ABSTRACT. Natural selection is the strongest known antientropic process in the universe when operating at the
biological level and may also operate at the cosmological level. Consideration of how biological natural
selection creates adaptations may illuminate the consequences and significance of cosmological natural
selection. An organismal trait is more likely to constitute an adaptation if characterized by more
improbable complex order, and such order is the hallmark of biological selection. If the same is true of
traits created by selection in general, then the more improbably ordered something is (i.e., the lower its
entropy), the more likely it is to be a biological or cosmological adaptation.
By this logic, intelligent life (as the least-entropic known entity) is more likely than black holes or anything
else to be an adaptation designed by cosmological natural selection. This view contrasts with Smolin’s
suggestion that black holes are an adaptation designed by cosmological natural selection and that life is
the by-product of selection for black holes. Selection may be the main or only ultimate antientropic
process in the universe/multiverse; that is, much or all observed order may ultimately be the product or
by-product of biological and cosmological selection.

ABSTRACT. Evolutionary development, evo devo or ED is a term that can be used as a replacement for the more
general term “evolution”, whenever a scholar thinks experimental, selectionist, contingent and stochastic
or “evolutionary” processes, and also convergent, statistically deterministic (probabilistically predictable)
or “developmental” processes, including replication, may be simultaneously occurring in any complex
system. The hyphenated “evo-devo” is commonly used for living systems, most prominently in evo-devo
genetics, and the unhyphenated “evo devo” can be used for the theory of any potentially replicating and
adapting complex system (star, prebiotic system, gene, cell, meme (concept), behavior, technology),
whether living or nonliving. In replicating biological and other complex systems, evolutionary processes
generate new information, and developmental processes conserve old information. Both processes can
be differentiated in any replicating complex system, and both are presumably fundamental to adaptation,
and the ways each system encodes representations (models, intelligence) of itself and its environment.
Like living systems, our universe broadly exhibits both stochastic and deterministic components, in all
historical epochs and at all levels of scale. It has a definite birth and it is inevitably senescing toward heat
death. The idea that we live in an “evo devo universe,” one that has self-organized over past replications
both to generate multilocal evolutionary variation (preselected diversity), and to convergently develop and
pass to future generations selected aspects of its accumulated complexity ("intelligence") is an obvious
hypothesis. Yet very few cosmologists or physicists, even in the community that theorizes universal
replication and the multiverse, has entertained the idea that our universe may be both evolving and
developing (engaging in self-organized, goal-driven, directional change and a replicative life cycle). Even
our leading current models of universal replication, like Lee Smolin's cosmological natural selection
(CNS), do not yet use the concept of universal development, or refer to development literature.

I will argue that such evo devo universe models must emerge, including models of CNS with Intelligence
(CNS-I), which explore the ways emergent intelligence constrains and directs “natural” selection, if we are
to understand such puzzles as the fine-tuned universe hypothesis, convergent evolution, and our own
planet’s history of unreasonably smooth and resilient acceleration of “leading edge” complexity and
intelligence, even while suffering continual catastrophic selection events. If they are to become
predictively validated in living systems and in all other adaptive replicators, evo devo models will require
both better simulation capacity and advances in a variety of theories, which I shall briefly review.

ABSTRACT. Ethical intuitions or systems emerged within human evolution and the rise of human societies. Most of
the time, they are anthropocentric in the sense that they value human beings above anything else. How
can we think about a universal ethics that could apply to all living things, but also to the rising population
of cyborgs, intelligent machines, intelligent algorithms or even potential extraterrestrial life? We generally
give value to complex structures, to objects resulting from a long history, to systems with many elements
and with many links finely adjusted. These include living beings, books, works of art or scientific theories.
Intuitively, we want to keep, multiply, and share such structures, as well as prevent their destruction.
Such objects are not any kind of information where more information (in bits) would simply mean more
value. Instead, they have value because they require a long computational history, numerous
interactions that we can assimilate to a computation, and that we call “organized complexity”.
To propose the foundations of a universal ethics based on the intrinsic value of organized complexity, we
explore three key issues: (1) How can we characterize mathematically this organized complexity?
Charles Bennett proposed a notion that is certainly a first approximation of what we are looking for:
“logical depth”. (2) Could such an organized complexity be useful to propose and develop a clear and
consistent universal ethics? (3) What would be the consequences of such a universal ethics, and how
does it fit with other general value systems?

ABSTRACT. Our species has had three major technological revolutions: agricultural, industrial, and informational. We can say that each of them has harnessed matter, energy, and information, respectively. Also, each revolution has transformed societies in general and education in particular. I will review the changes that our technologies have produced in education and focus on how computers have enabled us to study complex systems. Our recent understanding of complexity forces us to change how we teach. These changes will have an important impact on our species.

ABSTRACT. Generations have long been described by the science of their time. The last 200 years provides a set of familiar examples: the landscape of the Gilded Age was dominated by labor, vast in its quantity and skill. The Second Industrial Revolution was characterized by machines and the science of mass production. In the Information Age, science and technology enabled exponential increases in speed and miniaturization. However, those countries, businesses, and people who remain fixated on the science of the day are left behind tomorrow. These shifts to the next generation of scientific advancement continue. Today the world needs new science to understand the implications of large scale connectedness - less at the individual level, and more at that of the entire system. Network Science provides a framework with which to study our world as it is becoming defined by connectivity. As the world continues to change, one may expect education to likewise evolve and adapt. However, in many ways (though not all) our educational system functions in the same way as it did in the early 20th century when we broke apart the "one-room schoolhouse" and began teaching different age groups at different times in different rooms. Linear thinking persists in our educational system now just as it did then. Although we now live in a highly non-linear environment that can be characterized by networks and connectivity, network elements do not enter the educational lexicon until the Master’s and PhD levels of education. Yet younger students already have a natural intuition about networks because they live connected lives. This natural inclination could and should be leveraged as we introduce alternative methods for teaching the challenging subjects that include network elements, making these subjects more attainable to more students. Many educators have recognized that it is past time for our educational system to be updated to reflect the interconnected world in which we live.

We organized the NiCE (Networks in Classroom Education) Teacher Workshop (http://bit.ly/2s9GlRH) to facilitate the conversations and collaborations between educators and scientists necessary to address this demand for a revolutionary update to our educational system. It was held at the United States Military Academy at West Point as a four-day workshop on July 10-13, 2017, with around 30 participants that encompassed teachers and administrators (whose expertise spanned disciplines from math to remedial reading, across the whole K-12 range), and a number of experts in Network Science. Its goal was to educate K-12 teachers and administrators about Network Science, and to enable and empower those teachers and administrators to bring network thinking and ideas to their students, schools, and districts. During the workshop, network thinking was not only presented as concepts to be taught to students, but also actively utilized as a tool to make curriculum development and delivery easier and more successful, and to explore and explicate school-wide challenges. The participating teachers and administrators developed presentations and concrete lesson plans that utilized Network Science and network thinking. These ideas and plans collectively demonstrate tremendous opportunities to improve education, both by quantitatively identifying curricular elements central to interdisciplinary learning and describing them with this lens, and by systematically examining the curriculum and standards, and exploiting network thinking to meet their requirements. Ultimately, these central topics may be accessible to a greater range of students, via methods that are preferential to and easier for the educators delivering them.

ABSTRACT. The Mathematics of Patterns is an award-winning website containing educational resources for the study of pattern formation in nature, through Turing or diffusion-limited instabilities.  While undergraduate-level mathematics courses in dynamical systems focus heavily on technical aspects of solving differential equations, many of the concepts in these courses lend themselves to being understood visually in a more intuitive fashion. As such, the aim of the project was to produce easily-digestible audio and visual media for learning theoretical and numerical concepts in mathematical modelling, reaction-diffusion equations, and dynamical systems.

​

 I will present a summary of the project and how it has been received by the community of applied mathematicians, lecturers, and students worldwide in the four years from when the notes were first posted. Although the original target audience were undergraduate mathematics students, surprisingly, the website has reached a wider non-specialist audience through popular science communication channels that have used the online materials (e.g. Minute Earth and Blablalogia). I will discuss what I learned from this experience, and the workflow and commitment involved in creating such materials. Finally, I hope to discuss the challenges that lie ahead for academics who are interested in creating such audio-visual media as part of their teaching or public engagement.

ABSTRACT. In the present work a model is proposed in terms of network formation and evolution over time where the agents are leveraged institutions in the financial system made up of inter-bank connections. With heterogeneous types of units, we have fluctuations in terms of created links with a phase transition for various proportions of agents’ types. The target of the model is to observe and describe emergent structures that characterize the entire system in terms of critical pattern formation and resilience to perturbations.

ABSTRACT. In adaptive systems heterogeneous, error-prone agents extracting regularities from noisy data make strategic decisions in evolutionary, developmental, or learning time. In order to understand how these noisy strategies collectively combine to produce emergent function at the aggregate scale we need a theory of information aggregation. Given noisy data, noisy sensors, and finite time, information aggregation is a central challenge faced by adaptive systems from biofilms to neural systems to primate societies to financial markets to hybrid human-ai systems. In this talk I will discuss different mechanisms for information aggregation and outline open questions. 

ABSTRACT. We present results from our most recent simulations and data analysis of the positive and negative
feedback loops between the characteristics of evolving complex systems. Those loops lead to
exponential growth with sinusoidal oscillations which can be modeled with a system of coupled harmonic
oscillators.
The solutions of this model provide an exponential equation superimposed with sinusoidal oscillations.
The amplitude and frequency of those oscillations also grow exponentially with time. This solution
matches data well and gives us an insight into the feedback loops in evolving complex systems as a
proposed mechanism for their observed exponential rates of self-organization and progressive
development.

ABSTRACT. We review the definition and use of the complexity profile which captures the way system collective
behaviors reduce the freedom of individual components. In combination with Ashby’s law (of requisite
variety) it provides a means of relating the organizational structure of a system to the ability of a system
to perform tasks at multiple scales. Applications to social systems are helpful in clarifying how
organizational structures are or are not effective at performing certain tasks. Implications for the
difficulties of social organizations today will be discussed.

ABSTRACT. Self-propelled particles have long been used to describe the emergence of collective behaviours
according to various physical or social rules. Here we show a recently discovered, novel, and
extraordinary simple motion law for governing self-propelled particles that leads to a new quality of
collective behaviour: Identical particles start interacting and forming a self-structuring, self-reproducing
and self-sustaining life-like system, starting from an initial randomized spatial configuration. On an
individual level, these structures exhibit many properties also found in living organisms, such as isolation
to seclude them from their surrounding environment, growth and self-regulation of internal state
variables, self-replication, distinct behaviour that changes according to environmental alterations during
their lifetime or the ability to die. Furthermore, these structures show to be very resilient against sensor
or actuation noise. The emergent structures show a distinct life cycle and create their own ecosystem by
interacting with each other.
On this ecological scale, specific population dynamics arise with macroscopic properties resembling
population dynamics found in nature. We show the observed dynamics, the emerging spatio-temporal
structures with basic properties (size distributions, longevity), a macroscopic top-down model as well as
a probabilistic microscopic bottom-up model of the system. By changing parameters of the motion law, a
multiplicity of different particle behaviours can be achieved as a rudimentary parameter sweep shows.
The generality and simplicity of the motion law provokes the thought that one fundamental rule,
described by one simple equation yields various structures in nature: it may work on different time- and
size scales, ranging from the self-structuring universe, to emergence of living beings, down to the
emergent subatomic formation of matter.

ABSTRACT. This paper is an introduction to a novel method for visualizing the dynamics of evolutionary algorithms in
the form of networks. The idea is based on the obvious similarity between interactions between
individuals in a swarm and evolutionary algorithms and for example, users of social networks, linking
between web pages, etc.
In this paper two different areas of complexity research are productively merged: (complex) networks and
evolutionary computation. As mentioned, interactions amongst the individuals in a swarm and in
evolutionary algorithms can be considered like user interactions in social networks or like collections of
interacting people in society. When systems of evolutionary algorithms are best modeled as social
interactions or as swarm colonies is an open question of research.
The analogy between individuals in populations in an arbitrary evolutionary algorithm and vertices of a
network is discussed, as well as between edges in a network and communication between individuals in a
population. The possibility of visualizing the dynamics of network using the coupled map lattices method
and control using chaos control techniques are also discussed.

ABSTRACT. Due to the interdisciplinary nature and reach of complex systems as a field, students undertaking courses in complex systems at University level have diverse backgrounds across physics, mathematics, computer science, engineering, biology, neuroscience, economics, social sciences and the humanities. This brings challenges (e.g. diversity of skills, computer programming and analysis ability) but also opportunities (e.g. facilitating interdisciplinary interactions and projects, and applications that meet disciplinary needs) for the classroom. However, there is little published regarding how these challenges and opportunities are handled in teaching and learning Complex Systems as an explicit subject in higher education, and how this differs in comparison to other subject areas. We seek to explore these particular challenges and opportunities in more depth, by examining the primary body of knowledge currently residing in the experience of pioneering teachers and learners in this space. We report an interview-based study of several such subjects (conducted amongst the authors) on their experiences, and a discussion and analysis comparing and contrasting those experiences. Our discussions explore: how curriculum design was approached, how theories/models/frameworks of teaching and learning informed their decisions and experience, how diversity in student backgrounds was addressed, and assessment task design. We find a striking level of commonality in the issues expressed as well as the strategies to handle them. For example: there was a significant focus on problem- or activity-based learning; a focus on understanding and applying key principles with technical analysis and programming implementation as a means to this end; and the use of major student-led creative projects for both achieving and assessing learning outcomes. While similar approaches to curriculum design (e.g. constructive alignment) were observed, curriculum content was the main area recognised as being contested since the field is still rapidly evolving, however this can be interpreted as a strength of the field in tightly knitting research and teaching into the one community.

ABSTRACT. The Santa Fe Institute is using online learning to promote the understanding and application of key concepts in complexity science. Along with the "traditional" use of online teaching methods such as MOOCS, the Institute has created a novel, capstone learning experience for our online students that encourages them to apply the key concepts they learned from the various MOOCS to generate a solution to an open-ended project tied to a real-world application.  This Complexity Challenge, piloted in August and September of 2017, provides a nice method by which students can actively apply and synthesize their online learning in a context that is inherently self motivating.  Submitted solutions are initially peer reviewed and the better ones are then ranked by faculty mentors and outside experts, with prizes awarded to the top tier.  Here we outline the key elements of this Challenge-based methodology, and summarize lessons learned from the pilot project.

ABSTRACT. This study is concerned with the question whether the creation of small high schools has a favorable impact on daily attendance rates in those schools. The determinants of daily school attendance are under-researched in education, particularly considering the ongoing concerns about students leaving school prematurely. Reducing the likelihood of students dropping out through individualized attention and support is part of the rationale for the small high schools movement, and therefore, the impact of small high schools creation on student attendance is relevant. Are attendance rates higher in a given school after than before the initiation of such initiatives and do they show greater stability and predictability afterward?  The former question can be addressed through conventional summary statistics (mean, range), the latter question requires a detailed description of attendance rates sequentially ordered over a longer time period. The methodology for analyzing such information is available (e.g., Beran, 1994), but rarely used in education.

​

In this study, the daily attendance rates in one public school in New York City (School A) are analyzed over a seven-year period, i.e., from September 2007 through June 2014 (N = 1,245). School A enrolled approximately 900 students up to and through the 2009-2010 school year. Afterward, enrollment was reduced to about 250 students and remained stable for the period under study. Two complementary analytical approaches were used to estimate the effects of the enrollment reduction in this case. First, the impact of perturbation on the series was estimated, particularly in connection with the point of transition to a small school, using intervention analysis (Peña, 2001). Second, the statistical properties of the temporal process before and after the transition were examined for signs of meta-stability and self-organized criticality (Bak, 1996), using a fractional differencing approach (Beran, 1994).

​

Higher average attendance, as well as a gradual increase in the rates, was found in the period after the transition, although the immediate upward jump at the beginning of the 2010-11 school year is followed by a relapse. However, the findings indicate that overall, the underlying dynamics of attendance rates were favorably impacted by the small high schools implementation, i.e., they were more stable after than before the transition to the small high school format. Specifically, the analyses reveal meta-stability (edge of chaos) after the change, and instability (Brownian motion) before. These processes would have remained hidden if conventional statistics were used, making the case for a thorough analysis of temporal patterns in educational research, and the engagement of educational researchers in the use of the requisite methodology. Implications and limitations of the study are discussed.

​

References

Bak, P. (1996). How nature works: The science of self-organized criticality. New York: Springer.

Beran, J. (1994). Statistics for long-memory processes. Boca Raton, FL: Chapman & Hall/CRC.

Peña, D. (2001). Outliers, influential observations, and missing data. In D. Peña, G. C. Tiao, & R. S. Tsay (Eds.) A course in time series analysis (pp. 136-170). New York: Wiley & Sons, Inc.

ABSTRACT. Valisner (2008) defines education as the process of setting up conditions for a developing person to be open to innovation or change. From this perspective, educators should be invested in understanding how to meaningfully direct and support the transformation of the developing person into that which the person is becoming, rather than invested in investigation of the state of a person at any given point in time. Educational researchers have begun to explicitly use complex systems as a research paradigm to more adequately design and conduct research capable of capturing dynamic processes inherent in education (Jacobson, et al., 2016; Kaplan & Garner, in press; Koopmans & Stamovlasis, 2016). However, designing research in education from a complex systems perspective places a new set of demands on scholars, which includes engaging in a much deeper consideration of the nature of change of the phenomena of interest and how to treat time. Choices that researchers make about when to begin a study to determine the initial condition of the system or spacing of measurement points creates assumptions about causality, feedback, and patterns of change of our educational process of interest. This conceptual paper draws upon empirical examples to help researchers more explicitly consider how to address time in complex systems research in education.

​

Drawing upon the educational construct of engagement, we provide examples of the types of questions education researchers may consider asking of time and how to explicitly bind time and address issues of time from a design, measurement, and analytic standpoint. Theory suggests that student engagement is a dynamic process that exists both within and between people (Skinner et al., 2016). Ongoing reciprocal exchanges between person and context over time create self-reinforcing adaptive or maladaptive patterns of engagement, which in turn influence learning. These types of effects are described as “self amplifying” (Skinner & Pitzer, 2012, p. 34) and become autocatalytic (Kauffman, 1989). When engagement becomes autocatalytic it self-regulates and self-replicates with less energy and conscious deliberation. Producing empirical evidence of these states is a challenge for the field.  We might ask the following questions: How much time needs to be observed to understand the dynamics of engagement at different levels of analysis? How many observations within a given time period are required to understand the self-amplifying states that provide enhanced learning? At what point in a dynamic period does adaptive engagement translate to learning benefits or create a buffer against academic challenges or setbacks?

​

The complex systems education researcher cannot define their system of interest with a cursory understanding of time, bounding the change process in convenience (e.g., the beginning and end of a semester) even if eventually practical considerations dictate a balance between theoretically or empirically indicated change processes and convenience during the design of the study itself. Taking time seriously includes forefronting time as a critical element in creating boundaries around the system and including time as a design element in research.

​

References

Jacobson, M. J., Kapur, M., & Reimann, P. (2016). Conceptualizing Debates in Learning and Educational Research: Toward a Complex Systems Conceptual Framework of Learning. Educational Psychologist, 1-9. http://dx.doi.org/10.1080/00461520.2016.1166963

Kaplan, A. & Garner, J. (in press). A Complex Dynamic Systems Perspective on Identity and Its Development: The Dynamic Systems Model of Role Identity. Developmental Psychology.

Kauffman, S. A. (1989). Principles of adaptation in complex systems. Lectures in the Sciences of Complexity, 1, 619-712.

Koopmans, M., & Stamovlasis, D. (Eds.). (2016). Complex Dynamical Systems in Education: Concepts, Methods and Applications. Springer. http://dx.doi.org/10.1007/978-3-319-27577-2

Skinner, E. A., & Pitzer, J. (2012). Developmental dynamics of engagement, coping, and

everyday resilience. In S. Christenson, A. Reschly, & C. Wylie (Eds.), The Handbook of Research on Student Engagement (pp. 21-45). New York: Springer Science.

Skinner, E. A. (2016). Engagement and disaffection as central to processes of motivational

resilience and development. Handbook of Motivation at School. 2nd ed. New York, NY: Routledge, 145-68.

Valsiner, J. (2008). Open intransitivity cycles in development and education: Pathways to

synthesis. European journal of psychology of education, 23(2), 131-147.

​

ABSTRACT. .

ABSTRACT. In this paper we study the European bank network and its vulnerability to stressing different bank assets as well as sudden drops in equity levels. The importance of macro-prudential policy is emphasized by the inherent vulnerability of the financial system, high level of leverage, interconnectivity of system’s entities, similar risk exposure of financial institutions, and potential for systemic crisis propagation through the system. Our findings suggest that to better manage systemic risk regulators need sophisticated tools for real time monitoring and systemic stress-testing of the financial system.

ABSTRACT. In this paper we focus our attention on the exploitation of the information contained in financial news to enhance the performance of a classifier of bank distress. Among the different models proposed for such purpose, we investigate one of the possible deep learning approaches, based on a doc2vec representation of the textual data. The final aim is not only the improvement of the predictive performance of the classifier but also to assess the importance of news data in the classification process. Does news data really bring more useful information not contained in standard financial variables? Our results seem to confirm such hypothesis.

ABSTRACT. Countering evolving threats
Ana Isabel Barros (*), B. van der Vecht, B. Boltjes, B. Keijser, N. de Reus
TNO, The Hague, The Netherlands
(*) UvA-IAS

The hyper connectivity of the society, has enlarged the dynamics and volatility of conflicts and criminality and ensured that these are no longer contained within national borders or even ideologies. In fact, criminal activities provide new tactics for war fighting as well as extra resources to fund conflicts and terrorist activities. This entanglement accelerates not only this so-called hybrid threats but it also accelerates their evolution, and demands adaptive deterrence and countering strategies that are effective on the short and long term. A Complex system perspective offers new insights into some of the underlying mechanisms of conflict and violence at all their different levels. For instance, at an environment (macro) level identical patterns of violence in different conflicts can be identified. At an organizational (meso) level, modelling of the behaviour of guerrilla groups provides insight into effective strategies to counter the group activities, while at individual (micro) level models have been developed to achieve insight into individual behaviour and interventions to influence it. Each of these approaches (environmental, organizational, individual) brings different perspectives to the table. However, in order to derive effective strategies in the short and long run it is essential to connect all these three levels and to model the interactions among them. Here for we will explore the benefits of a generic multi-methodology framework combining Agent-Based Modelling and System Dynamics.

ABSTRACT. Many living and non-living complex systems can be modeled and understood as collective systems that
consist of heterogeneous components that self-organize and generate nontrivial morphological
structures and behaviors. In this talk, we propose a theoretical model of four distinct complexity levels of
such morphogenetic collective systems, according to the nature of components and their interactions.
We conducted a series of computational experiments using heterogeneous self-propelled particle swarm
models to investigate the effects of (1) heterogeneity of components, (2) differentiation/re-differentiation
of components, and (3) local information sharing among components, on the self-organization of the
system.
Results showed that (a) heterogeneity of components had a strong impact on the system's structure and
behavior, (b) dynamic differentiation /re-differentiation of components and local information sharing
helped the system maintain spatially adjacent, coherent organization, (c) dynamic differentiation /
re-differentiation contributed to the development of more diverse structures and behaviors, and (d)
stochastic re-differentiation of components naturally realized a self-repair capability of self-organizing
morphologies. We also explored evolutionary methods to design novel, non-trivial self-organizing
patterns, using interactive evolutionary computation and spontaneous evolution within an artificial
ecosystem. We also demonstrate that these self-organizing swarm systems are remarkably robust
against dimensional changes from 2D to 3D, although evolution works efficiently only in a 2D setting.

ABSTRACT. Individual cells within multicellular tissues communicate in order to self-organise into complex organs. In
plants, development is continuous and modular, where new tissues arise from groups of self-organising
stem cells in the meristem. Plant cells cannot move, so physical interactions with their neighbours are
fixed. To model the role of these associations, a complex systems approach to capturing, modelling and
predicting the self-organising outputs of multicellular organisation was used. Using live 3D imaging and
image analysis, cellular connectivity networks were extracted, describing all cell-cell interactions in the
meristem. Network analysis was used to identify features in the cell connectivity network, and revealed a
counter-optimised global topological feedback across the multicellular system, regulated through cell
division.
A classical study of cell division by Errera in 1888, shows that the shortest wall that bisects the cell
equally is often observed. In a comparison to this rule, predictions of division plane orientation were
made using topology based division rules, and simulated topological cell divisions mostly conformed to,
or outperformed Errera’s rule. The local geometric property of individual cell divisions in fact encodes a
global topological property of the multicellular system.

ABSTRACT. In evolutionary biology, the expression “survival of the fittest” is often heard as a summary of Charles Darwin’s idea of natural selection. This idea has definitely helped us understand how life is shaped by variation (and selective survival), but even Darwin himself acknowledged our lack of understanding of what causes all the variation he observed. Hugo de Vries, one of the first geneticists, famously said: “Natural selection may explain the survival of the fittest, but it cannot explain the arrival of the fittest.” More recently, genomics and bioinformatics have added pieces to the puzzle: there is large redundancy in the genome, caused by neutral mutations, implying that multiple genotypes can produce the same phenotypes. That naturally raises questions about how genotypes are distributed over phenotypes, and about biases in that distribution.

In this work, we address these questions using computational models for gene regulatory networks. In particular, we look at the gene network that regulates the fission yeast cell cycle. By working with a coarse-grained model of this gene network, we find that the design space of gene networks has a large bias in the distribution of genotypes mapping to phenotype, which is related to properties such as mutational robustness and evolvability. Moreover, we find that this bias is can also be characterised by applying concepts from algorithmic information theory, such as Kolmogorov complexity and Levin’s coding theorem, which suggest that the most likely phenotypes will be the ones with lower complexity.

ABSTRACT. The purpose of this presentation is to describe an approach to research design that can be used to guide complex systems research in educational contexts. Theoretical postulations in education often describe interaction dominant phenomena with complex, dynamic, and emergent properties. However, within a linear deterministic paradigm, these properties are often left out of theoretical models, which utilize mechanical assumptions that are conducive to statistical tests. In this presentation we will define the ontological differences between interaction dominant models (i.e. those used to model complex systems) and component dominant models (i.e. those used to model mechanical systems). We then discuss some analytical possibilities for drawing conclusions about interaction dominant systems. We end with some existing examples that help to illustrate the techniques. A complex system is a collection of interacting components that gives rise to complex behavior (Mitchell, 2009). In education contexts system components can take material, conceptual, or semiotic forms such as individual students teachers and technological objects, motivation, behavioral, affective, epistemological, and cognitive variables, or words, text, symbols, and discourses (Bunge, 2004). Components within complex systems interact over time to produce outcomes at higher levels of analysis that are more than the sum of their parts, meaning the complex behavior cannot be reduced to the components that make up the system (Holland, 2006). ​ Interaction dominant systems (i.e. those that are complex) are based on the philosophical notion that outcomes emerge from the coordination of system components across temporal scales (Kaufmann, 1993). Although the system cannot be reduced, critical indicators can be identified that give insight into the overall functioning of the system. The relationship among the system components is considered nonlinear, meaning that the strength and direction of the relationships change over time and are dependent upon other system components. Supervenience describes system maintenance, where emergent states influence the equilibrium of interactions among system components. Because all system components are dependent upon one another, the dynamic behavior of critical indicators over time contains information about the entire system (Takens, 1981). In contrast, a component dominant system can be precisely defined by its components, so that outcomes can be nearly perfectly reduced (Holden, Van Orden, & Turvey, 2009). Component dominant systems are based on the philosophical notion that a system can be reduced to components that adequately describe it. Further, the relationship among the components is considered linear, meaning that the strength and direction of the relationships are stable across time, or nomothetic (Hempel, 1965). Feedback loops describe system maintenance, where independent variables influence dependent variables in a cyclical fashion. Because all system components are considered independent of each other, the behavior of the systems is defined by a linear, deterministic causal mechanism (Chronbach & Meehl, 1955). ​ Theory building and testing are accomplished through the development of two forms of related submodels: a) the conceptual or theoretical model and b) the statistical model. These two models are considered a simplification or approximation of more complex conceptual or social systems (Sloane, 2006). This presentation will lay out the ontological assumptions of interaction dominant model building for complex systems research in a way that is useful for education researchers.

References Bunge, M. (2004). Clarifying some misunderstandings about social systems and their mechanisms. Philosophy of the social sciences, 34(3), 371-381. Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological bulletin, 52(4), 281. Hempel, C. G. (1965). Aspects of scientific explanation and other essays in the philosophy of science. New York. Holden, J. G., Van Orden, G. C., & Turvey, M. T. (2009). Dispersion of response times reveals cognitive dynamics. Psychological Review, 116(2), 318. Holland, J. H. (2000). Emergence: From chaos to order. OUP Oxford. Kauffman, S. A. (1993). The origins of order: Self organization and selection in evolution. Oxford University Press, USA. Mitchell, M. (2009). Complexity: A guided tour. Oxford University Press. Sloane, F. (2006). Normal and design sciences in education: Why both are necessary. Educational design research, 19-44. Takens, F. (1981). Detecting strange attractors in turbulence. In D. A. Rand and L.-S. Young. Dynamical Systems and Turbulence, Lecture Notes in Mathematics, vol. 898. Springer-Verlag. p. 366–381.

ABSTRACT. Our research concerning impact of musical learning on broader educational, cognitive, and social progress supports our hypothesis that the brain has variety of strategies for developing mental skills. Some are initially easier to develop but limit the extent of skill development in a particular area of application until they are augmented or replaced. Implications for education will be discussed. For physical and mental skills of all kinds to advance, the brain must form complex dynamical engagement systems to address dynamically demands that cannot fully be anticipated in advance. Our research provides some evidence concerning the strategic formation of such systems.

ABSTRACT. .

ABSTRACT. In this paper, we show both theoretically and empirically that the size of over-the-counter (OTC) markets can be reduced without affecting individual net positions. First, we find that the networked nature of these markets generates an excess of notional obligations between the aggregate gross amount and the minimum amount required to satisfy each individual net position. Second, we show conditions under which such excess can be removed. We refer to this netting operation as compression. Third, we apply our framework to a unique and comprehensive transaction-level dataset on OTC derivatives including all firms based in the European Union.

ABSTRACT. David Wolpert: “People sometimes refer to physical systems undergoing Markovian dynamics as “processing information” (e.g., in statistical physics) and sometimes even refer to them as “performing computation” (e.g., in biology). Arguably though, unless we have theorems that apply to physical systems that “perform computation” but do not apply to arbitrary Markov processes, it is vacuous to use the term “computation” to describe the dynamics of physical systems. What might such theorems be?”

Carlos Gershenson: “Everything can be described in terms of information. Thus, information is a promising framework to explore general principles in science. Nevertheless, there is no agreement on what information is. The evolution of life, cognition, intelligence, and consciousness can be generalized in terms of the evolution of information. This prevents us from falling into a dualist trap. Moreover, we can gain insights into the evolution of complexity.”

Gershenson, C. (2012). The world as evolving information. In Minai, A., Braha, D., and Bar-Yam, Y., editors, Unifying Themes in Complex Systems, volume VII, pages 100–115. Springer, Berlin Heidelberg. http://arxiv.org/abs/0704.0304

Mile Gu: “Complexity depends on what sort of information theory we use. The discovery of new physics (e.g. quantum mechanics) and lead to new methods of processing information,  and this can in turn, fundamentally change what we perceive to be complex.”

ABSTRACT. There are over 7 billion people in the world, each with a different brain containing 15 to 33 billion neurons. These people are intelligent entities who develop and change over time and who interact with each other in complicated social structures. Consequently, human behavior is likely to be complex. Lines of research have recently emerged that conceptualize behavior as complex systems of interacting psychological, biological, and sociological components, characterized by the use of dynamical systems and network models. This conceptualization allows for novel insights in psychological phenomena: phenomena, such as the occurrence of symptoms, can be considered emergent behavior in a complex system, making it possible to investigate symptom based therapy, attractor states (e.g., a depression) and critical transitions (e.g., relapse into substance abuse).

This presentation will provide a conceptual introduction to the topic of this symposium and will discuss four questions of this emerging field: (1) Can we estimate the structure of psychological dynamics (networks in which nodes are variables and edges are statistical relationships). (2) Does psychological behavior feature attractor states and critical transitions? (3) Are attractor states resilient to change?, and (4) How can we assert control over the complex system? For the last topic, we introduce a new direction in psychological complexity research: process modeling, which will form the overall theme of the remainder of the symposium.

ABSTRACT. The goal of psychopathological research is to find interventions that move patients in a state without symptoms. Recently network models of symptom data have been proposed to generate possible interventions and judge their effectiveness. While these models are attractive because they can be estimated directly from symptom data, they have two downsides: first, symptoms may not be the ideal level of analysis because they are summaries of more fundamental variables. Second, the expressiveness of these network models does not match the complexity of psychological disorders. Here we present first steps towards a realistic computational model of major depression based on coupled differential equations. We begin by simulating a functioning individual and then discuss how to formalize different etiologies of depressive symptomatology. In addition, we discuss how to support the model with data and how to maximize its usability for other researchers.

ABSTRACT. Education invariably ends with an exam. About a century ago it became possible, and popular, to administer the same exam to large numbers of learners at once. This paved the way for the statistical analysis of the responses of learners to exam questions. A new field in the statistical sciences emerged, called psychometrics. Psychometric models take the form of a weighted directed bipartite graph comprised of an observable on the head side and one of a small number of unobserved “explanatory” variables on the tail side of the arrow. These models are eerily accurate at describing the outcomes of learning. Regardless of the age of the learners, their geographic location, the content area, a weighted directed bipartite graph with a hidden layer does the trick. Sure, the weights are different at different ages, the number of hidden nodes depends on the content, etc. But a century of analyzing exam data has found that there always is such a weighted directed bipartite graph with a hidden layer.

Why this should be the case is a question that is not only not answered but is even not asked. In this presentation we’ll deal with exactly the “why” question, and how “it” emerges from the interaction of what we teach (can’t change that) and how we teach it (could change that).

The reason for doing so is not without importance. In a psychometric model either the weights or intercepts change (which effects all learners at once), or the value of hidden nodes change (which effects all questions at the head of its arrows at once). That is not how learning nor teaching happens, and consequently has little to offer to learners or teachers. We propose an alternative view that a) does justice to the fact that weighted directed bipartite graphs with a hidden layer fit the data to (near) perfection, yet b) opens up the way to make it preventable through teaching.

ABSTRACT. A network of dynamical units can generate a complex systemic behavior. Examples include human cognition emerging from a network of neural cells, ecosystems from food webs, and cellular regulatory processes from protein-protein interactions. A first important question is: which agents are the ‘drivers’ of the systemic behavior? A second question is: can we detect emergent phenomena, particularly ‘criticality’ (susceptibility to small perturbations)? We address these questions using the concept of ‘information dissipation’ which we are developing. This is the idea that Shannon information is first stored in an agent’s state, and then percolates through the network due to the agent-agent interactions. One of the main findings is that the most influential information spreader are not the most well-connected nodes. I will present recent work on addressing the above questions through analytical results, computational modeling, and a preliminary analysis of a psychological symptom network.

ABSTRACT. Kubernetes is a fiction “edudrama” about the past, present and future of Cybernetics. The writer and producer of the film was a very close friend and disciple for more than 20 years of Stafford Beer, the creator of Management Cybernetics and author of many groundbreaking books, among them The Brain of the Firm and The Heart of Enterprise. Stafford Beer was also the chief scientist behind the creation of PROJECT CYBERSYN in the early 1970’s. The project died with the abrupt ending of Salvador Allende’s tenure as President of Chile in September 11th, 1973.

In this film, many of Stafford Beer’s ideas are discussed, including a look at the iconic “operations room”. The plot has ramifications to the 1970’s as a group of beautiful minds with very different cybernetic backgrounds are invited to a meeting to find a way to change the world by changing the way organizations of all types operate. Each one of the organizer’s guest brings a statement to the meeting. Once there, they ask questions about the issue of complexity, religion, government, Newtonian science, the purpose of human beings, criticize organizations and reflect on the possible existence of God. 

The screening will be followed by a Q&A session with the writer and producer of Kubernetes.

[Mexico, 2017, 95 min. In Spanish with English subtitles]