David Krakauer

Santa Fe Institute · Behavioral Sciences

Complex phenomena are made possible when: (i) fundamental physical symmetries are broken and (ii) from the set of broken symmetries historically selected ground states are applied to performing mechanical work and storing adaptive information. Over the course of several decades Philip Anderson enumerated several key principles that can follow from...

We survey a current, heated debate in the artificial intelligence (AI) research community on whether large pretrained language models can be said to understand language-and the physical and social situations language encodes-in any humanlike sense. We describe arguments that have been made for and against such understanding and key questions for th...

Organisms are able to partition resources adaptively between growth and repair. The precise nature of optimal partitioning and how this emerges from cellular dynamics including insurmountable trade-offs remains an open question. We construct a mathematical framework to estimate optimal partitioning and the corresponding maximal growth rate constrai...

We survey a current, heated debate in the AI research community on whether large pre-trained language models can be said to "understand" language -- and the physical and social situations language encodes -- in any important sense. We describe arguments that have been made for and against such understanding, and key questions for the broader scienc...

Adaptation to changing environments is a universal feature of life. How should adaptation -- to include the duration of memory -- scale with environmental rate of change given trade-offs in remembering vs. forgetting and active modification of the environment (e.g. niche construction)? We derive a universal scaling law for optimal memory duration a...

Adaptation to changing environments is a universal feature of life and can involve the organism modifying itself in response to the environment as well as actively modifying the environment to control selection pressures. The latter case couples the organism to environment. Then, how quickly should the organism change in response to the environment...

Institutions have been described as ‘the humanly devised constraints that structure political, economic, and social interactions.’ This broad definition of institutions spans social norms, laws, companies, and even scientific theories. We describe a non-equilibrium, multi-scale learning framework supporting institutional quasi-stationarity, periodi...

The original "Seven Motifs" set forth a roadmap of essential methods for the field of scientific computing, where a motif is an algorithmic method that captures a pattern of computation and data movement. We present the "Nine Motifs of Simulation Intelligence", a roadmap for the development and integration of the essential algorithms necessary for...

The original "Seven Motifs" set forth a roadmap of essential methods for the field of scientific computing, where a motif is an algorithmic method that captures a pattern of computation and data movement. We present the "Nine Motifs of Simulation Intelligence", a roadmap for the development and integration of the essential algorithms necessary for...

We argue for multiple forms of life realized through multiple different historical pathways. From this perspective, there have been multiple origins of life on Earth—life is not a universal homology. By broadening the class of originations, we significantly expand the data set for searching for life. Through a computational analogy, the origin of l...

Armed conflict data display features consistent with scaling and universal dynamics in both social and physical properties like fatalities and geographic extent. We propose a randomly branching armed conflict model to relate the multiple properties to one another. The model incorporates a fractal lattice on which conflict spreads, uniform dynamics...

Recent work suggests that collective computation of social structure can minimize uncertainty about the social and physical environment, facilitating adaptation. We explore these ideas by studying how fission-fusion social structure arises in spider monkey (Ateles geoffroyi) groups, exploring whether monkeys use social knowledge to collectively com...

We consider biological individuality in terms of information theoretic and graphical principles. Our purpose is to extract through an algorithmic decomposition system-environment boundaries supporting individuality. We infer or detect evolved individuals rather than assume that they exist. Given a set of consistent measurements over time, we discov...

Armed conflict data display scaling and universal dynamics in both social and physical properties like fatalities and geographic extent. We propose a randomly branching, armed-conflict model that relates multiple properties to one another in a way consistent with data. The model incorporates a fractal lattice on which conflict spreads, uniform dyna...

Large-scale armed conflict is a characteristic feature of modern civilization. The statistics of conflict show remarkable regularities like power law distributions of fatalities and durations, but these properties have remained disparate, albeit prominent, features of conflict. We explore a large, detailed data set of $10^5$ armed conflict reports...

In many biological systems, the functional behavior of a group is collectively computed by the system’s individual components. An example is the brain’s ability to make decisions via the activity of billions of neurons. A long-standing puzzle is how the components’ decisions combine to produce beneficial group-level outputs, despite conflicts of in...

What is history anyway? Most people would say it’s what happened in the past, but how far back does the past extend? To the first written sources? To what other forms of evidence reveal about pre-literate civilizations? What does that term mean—an empire, a nation, a city, a village, a family, a lonely hermit somewhere? Why stop with people: should...

We explore how ideas from infectious disease and genetics can be used to uncover patterns of cultural inheritance and innovation in a corpus of 591 national constitutions spanning 1789–2008. Legal “ideas” are encoded as “topics”—words statistically linked in documents—derived from topic modeling the corpus of constitutions. Using these topics we de...

In biological systems, prolonged conflict is costly, whereas contained conflict permits strategic innovation and refinement. Causes of variation in conflict size and duration are not well understood. We use a well-studied primate society model system to study how conflicts grow. We find conflict duration is a ‘first to fight’ growth process that sc...

A central question in cognitive neuroscience is how unitary, coherent decisions at the whole organism level can arise from the distributed behavior of a large population of neurons with only partially overlapping information. We address this issue by studying neural spiking behavior recorded from a multielectrode array with 169 channels during a vi...

Recent advances suggest that the concept of information might hold the key to unravelling the mystery of life's nature and origin. Fresh insights from a broad and authoritative range of articulate and respected experts focus on the transition from matter to life, and hence reconcile the deep conceptual schism between the way we describe physical an...

Supplementary figures, supplementary table, supplementary notes and supplementary references.

A critical feature of all cellular processes is the ability to control the rate of gene or protein expression and metabolic flux in changing environments through regulatory feedback. We review the many ways that regulation is represented through causal, logical, and dynamical components. Formalizing the nature of these components promotes effective...

A critical feature of all cellular processes is the ability to control the rate of gene or protein expression and metabolic flux in changing environments through regulatory feedback. We review the many ways that regulation is represented through causal, logical and dynamical components. Formalizing the nature of these components promotes effective...

In biological function emerges from the interactions of components with only partially aligned interests. An example is the brain—a large aggregation of neurons capable of producing unitary, coherent output. A theory for how such aggregations produce coherent output remains elusive. A first question we might ask is how collective is the behavior of...

Over the last decade new technologies for making large numbers of fine-grained measurements have led to the surprising discovery that many biological systems sit near a critical point. These systems are potentially more adaptive in that small changes to component behavior can induce large-scale changes in aggregate structure and function. Accountin...

We explore how ideas from infectious disease and genetics can be used to uncover patterns of cultural inheritance and innovation in a corpus of 591 national constitutions spanning 1789 - 2008. Legal "Ideas" are encoded as "topics" - words statistically linked in documents - derived from topic modeling the corpus of constitutions. Using these topics...

I consider the many ways in which evolved information-flows are restricted and metabolic resources protected and hidden -- the thesis of living phenomena as evolutionary cryptosystems. I present the information theory of secrecy systems and discuss mechanisms acquired by evolved lineages that encrypt sensitive heritable information with random keys...

Genetic redundancy is often associated with the duplication of an open reading frame within a genome or a multiplicity of regulatory elements sharing a target. Genetic redundancy is inferred when the modification or deletion of a portion of functional genetic material results in minimal changes to a trait or organismal phenotype – robustness – and...

We discuss a set of computational techniques, called Inductive Game Theory, for extracting strategic decision-making rules from time series data and constructing probabilistic social circuits. We construct these circuits by connecting component individuals and groups with strategies in a game and propose an inductive approach to reconstructing the...

This figure shows show the sensitivity of each algorithm to source bias in a shuffled primate communication matrix. For each algorithm, we report the drop in rank induced when a node receives all of its edges from one of its neighbors. The point shows the mean correlation and the bars show plus or minus one standard deviation. The algorithms are or...

Flow chart for the calculation of the various algorithms. The color indicates our intuitions about the complexity of each algorithm, with darker grays corresponding to more complex calculations. (PDF)

The distribution of the scores resulting from each algorithm as applied to the subordination signaling network. (PDF)

The distribution of the logarithms of the scores resulting from each algorithm as applied to the physicist collaboration network. (PDF)

This figure shows for the physicist collaboration network the fit of each algorithm to the functional data. The x-axis indicates which subset of nodes are being considered– 1 is the top quartile, 2 is the top half, 3 is the top three quartiles, 4 is all nodes, 5 is the bottom three quartiles, 6 is the bottom half, and 7 is the bottom quartile– wher...

This figure shows for the functional linkage network of genes the fit of each algorithm to the functional data. The x-axis indicates which subset of nodes are being considered– 1 is the top quartile, 2 is the top half, 3 is the top three quartiles, 4 is all nodes, 5 is the bottom three quartiles, 6 is the bottom half, and 7 is the bottom quartile–...

The worst case error of the naive entropy estimator as a function of the naive estimator. Each data point represents the naive entropy estimate and worst case error of one individual's receiving distribution. Figure was created by Simon DeDeo, 2011. (PDF)

This figure shows the effects of varying the length of the random walk, , on the power scores computed using David's Score. The axis shows David's score for each individual, normalized so that the highest score is and the lowest score is . Each line corresponds to the generalization of David's score for different , as specified in the Figure legend...

The distribution of the scores resulting from each algorithm as applied to the yeast functional linkage network. (PDF)

This figure shows how predictive value of eigenvector centrality changes as a function of the redistribution parameter. A. shows how the value from a multivariate regression of eigenvector centrality on the subordination signaling network against the external data depends on the redistribution probability. B. shows how the value from a regression o...

Each plot shows the (normalized) entropy scores and (normalized) algorithm scores for the four algorithms that do not correlate well with entropy on four artificial data sets constructed to show this lack of correlation. The solid lines are the entropy scores and the dashed lines are the algorithms. (PDF)

This figure shows how predictive value and skewness of eigenvector centrality scores are related. A. shows how the value from a multivariate regression of eigenvector centrality on the primate communication network against the external data depends on the skewness of the scores. B. shows how the value from a regression of eigenvector centrality on...

Biological and social networks are composed of heterogeneous nodes that contribute differentially to network structure and function. A number of algorithms have been developed to measure this variation. These algorithms have proven useful for applications that require assigning scores to individual nodes-from ranking websites to determining critica...

A multidisciplinary examination of cognitive mechanisms, shaped over evolutionary time through natural selection, that govern decision making. How do we make decisions? Conventional decision theory tells us only which behavioral choices we ought to make if we follow certain axioms. In real life, however, our choices are governed by cognitive mechan...

Animals living in groups collectively produce social structure. In this context individuals make strategic decisions about when to cooperate and compete. This requires that individuals can perceive patterns in collective dynamics, but how this pattern extraction occurs is unclear. Our goal is to identify a model that extracts meaningful social patt...

Literature is a form of expression whose temporal structure, both in content and style, provides a historical record of the evolution of culture. In this work we take on a quantitative analysis of literary style and conduct the first large-scale temporal stylometric study of literature by using the vast holdings in the Project Gutenberg Digital Lib...

A common feature of biological networks is the geometrical property of self-similarity. Molecular regulatory networks through to circulatory systems, nervous systems, social systems and ecological trophic networks show self-similar connectivity at multiple scales. We analyse the relationship between topology and signalling in contrasting classes of...

We study a mechanism for reliable switching in biomolecular signal-transduction cascades. Steady bistable states are created by system-size cooperative effects in populations of proteins, in spite of the fact that the phosphorylation-state transitions of any molecule, by means of which the switch is implemented, are highly stochastic. The emergence...

We review an empirically grounded approach to studying the emergence of collective properties from individual interactions in social dynamics. When individual decision-making rules, strategies, can be extracted from the time-series data, these can be used to construct adaptive social circuits. Social circuits provide a compact description of collec...

Natural selection is shown to be an extended instance of a Maxwell's demon device. A demonic selection principle is introduced that states that organisms cannot exceed the complexity of their selective environment. Thermodynamic constraints on error repair impose a fundamental limit to the rate that information can be transferred from the environme...

We present statistical evidence and dynamical models for the management of conflict and a division of labor (task specialization) in a primate society. Two broad intervention strategy classes are observed--a dyadic strategy--pacifying interventions, and a triadic strategy--policing interventions. These strategies, their respective degrees of specia...

Teaming up with science writer Highfield to address lay readers, Nowak argues for the fundamental importance of cooperation throughout life's history.

a b s t r a c t Scientific theories seek to provide simple explanations for significant empirical regularities based on fundamental physical and mechanistic constraints. Biological theories have rarely reached a level of generality and predictive power comparable to physical theories. This discrepancy is explained through a combination of frozen ac...

We analyse the timescales of conflict decision-making in a primate society. We present evidence for multiple, periodic timescales associated with social decision-making and behavioural patterns. We demonstrate the existence of periodicities that are not directly coupled to environmental cycles or known ultraridian mechanisms. Among specific biologi...

Stochastic multi-modal communication Biological systems are inherently noisy and typically comprised of distributed, partially autonomous components. These features require that we understand evolutionary traits in terms of probabilistic design principles, rather than traditional deterministic, engineering frameworks. This characterisation is parti...

Sigmund offers a fairly rigorous introduction to game-theory approaches to "interactions among individuals guided by self-interest and caught in social traps."

We consider the value of structured priors in the analysis of data sampled from complex adaptive systems. We propose that adaptive dynamics entails basic constraints (memory, information processing) and features (optimization and evolutionary history) that serve to significantly narrow search spaces and candidate parameter values. We suggest that t...

Supporting information for the main MS. (0.16 MB PDF)

Author Summary Persistent conflict is one of the most important contemporary challenges to the integrity of society and to individual quality of life. Yet surprisingly little is understood about conflict. Is resource scarcity and competition the major cause of conflict, or are other factors, such as memory for past conflicts, the drivers of turbule...

Chromatin regulation is understood to be one of the fundamental modes of gene regulation in eukaryotic cells. We argue that the basic proteins that determine the chromatin architecture constitute an evolutionary ancient layer of transcriptional regulation common to all three domains of life. We explore phylogenetically, sources of innovation in chr...

Even the recognition of an individual whom we see every day is only possible as the result of an abstract idea of him formed by generalization from his appearances in the past James G. Frazer I don't paint things. I only paint the difference between things Henri Matisse Mathematics is the art of giving the same name to different things Henri Poinca...

The precise elucidation of the gene concept has become the subject of intense discussion in light of results from several, large high-throughput surveys of transcriptomes and proteomes. In previous work, we proposed an approach for constructing gene concepts that combines genomic heritability with elements of function. Here, we introduce a definiti...

A moral system is an adaptive system for conflict management based on prescriptive, internalized social rules. We decompose moral systems into the sense of fairness, moral judgments, and rules at the aggregate level. We explore how each of these levels is constructed, including how this process is influenced by cognitive and organizational constrai...

Résumé L’histoire cherche à combiner des descriptions particulières dans des cadres généraux dans le but d’expliquer des séquences d’événements. Dans cet esprit, l’histoire adopte une approche transdisciplinaire qui couvre une variété de champs allant de la biologie à la géologie, en passant par l’anthropologie et l’histoire humaine. Je cherche à c...

The behavior of organisms can contribute to the transformation of their environments. When organismal impacts on the environment feed back to influence organismal density, viability, fertility, or persistence, the environment can be construed as an extension of the organism. This process of fitness-enhancing environmental transformation has been ca...

The first comprehensive general resource on state-of-the-art protocell research, describing current approaches to making new forms of life from scratch in the laboratory. Protocells offers a comprehensive resource on current attempts to create simple forms of life from scratch in the laboratory. These minimal versions of cells, known as protocells,...

We explore adaptive theories for the diversity of protein translation based on the genetic code viewed as a primitive immune system. Immunity is acquired through a genetic mechanism of non-recognition of parasite genomes. Modifying the set of codons bound by tRNA anticodon molecules or changing the specificity of binding, reduces the replication ra...

Studies of the evolution of development characterize the way in which gene regulatory dynamics during ontogeny constructs and channels phenotypic variation. These studies have identified a number of evolutionary regularities: (1) phenotypes occupy only a small subspace of possible phenotypes, (2) the influence of mutation is not uniform and is ofte...

A loglog plot of the distribution of degeneracy levels among visible phenotypes for varying regulatory densities. The settings are again identical to those described in Figure 3A in the main text, but with the matrix density, c, set to (A) 0∶1, (B) 0∶25, (C) 0∶5, and (D) 1. Each point denotes the expected number of distinct phenotypes with a certai...

A loglog plot of the distribution of degeneracy levels among visible phenotypes using varying number of regulatory levels. The settings are identical to those described in Figure 3A in the main text, but using (A) 1, (B) 2, (C) 5, (D) 10, (E) 25, and (F) 50 regulatory layers. Each point denotes the expected number of distinct phenotypes with a cert...

Supporting Text: Numerical Analysis (1.27 MB PDF)

(A) A loglog plot of the distribution of degeneracy levels among visible phenotypes as obtained by the numerical analysis. Each point denotes the expected number of developmental plans in which the ‘half ones’ phenotype obtains a certain degeneracy level, and is averaged over 1,000,000 different plans. From symmetry considerations, this distributio...

During the course of development cells undergo division producing a variety of cell types. Proliferation and differentiation are dependent on both genetic programs, encoded by the cellular genome, and environmental cues produced by the local cellular environment imposing local selection pressures on cells. We explore the role that cellular signals...

Genetic redundancy typically relates to the duplication of an open reading frame within a genome. Genetic redundancy is often inferred when the modification or deletion of a portion of genetic material in a duplicated genome results in minimal changes in trait or organismal phenotype in reference to the nonduplicated wildtype. This invariance has b...

We consider the concerted evolution of viral genomes in four families of DNA viruses. Given the high rate of horizontal gene transfer among viruses and their hosts, it is an open question as to how representative particular genes are of the evolutionary history of the complete genome. To address the concerted evolution of viral genes, we compared g...

Dynamics of Cancer. Incidence, Inheritance, and Evolution. By Steven A. Frank. Princeton University Press, Princeton, NJ, 2007. 392 pp. $99.50, £59.95. ISBN 9780691133652. Paper, $39.50, £23.95. ISBN 9780691133669. Princeton Series in Evolutionary Biology. The author combines molecular, demographic, and evolutionary analyses to examine genetic and...

A recent workshop discussed the recognition of multiple distinct ligands by individual T cell and B cell receptors and the implications of this discovery for lymphocyte biology. The workshop recommends general use of the term polyspecificity because it emphasizes two fundamental aspects, the inherent specificity of receptor recognition and the abil...

Of considerable interest are the evolutionary and developmental origins of complex, adaptive structures and the mechanisms that stabilize these structures. We consider the relationship between the evolutionary process of gene duplication and deletion and the stability of morphogenetic patterns produced by interacting activators and inhibitors. We c...

— When the same sequence of nucleotides codes for regions of more than one functional polypeptide, this sequence contains overlapping genes. Overlap is most common in rapidly evolving genomes with high mutation rates such as viruses, bacteria, and mitochondria. Overlap is thought to be important as: (1) a means of compressing a maximum amount of in...

We provide a geometric framework for investigating the robustness of information flows over biological networks. We use information measures to quantify the impact of knockout perturbations on simple networks. Robustness has two components, a measure of the causal contribution of a node or nodes, and a measure of the change or exclusion dependence,...

Among microbial genomes, genetic information is frequently compressed, exploiting redundancies in the genetic code in order to store information in overlapping genes. We investigate the length, phase and orientation properties of overlap in 58 prokaryotic species evaluating neutral and selective mechanisms of evolution. Using a variety of statistic...