Research Grants

Computer Simulations in the Humanities

This project will advance research in the humanities by adding a variety of simulation techniques to the standard repertoire of methods already employed by humanists. Interested humanists from a range of disciplines including philosophy, history, archeology, linguistics, anthropology and political science, among others, will work not only with technical experts but also with humanists already familiar with methods involving computer simulations and models. Our aim in bringing technologists and humanists together in precisely this way is to promote the dual notion of “the humanities shaping technology” as well as “technology shaping the humanities.” Modeling experts will be pressed to not merely present existing techniques but to shape those techniques in ways that address questions and on-going inquiries pursued by humanists.

To make this institute be of genuine value to both humanists and modeling experts, time will be divided between hands-on training with existing modeling techniques, design sessions discussing how these models can be improved to address humanist questions/projects, and general presentations on the how to enhance the interaction between technology and the humanities. Thus, the institute will bring together 24 humanists from a variety of disciplines to gather daily over three weeks at the beginning of June, 2011 to learn from talks and discussion each morning, work on models in the afternoon, and attend evening lectures. The topics will range over various aspects of the humanities-technology interaction and the nature of collaborative, interdisciplinary research within the humanities. This project will run during the first three weeks of June, 2011 and will include a 3-day follow-up session in 2012.

Marvin Croy, Mirsad Hadzikadic, Tony Beavers, Patrick Grim; NEH, $158,225.

Back to Top

2010 AAAI Complex Adaptive Systems Symposium

Our goal for this year’s symposium was to address the phenomena underlying the process of designing non-trivial, real-life complex adaptive systems (CAS) applications. It is generally accepted that designing CAS applications is more of an art than a science. This is due to the fact that emergence and self-organization, the key properties of nonlinear systems, cannot be built into the systems in a top-down fashion. Rather, they emerge from bottom-up processes. As such, they seem more like a phenotype than a genotype of a particular species. As nature taught us, the most beneficial designs are arrived at through the painstaking process of candidate design generation and subsequent evaluation in the environment in which the candidate design is intended to operate.

This symposium focused on the fundamental aspects of complex adaptive systems, primarily human/social and ecological systems, and provided the framework for addressing fundamental questions. The program committee of the symposium decided to start this inquiry by focusing on resilience, robustness, and evolvability. Our assumption is that the properties of resilience, robustness, and evolvability will help us address the issues of designing best models for studying complex systems, understanding the relationship between systems’ structure and the resulting emergent behaviors, understanding the processes of their (systems’) change (evolution), and system validation and verification.

Complex Adaptive Systems have proven to be a powerful tool for exploring these and other related phenomena. We characterize CAS systems as having a significant number of self-similar agents that:

  • utilize one or more levels of feedback
  • exhibit emergent properties and self-organization
  • produce nonlinear dynamic behavior

For some practitioners in the field the terms “resilience” and “robustness” may seem largely redundant. Indeed, there are many other terms from various domains that possibly overlap as well: from “basins of attraction” (physics, mathematics), to “homeostasis” (biology), to “sustainability” (ecology), to economics (equilibrium). This is precisely the point: different disciplines often have their own language, even as they are describing identical or similar phenomena.

Issues of robustness, resilience, and evolvability are critical for understanding complex systems. Furthermore, communicating the properties of such systems across domains allows researchers to gain insight from fields that they normally may not encounter. As such, this symposium was designed to include organizers and participants representing expertise in a wide variety of natural, physical, social, and virtual/artificial systems.

Back to Top

ACSES: Actionable Capability for Social and Economic Systems

At an abstract level, a society can be characterized as a collection of open, nonlinear, dynamic networks of imperfectly rational agents having access to partial information and providing both positive and negative feedback. Such systems typically exhibit properties such as emergence, path-dependence and self-organization, and may possess multiple equilibria, cycles, or no stable states at all. This makes analysis using mainstream, traditional approaches in social science inadequate, for these approaches almost always assume that the phenomenon under investigation reaches a single equilibrium through a deterministic process. Outside of social science, the 20th and 21st centuries have seen tremendous gains in the scientific understanding of dynamic, nonlinear systems. In the past few decades this has accelerated with the increases in computing power that have made sophisticated computer simulations of complex processes feasible, for example, in physics (e.g., plasma and flow processes), biology (e.g., cellular processes), and applied sciences such as meteorology. Despite the apparent complexity of social processes, however, progress in the understanding of nonlinear social systems and, specifically, in the use of computer simulation as a tool for investigating them, has lagged.

The United States Government is interested in obtaining a capability to analyze, model, and simulate the social fabric of a country in order to determine the best ways to interact with that country. Such a capability involves developing a high-level modeling and simulation system that takes into account a complex network of social, economic, political, cultural, religious, demographic, historical, and geographical factors.

Over the course of the ACSES project, funded by DARPA, we have designed tools that embody the characteristics of nonlinear, dynamic systems outlined above. These tools as part of the agent-based system (ABS) model of Afghanistan reflect the strides we have made towards the ultimate goal of a rigorous and actionable modeling paradigm. Such a modeling paradigm must possess two key attributes: 1) The system must allow for multiple theoretical models of social theory to be integrated together in ways that are instantiated, driven, and ultimately validated by hard static and time-series data pertaining to the country or region of interest; and 2) commanders must be able to understand, trust, and gain insights from the model’s outputs, such that causal factors are transparent and various Diplomatic, Information, Military, and Economic (DIME) options can be evaluated. The ACSES project is the first step towards achieving such modeling capability.

Back to Top

2009 AAAI Complex Adaptive Systems Symposium

Most interesting phenomena in natural and social systems include constant transitions and oscillations among their various phases. Wars, companies, societies, markets, and humans rarely stay in a stable, predictable state for long. Randomness, power laws, and human behavior ensure that the future is both unknown and challenging. How do events unfold? When do they take hold? Why do some initial events cause an avalanche while others do not? What characterizes these events? What are the thresholds that differentiate a sea change from a non-event?

Complex Adaptive Systems have proven to be a powerful tool for exploring these and other related phenomena. We characterize a general CAS model as having a large number of self-similar agents that: 1) utilize one or more levels of feedback; 2) exhibit emergent properties and self-organization; and 3) produce nonlinear dynamic behavior. Advances in modeling and computing technology have led not only to a deeper understanding of complex systems in many areas, but they have also raised the possibility that similar fundamental principles may be at work across these systems, even though the underlying principles may manifest themselves differently.

Threshold effects are found all around us. In economics, this could be movement from a bull market to a bear market; in sociology, it could be the spread of political dissent, culminating in rebellion; in biology, the immune response to infection or disease as the body moves from sickness to health.

The goal of this – and future related symposia – is to bring together researchers from diverse fields who study these complex systems using the tools and techniques of CAS. The 2009 symposium focused on threshold effects in various disciplines as one avenue towards exposing common dynamics that are found in these disparate domains. In the past, knowledge gained in each domain about threshold effects has remained mostly exclusive to that domain, especially when the disciplines are far apart. It is our belief that by bringing together scholars who study these phenomena, we can leverage a deep knowledge of one domain to gain insight into others.

Back to Top

Marine Ecosystem

One of the foundations of ecology dynamics is the Lotka-Volterra equations for predator-prey populations. These equations, while mathematically robust and widely accepted, are general in nature. Thus, they are limited by the assumptions imposed upon them, including, for example, the assumption of unlimited resources available to the prey population. This project will bring together computer scientists, biologists, environmental researchers, mathematicians, economists, and industry experts in order to: 1) create computer simulations of multiple trophic levels in a general marine ecosystem model; 2) use these simulations to better understand the complex dynamics of the Lotka-Volterra equations in a more realistic setting; 3) conduct both controlled experiments and real-world studies in order to validate the results of these computer models; and 4) use these results to suggest more robust economic models of sustainable industrial practices.

Back to Top

Nanoperception: The Public Perception of Nanotechnology

This project encompasses an analysis of the public perception of nanotechnology as a complex system – a system thus labeled nanoperception. While nanoperception is certainly influenced by the science itself, there are many other agents that have as much or more impact on the system. From military to commercial to aesthetic interests broadly defined, the perception of nanotechnology is truly a system that exists on the edge of chaos in that tense zone between order and disorder. The interaction of each of the agents help to create nanoperception as a self-organizing, quasi-stable pattern – identifiable but evolving, intelligible but not necessarily predictable. Thus, the interdisciplinarity required of complexity science and of this project in particular is a recognition of the fact that in order to grasp nanoperception one must reach beyond separate agents or influences to gain a true appreciation of the overall behavioral pattern of the system. While each agent by itself could provide the basis for fruitful research, nanoperception is yet another example in complexity science of the whole being greater than simply the sum of its parts.

Back to Top

Hospital Environment Simulation

Hospitals represent a complex environment for patient care, with many variables affecting patient outcomes. These variables can often interact in surprising ways, producing nonlinear effects in a dynamic environment. Complex Adaptive Systems techniques can be used to design an Agent-based Model for simulating these systems, allowing for a meaningful exploration of large datasets related to patient outcomes. Furthermore, since every hospital environment is different, a well-designed simulation tool should allow an operator to input various theories of patient care. In this way, the operator can tailor an experimental design that more precisely fits the environment being studied, while still benefiting from the vast amount of data available across many different environments.

Back to Top


This project involves developing, testing, and deploying a general Complex Adaptive Systems model as a new technology for supporting human creativity. Our general CAS model will use an agent-based paradigm to develop a common language for exploring and describing complex phenomena in various domains. By constraining these phenomena within the paradigm of CAS and agent-based systems, and using this general model in research and education, we believe that a deep knowledge of one domain can inform understanding of similar phenomena in another domain, even when the disciplines are far apart. In this way, multiple concepts from disciplines that do not generally interact at such a fundamental level can be synthesized into new forms for understanding, researching, and communicating novel ideas.

Back to Top

Hyperlocal Community Platform

The world is shrinking. Ease of travel, cheap communication, Internet connectivity: all have allowed us to broaden our world-view and interact globally as never before. Some believe this has led to a decline in local participation and engagement. Newspapers, once the glue that held a community together, are struggling, and the growth of interest-based social media seem to indicate that geography is no longer important. We believe, however, that geographic community is as important now as ever before. It remains the center of our public life, and the quality of our local environment greatly impacts the quality of our lives. Thus, this project aims to do two things: 1) to build a web platform that will serve as local hubs for the community, encouraging the emergence of self-organized groups that are defined by geography as much as topical interest; and 2) to use the tools of Complex Adaptive Systems to model and study these emergent clusters, so that these local communities are transparent and understandable to the users, allowing for positive feedback and influence towards even greater local engagement.

Back to Top