Toward a theory of evolution as multilevel learning

'Snakes and ladders' in paleoanthropology: From cognitive surprise to skillfulness a million years ago

A paradigmatic account may suffice to explain behavioral evolution in early Homo. We propose a parsimonious account that (1) could explain a particular, frequently-encountered, archeological outcome of behavior in early Homo - namely, the fashioning of a Paleolithic stone 'handaxe' - from a biological theoretic perspective informed by the free energy principle (FEP); and that (2) regards instances of the outcome as postdictive or retrodictive, circumstantial corroboration. Our proposal considers humankind evolving as a self-organizing biological ecosystem at a geological time-scale. We offer a narrative treatment of this self-organization in terms of the FEP. Specifically, we indicate how 'cognitive surprises' could underwrite an evolving propensity in early Homo to express sporadic unorthodox or anomalous behavior. This co-evolutionary propensity has left us a legacy of Paleolithic artifacts that is reminiscent of a 'snakes and ladders' board game of appearances, disappearances, and reappearances of particular archeological traces of Paleolithic behavior. When detected in the Early and Middle Pleistocene record, anthropologists and archeologists often imagine evidence of unusual or novel behavior in terms of early humankind ascending the rungs of a figurative phylogenetic 'ladder' - as if these corresponded to progressive evolution of cognitive abilities that enabled incremental achievements of increasingly innovative technical prowess, culminating in the cognitive ascendancy of Homo sapiens. The conjecture overlooks a plausible likelihood that behavior by an individual who was atypical among her conspecifics could have been disregarded in a community of Hominina (for definition see Appendix 1) that failed to recognize, imagine, or articulate potential advantages of adopting hitherto unorthodox behavior. Such failure, as well as diverse fortuitous demographic accidents, would cause exceptional personal behavior to be ignored and hence unremembered. It could disappear by a pitfall, down a 'snake', as it were, in the figurative evolutionary board game; thereby causing a discontinuity in the evolution of human behavior that presents like an evolutionary puzzle. The puzzle discomforts some paleoanthropologists trained in the natural and life sciences. They often dismiss it, explaining it away with such self-justifying conjectures as that, maybe, separate paleospecies of Homo differentially possessed different cognitive abilities, which, supposedly, could account for the presence or absence in the Pleistocene archeological record of traces of this or that behavioral outcome or skill. We argue that an alternative perspective - that inherits from the FEP and an individual's 'active inference' about its surroundings and of its own responses - affords a prosaic, deflationary, and parsimonious way to account for appearances, disappearances, and reappearances of particular behavioral outcomes and skills of early humankind.

Long range segmentation of prokaryotic genomes by gene age and functionality

Bacterial and archaeal genomes encompass numerous operons that typically consist of two to five genes. On larger scales, however, gene order is poorly conserved through the evolution of prokaryotes. Nevertheless, non-random localization of different classes of genes on prokaryotic chromosomes could reflect important functional and evolutionary constraints. We explored the patterns of genomic localization of evolutionarily conserved (ancient) and variable (young) genes across the diversity of bacteria and archaea. Nearly all bacterial and archaeal chromosomes were found to encompass large segments of 100-300 kilobases that were significantly enriched in either ancient or young genes. Similar clustering of genes with lethal knockout phenotype (essential genes) was observed as well. Mathematical modeling of genome evolution suggests that this long-range gene clustering in prokaryotic chromosomes reflects perpetual genome rearrangement driven by a combination of selective and neutral processes rather than evolutionary conservation.

Neural representation in active inference: Using generative models to interact with-and understand-the lived world

This paper considers neural representation through the lens of active inference, a normative framework for understanding brain function. It delves into how living organisms employ generative models to minimize the discrepancy between predictions and observations (as scored with variational free energy). The ensuing analysis suggests that the brain learns generative models to navigate the world adaptively, not (or not solely) to understand it. Different living organisms may possess an array of generative models, spanning from those that support action-perception cycles to those that underwrite planning and imagination; namely, from explicit models that entail variables for predicting concurrent sensations, like objects, faces, or people-to action-oriented models that predict action outcomes. It then elucidates how generative models and belief dynamics might link to neural representation and the implications of different types of generative models for understanding an agent's cognitive capabilities in relation to its ecological niche. The paper concludes with open questions regarding the evolution of generative models and the development of advanced cognitive abilities-and the gradual transition from pragmatic to detached neural representations. The analysis on offer foregrounds the diverse roles that generative models play in cognitive processes and the evolution of neural representation.

Evolution at the Origins of Life?

The role of evolutionary theory at the origin of life is an extensively debated topic. The origin and early development of life is usually separated into a prebiotic phase and a protocellular phase, ultimately leading to the Last Universal Common Ancestor. Most likely, the Last Universal Common Ancestor was subject to Darwinian evolution, but the question remains to what extent Darwinian evolution applies to the prebiotic and protocellular phases. In this review, we reflect on the current status of evolutionary theory in origins of life research by bringing together philosophy of science, evolutionary biology, and empirical research in the origins field. We explore the various ways in which evolutionary theory has been extended beyond biology; we look at how these extensions apply to the prebiotic development of (proto)metabolism; and we investigate how the terminology from evolutionary theory is currently being employed in state-of-the-art origins of life research. In doing so, we identify some of the current obstacles to an evolutionary account of the origins of life, as well as open up new avenues of research.

A Variational Synthesis of Evolutionary and Developmental Dynamics

This paper introduces a variational formulation of natural selection, paying special attention to the nature of 'things' and the way that different 'kinds' of 'things' are individuated from-and influence-each other. We use the Bayesian mechanics of particular partitions to understand how slow phylogenetic processes constrain-and are constrained by-fast, phenotypic processes. The main result is a formulation of adaptive fitness as a path integral of phenotypic fitness. Paths of least action, at the phenotypic and phylogenetic scales, can then be read as inference and learning processes, respectively. In this view, a phenotype actively infers the state of its econiche under a generative model, whose parameters are learned via natural (Bayesian model) selection. The ensuing variational synthesis features some unexpected aspects. Perhaps the most notable is that it is not possible to describe or model a population of conspecifics per se. Rather, it is necessary to consider populations of distinct natural kinds that influence each other. This paper is limited to a description of the mathematical apparatus and accompanying ideas. Subsequent work will use these methods for simulations and numerical analyses-and identify points of contact with related mathematical formulations of evolution.

Evolution is driven by natural autoencoding: reframing species, interaction codes, cooperation and sexual reproduction

The continuity of life and its evolution, we proposed, emerge from an interactive group process manifested in networks of interaction. We term this process survival of the fitted. Here, we reason that survival of the fitted results from a natural computational process we term natural autoencoding. Natural autoencoding works by retaining repeating biological interactions while non-repeatable interactions disappear. (i) We define a species by its species interaction code, which consists of a compact description of the repeating interactions of species organisms with their external and internal environments. Species interaction codes are descriptions recorded in the biological infrastructure that enables repeating interactions. Encoding and decoding are interwoven. (ii) Evolution proceeds by natural autoencoding of sustained changes in species interaction codes. DNA is only one element in natural autoencoding. (iii) Natural autoencoding accounts for the paradox of genome randomization in sexual reproduction-recombined genomes are analogous to the diversified inputs required for artificial autoencoding. The increase in entropy generated by genome randomization compensates for the decrease in entropy generated by organized life. (iv) Natural autoencoding and artificial autoencoding algorithms manifest defined similarities and differences. Recognition of the importance of fittedness could well serve the future of a humanly livable biosphere.

To copy or not to copy? That is the question! From chimpanzees to the foundation of human technological culture

A prerequisite for copying innovative behaviour faithfully is the capacity of observers' brains, regarded as 'hierarchically mechanistic minds', to overcome cognitive 'surprisal' (see 2.), by maximising the evidence for their internal models, through active inference. Unlike modern humans, chimpanzees and other great apes show considerable limitations in their ability, or 'Zone of Bounded Surprisal', to overcome cognitive surprisal induced by innovative or unorthodox behaviour that rarely, therefore, is copied precisely or accurately. Most can copy adequately what is within their phenotypically habitual behavioural repertoire, in which technology plays scant part. Widespread intra- and intergenerational social transmission of complex technological innovations is not a hall-mark of great-ape taxa. 3 Ma, precursors of the genus Homo made stone artefacts, and stone-flaking likely was habitual before 2 Ma. After that time, early Homo erectus has left traces of technological innovations, though faithful copying of these and their intra- and intergenerational social transmission were rare before 1 Ma. This likely owed to a cerebral infrastructure of interconnected neuronal systems more limited than ours. Brains were smaller in size than ours, and cerebral neuronal systems ceased to develop when early Homo erectus attained full adult maturity by the mid-teen years, whereas its development continues until our mid-twenties nowadays. Pleistocene Homo underwent remarkable evolutionary adaptation of neurobiological propensities, and cerebral aspects are discussed that, it is proposed here, plausibly, were fundamental for faithful copying, which underpinned social transmission of technologies, cumulative learning, and culture. Here, observers' responses to an innovation are more important for ensuring its transmission than is an innovator's production of it, because, by themselves, the minimal cognitive prerequisites that are needed for encoding and assimilating innovations are insufficient for practical outcomes to accumulate and spread intra- and intergenerationally.

Learning by Population Genetics and Matrix Riccati Equation

A model of learning as a generalization of the Eigen's quasispecies model in population genetics is introduced. Eigen's model is considered as a matrix Riccati equation. The error catastrophe in the Eigen's model (when the purifying selection becomes ineffective) is discussed as the divergence of the Perron-Frobenius eigenvalue of the Riccati model in the limit of large matrices. A known estimate for the Perron-Frobenius eigenvalue provides an explanation for observed patterns of genomic evolution. We propose to consider the error catastrophe in Eigen's model as an analog of overfitting in learning theory; this gives a criterion for the presence of overfitting in learning.

Energy-harnessing problem solving of primordial life: Modeling the emergence of catalytic host-nested parasite life cycles

All life forms on earth ultimately descended from a primordial population dubbed the last universal common ancestor or LUCA via Darwinian evolution. Extant living systems share two salient functional features, a metabolism extracting and transforming energy required for survival, and an evolvable, informational polymer-the genome-conferring heredity. Genome replication invariably generates essential and ubiquitous genetic parasites. Here we model the energetic, replicative conditions of LUCA-like organisms and their parasites, as well as adaptive problem solving of host-parasite pairs. We show using an adapted Lotka-Volterra frame-work that three host-parasite pairs-individually a unit of a host and a parasite that is itself parasitized, therefore a nested parasite pair-are sufficient for robust and stable homeostasis, forming a life cycle. This nested parasitism model includes competition and habitat restriction. Its catalytic life cycle efficiently captures, channels and transforms energy, enabling dynamic host survival and adaptation. We propose a Malthusian fitness model for a quasispecies evolving through a host-nested parasite life cycle with two core features, rapid replacement of degenerate parasites and increasing evolutionary stability of host-nested parasite units from one to three pairs.

Resilience and active inference

In this article, we aim to conceptualize and formalize the construct of resilience using the tools of active inference, a new physics-based modeling approach apt for the description and analysis of complex adaptive systems. We intend this as a first step toward a computational model of resilient systems. We begin by offering a conceptual analysis of resilience, to clarify its meaning, as established in the literature. We examine an orthogonal, threefold distinction between meanings of the word "resilience": (i) inertia, or the ability to resist change (ii) elasticity, or the ability to bounce back from a perturbation, and (iii) plasticity, or the ability to flexibly expand the repertoire of adaptive states. We then situate all three senses of resilience within active inference. We map resilience as inertia onto high precision beliefs, resilience as elasticity onto relaxation back to characteristic (i.e., attracting) states, and resilience as plasticity onto functional redundancy and structural degeneracy.

Cognition as Morphological/Morphogenetic Embodied Computation In Vivo

Cognition, historically considered uniquely human capacity, has been recently found to be the ability of all living organisms, from single cells and up. This study approaches cognition from an info-computational stance, in which structures in nature are seen as information, and processes (information dynamics) are seen as computation, from the perspective of a cognizing agent. Cognition is understood as a network of concurrent morphological/morphogenetic computations unfolding as a result of self-assembly, self-organization, and autopoiesis of physical, chemical, and biological agents. The present-day human-centric view of cognition still prevailing in major encyclopedias has a variety of open problems. This article considers recent research about morphological computation, morphogenesis, agency, basal cognition, extended evolutionary synthesis, free energy principle, cognition as Bayesian learning, active inference, and related topics, offering new theoretical and practical perspectives on problems inherent to the old computationalist cognitive models which were based on abstract symbol processing, and unaware of actual physical constraints and affordances of the embodiment of cognizing agents. A better understanding of cognition is centrally important for future artificial intelligence, robotics, medicine, and related fields.

Overcoming the limits of natural computation in biological evolution toward the maximization of system efficiency

The goal-directedness of biological evolution is realized via the anticipatory achievement of the final state of the system that corresponds to the condition of its perfection in self-maintenance and in adaptability. In the course of individual development, a biological system maximizes its power via synergistic effects and becomes able to perform external work most efficiently. In this state, defined as stasis, robust self-maintaining configurations act as attractors resistant to external and internal perturbations. This corresponds to the local energy–time constraints that most efficiently fit the integral optimization of the whole system. In evolution, major evolutionary transitions that establish new states of stasis are achieved via codepoiesis, a process in which the undecided statements of existing coding systems form the basis for the evolutionary unfolding of the system by assigning new values to them. The genetic fixation of this macroevolutionary process leads to new programmes of individual development representing the process of natural computation. The phenomenon of complexification in evolution represents a metasystem transition that results in maximization of a system’s power and in the ability to increase external work performed by the system.

Yeast cell death pathway requiring AP-3 vesicle trafficking leads to vacuole/lysosome membrane permeabilization

Unicellular eukaryotes have been suggested as undergoing self-inflicted destruction. However, molecular details are sparse compared with the mechanisms of programmed/regulated cell death known for human cells and animal models. Here, we report a molecular cell death pathway in Saccharomyces cerevisiae leading to vacuole/lysosome membrane permeabilization. Following a transient cell death stimulus, yeast cells die slowly over several hours, consistent with an ongoing molecular dying process. A genome-wide screen for death-promoting factors identified all subunits of the AP-3 complex, a vesicle trafficking adapter known to transport and install newly synthesized proteins on the vacuole/lysosome membrane. To promote cell death, AP-3 requires its Arf1-GTPase-dependent vesicle trafficking function and the kinase Yck3, which is selectively transported to the vacuole membrane by AP-3. Video microscopy revealed a sequence of events where vacuole permeability precedes the loss of plasma membrane integrity. AP-3-dependent death appears to be conserved in the human pathogenic yeast Cryptococcus neoformans.

Details about how mammalian cells die have yielded effective cancer therapies. Similarly, details about fungal cell death may explain failed responses to anti-fungal agents and inform next-generation anti-fungal strategies. Stolp et al. describe a potential mechanism of yeast cell death subversion, by inhibiting AP-3 vesicle trafficking to block vacuole/lysosome permeability.

Design for an Individual: Connectionist Approaches to the Evolutionary Transitions in Individuality

The truly surprising thing about evolution is not how it makes individuals better adapted to their environment, but how it makes individuals. All individuals are made of parts that used to be individuals themselves, e.g., multicellular organisms from unicellular organisms. In such evolutionary transitions in individuality, the organised structure of relationships between component parts causes them to work together, creating a new organismic entity and a new evolutionary unit on which selection can act. However, the principles of these transitions remain poorly understood. In particular, the process of transition must be explained by “bottom-up” selection, i.e., on the existing lower-level evolutionary units, without presupposing the higher-level evolutionary unit we are trying to explain. In this hypothesis and theory manuscript we address the conditions for evolutionary transitions in individuality by exploiting adaptive principles already known in learning systems. Connectionist learning models, well-studied in neural networks, demonstrate how networks of organised functional relationships between components, sufficient to exhibit information integration and collective action, can be produced via fully-distributed and unsupervised learning principles, i.e., without centralised control or an external teacher. Evolutionary connectionism translates these distributed learning principles into the domain of natural selection, and suggests how relationships among evolutionary units could become adaptively organised by selection from below without presupposing genetic relatedness or selection on collectives. In this manuscript, we address how connectionist models with a particular interaction structure might explain transitions in individuality. We explore the relationship between the interaction structures necessary for (a) evolutionary individuality (where the evolution of the whole is a non-decomposable function of the evolution of the parts), (b) organismic individuality (where the development and behaviour of the whole is a non-decomposable function of the behaviour of component parts) and (c) non-linearly separable functions, familiar in connectionist models (where the output of the network is a non-decomposable function of the inputs). Specifically, we hypothesise that the conditions necessary to evolve a new level of individuality are described by the conditions necessary to learn non-decomposable functions of this type (or deep model induction) familiar in connectionist models of cognition and learning.

Toward a theory of evolution as multilevel learning

We apply the theory of learning to physically renormalizable systems in an attempt to outline a theory of biological evolution, including the origin of life, as multilevel learning. We formulate seven fundamental principles of evolution that appear to be necessary and sufficient to render a universe observable and show that they entail the major features of biological evolution, including replication and natural selection. It is shown that these cornerstone phenomena of biology emerge from the fundamental features of learning dynamics such as the existence of a loss function, which is minimized during learning. We then sketch the theory of evolution using the mathematical framework of neural networks, which provides for detailed analysis of evolutionary phenomena. To demonstrate the potential of the proposed theoretical framework, we derive a generalized version of the Central Dogma of molecular biology by analyzing the flow of information during learning (back propagation) and predicting (forward propagation) the environment by evolving organisms. The more complex evolutionary phenomena, such as major transitions in evolution (in particular, the origin of life), have to be analyzed in the thermodynamic limit, which is described in detail in the paper by Vanchurin et al. [V. Vanchurin, Y. I. Wolf, E. V. Koonin, M. I. Katsnelson, Proc. Natl. Acad. Sci. U.S.A. 119, 10.1073/pnas.2120042119 (2022)].