticity through the AGMP framework in Neuraxon.NIA Volume 6: Conscious Machines vs Intelligent Organisms: AI Consciousness Explained — Explores AI consciousness through the lens of Global Workspace Theory, Integrated Information Theory, and predictive coding.
Qubic is a decentralized, open-source network. To learn more, visit qubic.org. Join the discussion on X, Discord, and Telegram.
1
1
121
Prihláste sa a preskúmajte ďalší obsah
Pripojte sa k používateľom kryptomien na celom svete na Binance Square
⚡️ Získajte najnovšie a užitočné informácie o kryptomenách.
💬 Dôvera najväčšej kryptoburzy na svete.
👍 Objavte skutočné poznatky od overených tvorcov.
E-mail/telefónne číslo
Populárne témy
1
#SpaceXIPOUSStocksOpenHigher
Počet zobrazení: 95,943
Diskutuje: 945
2
#OilSlidesOnMiddleEastPeaceDealProspects
Počet zobrazení: 49,464
Diskutuje: 641
3
#SpaceXSharesOpen29PercentAboveIPOPrice
Počet zobrazení: 93,925
Diskutuje: 706
4
#SBFAppealFails25YearSentenceUpheld
Počet zobrazení: 38,636
Diskutuje: 429
5
#IranDeniesSundayGenevaSigningDate
Počet zobrazení: 32,034
Diskutuje: 356
6
#USConsumerSentimentRisesEarlyJune
Počet zobrazení: 26,013
Diskutuje: 330
7
#TRUMPTokenMarketCapAbove$2.1B
Počet zobrazení: 9,690
Diskutuje: 123
8
#SpaceXNasdaqIPO$75BCryptoImpact
Počet zobrazení: 2,868
Diskutuje: 23
9
#SpaceXIPO$1.8TRecordDebut
Počet zobrazení: 8,490
Diskutuje: 48
10
#SPCXReboundsTo$183OnHyperliquid
Počet zobrazení: 6,658
Diskutuje: 59