Jonathan K. Alt

The Cultural Geography Model: Evaluating the Impact of Tactical Operational Outcomes on a Civilian Population in an Irregular Warfare Environment:

The civilian population has been described as ‘the center of gravity in irregular warfare’. Understanding the behavioral response of the civilian population in irregular warfare operations presents a major challenge area to the joint modeling and simulation community where there is a clear need for the development of models, methods, and tools to address civilian behavior response. This paper provides a conceptual and theoretical overview of the Cultural Geography (CG) model, a government-owned, open-source agent-based model designed to address the behavioral response of civilian populations in conflict environments. With an embedded case study, we describe the development of cognitive modules to represent the civilian population and their implementation as Bayesian belief networks (BBNs), the social structure module implemented using homophily, the process of adjudicating the effects of tactical level outcomes on a population segment within the model, and a sample case study analysis using a designed experiment.

Developing cognitive models for social simulation from survey data

The representation of human behavior and cognition continues to challenge the modeling and simulation community. The use of survey and polling instruments to inform belief states, issue stances and action choice models provides a compelling means of developing models and simulations with empirical data. Using these types of data to population social simulations can greatly enhance the feasibility of validation efforts, the reusability of social and behavioral modeling frameworks, and the testable reliability of simulations. We provide a case study demonstrating these effects, document the use of survey data to develop cognitive models, and suggest future paths forward for social and behavioral modeling.

Representing dynamic social networks in discrete event social simulation

One of the key structural components of social systems is the social network. The representation of this network structure is key to providing a valid representation of the society under study. The social science concept of homophily provides a conceptual model of how social networks are formed and evolve over time. Previous work described the results of social simulation using a static homophily network. In order to gain the full benefit of modeling societies a representation of how the social network changes over time is required. This paper introduces the implementation of a dynamic homophily network, along with a case study exploring the sensitivity of model outputs to the parameters describing the network and applying social network change detection methods (SNCD) to model output.

Developing social networks for artificial societies from survey data

Authentically representing large social collectivities remains a preeminent challenge throughout the social computing, and modeling and simulation communities. We demonstrate here a simple technique that uses survey and polling data to embed agents with attributes and endogenously elicit an authentic and theory-driven simulation social structure for an artificial society. We furthermore show that a representation of social structure based on internal agent attributes allows for the continuous representation of social dynamics that affect agent cognition and association, and that social structures for artificial societies can be generated without any loss to the granularity of the underlying data or simulation output. We provide a case study using social survey data to demonstrate the method and effects, document the visualization of social structure for the population of Indonesia, discuss the implications and uses of survey data for social simulation, and suggest several paths forward for social and behavioral predictive modeling.

Behavior Selection Using Utility-Based Reinforcement Learning in Irregular Warfare Simulation Models

The Theory of Planned Behavior TPB provides a conceptual model for use in assessing behavioral intentions of humans. Agent based social simulations seek to represent the behavior of individuals in societies in order to understand the impact of a variety of interventions on the population in a given area. Previous work has described the implementation of the TPB in agent based social simulation using Bayesian networks. This paper describes the implementation of the TPB using novel learning techniques related to reinforcement learning. This paper provides case study results from an agent based simulation for behavior related to commodity consumption. Initial results demonstrate behavior more closely related to observable human behavior. This work contributes to the body of knowledge on adaptive learning behavior in agent based simulations.

Representing trust in cognitive social simulations

Trust plays a critical role in communications, strength of relationships, and information processing at the individual and group level. Cognitive social simulations show promise in providing an experimental platform for the examination of social phenomena such as trust formation. This paper describes the initial attempts at representation of trust in a cognitive social simulation using reinforcement learning algorithms centered around a cooperative Public Commodity game within a dynamic social network.

A use-case approach to the validation of social modeling and simulation

The modeling and simulation (M&S) community is faced with the task of informing public policy decision makers, from both the defense community and from the civilian sector, of the impact of their decisions on the beliefs, values and interests (BVIs) of the populations in their areas of influence. The M&S techniques used for these types of analyses by necessity challenge the traditional boundaries and methods regarding validation efforts. Given that the analysis of populations is largely intractable via reductionist methodologies, we posit that insight must be garnered through experiment and iteration using simulated societies. Further, the designs of these simulation experiments must be based on the information needs of the community. We discuss the concept of developing social simulations by use case, the validation of data sources for model development, validation techniques guided by usefulness, and a case study approach to validation of social simulations.

Visualizing the human, social, cultural and behavioral components of a complex conflict ecosystem

Creating meaningful visualizations of multi-dimensional human, social, cultural, and behavioral (HSCB) data will provide greater insights to operational decision makers across a large variety of problem domains. Developing and deploying visualization tools presents a variety of challenges to the analytic community, and these are further compounded when presenting information on the non-traditional dimensions of the battlefield encompassed by HSCB domain. Given that the center of gravity in Irregular Warfare is the population, the need for battlefield commanders to understand this information in operationally relevant ways is clear. Here we provide an overview of the challenges in providing visual displays of HSCB data for decision support, and the methods chosen for communicating the output of a social simulation. We use the Cultural Geography model as an example of a social simulation with accompanying visualizations, discuss the importance of display configurations, and propose several paths forward for future work in HSCB data visualization.

A practical situation based agent architecture for social simulations

The concept of situation is central to the decision making processes of both human and software agents. The recognition of situation facilitates decision processes that ultimately result in action selection. Cognitive agent architectures that incorporate the concept of situation provide the opportunity for more sophisticated representations of human behavior and for more sophisticated decision support applications. This paper provides an overview of a general cognitive architecture for use in multi-agent simulation with the concept of situation central to the action selection and decision making process.

Iowa Gambling Task modified for military domain.

One key component of optimal military decision making is that the decision maker demonstrates reinforcement learning. The modification of psychological tasks gives insight into understanding how to effectively train military decision makers and how experienced decision makers arrive at optimal or near optimal decisions. We developed a task modeled after the Iowa Gambling Task (IGT) to measure military decision making performance. This new task focuses on high stakes and uncertain environments particular to military decision making conditions. Thirty-four U.S. military officers from all branches of service completed the tasks yielding decision data for validation. The new task retains essential characteristics of the foundational task and gives insight into reinforcement learning of military decision makers. Results indicate that the additional metric of regret defines higher performance at a trial-by-trial level, and clustering by multiple metrics defines high performance groups.