The construction of computational causal models for complex systems has typically been completed manually by domain experts and is a time-consuming, cumbersome process. Operational design is a method of structured team discourse used by military planners for rapidly envisioning complex systems and relationships; however, the products are typically static diagrams on whiteboards or slides. DARPAs Causal Exploration program seeks to leverage artificial intelligence (AI) assistance and causal analytics to enable rapid system modeling and analysis. We introduce Causeworks, an application in which operators “sketch” complex systems, leverage AI tools and expert knowledge to transform the sketches into computational causal models, and then apply analytics to understand how to influence the system. We walk through human–machine collaborative model building using Causeworks and discuss feedback and lessons learned about how to flexibly apply causal modeling and thinking for expert planners that are novice modelers.

Military planners use “Operational Design” (OD) methods to develop an understanding of systems and relationships in complex operational environments. Here, we present Causeworks, a visual analytics application for OD teams to collaboratively build causal models of environments and use analytics to understand and find solutions to affect them. Collaborative causal modelling can help teams craft better plans, but there are unique challenges in developing synchronous collaboration tools for building and using causal models. Collaboration systems typically organize information around varying degrees of synchronization between data “values” and user “views.” Our contribution is in extending this collaboration framework to include analytics as layers that are by nature derived from the data values but utilized and displayed temporarily as private views. We describe how Causeworks overlays analytics inputs and outputs over a shared causal model to flexibly support multiple modeling tasks simultaneously in a collaborative environment with minimal state management burden on users.

Causal Model building for complex problems has typically been completed manually by domain experts and is a time-consuming, cumbersome process. Operational Design defines a process of rapid, structured discourse for teams to envision systems and relationships about complex, “wicked” problems, however, the resulting models are simple diagrams produced on whiteboards or slides, and as such, do not support computational analytics, thus limiting usefulness. We introduce CauseWorks, an application that helps operators “sketch” complex systems and transforms sketches into computational causal models using automatic and semiautomatic causal model construction from knowledge extracted from unstructured and structured documents. CauseWorks then provides computational analytics to assist users in understanding and influencing the system. We walk through human-machine collaborative model-building with CauseWorks and describe its application to regional conflict scenarios. We discuss feedback from subject matter experts as well as lessons learned.

Collaborative interactions and transactions among traders, portfolio managers and risk managers can be transformed with synthetic worlds. It is easy to anticipate that by 2025, the reports, analyses and projections from financial industry participants exist in a shared workspace for dissemination and collaboration that is flexible, adaptable and with high visualization capabilities. This chapter describes a scenario with a futuristic portfolio analysis and planning process that is distributed in a global organization. It is made possible by powerful computer graphics hardware, advanced computational industry risk models and perceptually correct, cognitively enhancing, visual analytic user interfaces.

Humans are vulnerable to cognitive biases such as neglect of probability, framing effect, confirmation bias, conservatism (belief revision) and anchoring. Argument Mapper addresses these biases in intelligence analysis by providing an easy-to-use, theoretically sound, web-based interactive software tool that enables the application of evidence-based reasoning to analytic questions. Designed in collaboration with analytic methodologists, this tool combines structured argument mapping methodology with visualization techniques to help analysts make sense of complex problems and overcome cognitive biases. The tool uses Baconian probability and conjunctive logic to automatically calculate the inferential force on the upper level hypothesis. Evaluations with 16 analysts showed the tool was easy to use and easy to understand.

In this paper, we present in-progress work on “Influent”, a graph analysis tool that enables an intelligence analyst to visually and interactively “follow the money” or other transaction flow. Summary visualizations of transactional patterns and entity characteristics, a left-to-right semantic flow layout, interactive link expansion and hierarchical entity clustering enable Influent to operate effectively at scale with millions of entities and hundreds of millions of transactions, with larger data sets in progress.

Aperture is an open, adaptable and extensible Web 2.0 visualization framework, designed to produce visualizations for analysts and decision makers in any common web browser. Aperture utilizes a novel layer based approach to visualization assembly, and a data mapping API that simplifies the process of adaptable transformation of data and analytic results into visual forms and properties. This common visual layer and data mapping API, combined with core elements such as contextually derivable color palettes, layout and symbol ontology services is designed to enable highly creative and expressive visual analytics, rapidly and with less effort. This paper introduces the Aperture framework, describing key features of the programming API and reference implementation, presents example use cases, and proposes an approach for measuring technical performance metrics for software development, and operational performance metrics for visualization support of analysis and decision making.

In visual analytics, interactive data visualizations provide a bridge between analytic computations, often involving “big data”, and computations in the brain of the user. Visualization provides a high bandwidth channel from the computer to the user by means of the visual display, with interactions including brushing, dynamic queries, and generalized fisheye views designed to select and control what is shown. In this paper we introduce Visual Thinking Design Patterns (VTDPs) as part of a methodology for producing cognitively efficient designs. We describe their main components, including epistemic actions (actions to seek knowledge) and visual queries (pattern searches that provide a whole or partial solution to a problem). We summarize the set of 20 VTDPs we have identified so far and show how they can be used in a design methodology. Keywords—design patterns, visual thinking, data visualization, visual analytics.

New tools for raw data exploration and characterization of “big data” sets are required to suggest initial hypotheses for testing. The widespread use and adoption of web-based geo maps have provided a familiar set of interactions for exploring extremely large geo data spaces and can be applied to similarly large abstract data spaces. Building on these techniques, a tile based visual analytics system (TBVA) was developed that demonstrates interactive visualization for a one billion point Twitter dataset. TBVA enables John Tukey-inspired exploratory data analysis to be performed on massive data sets of effectively unlimited size.

Challenges with using graphs to visualize extremely large entityrelationship datasets include visibility, usability and high degree nodes. Visual aggregation techniques, tools and easily tailorable components are needed that will support answering analytical questions with data description, characterization and interaction without loss of information. We present two case studies of prototype implementations of JavaScript browser-based visualization tools leveraging the Louvain clustering algorithm. Two “big data” datasets were used to test aggregation of large networks to reveal communities and answer analytical questions.

Humans use intuition and experience to classify everything they perceive, but only if the distinguishing patterns are visible. Machine-learning algorithms can learn class information from data sets, but the created classes' meaning isn't always clear. A proposed mixed-initiative approach combines intuitive visualizations with machine learning to tap into the strengths of human and machine classification. The use of visualizations in an expert-guided clustering technique allows the display of complex data sets in a way that allows human input into machine clustering. Test participants successfully employed this technique to classify analytic activities using behavioral observations of a creative-analysis task. The results demonstrate how visualization of the machine-learned classification can help users create more robust and intuitive categories.

Adaptive user interfaces offer the potential to improve the learnability of software tools and analytic methodologies by tailoring the operation and experience to a user's needs. Scaffolding is an instructional strategy that can be applied by adaptive interfaces to achieve this. Scaffolding theory suggests that the level of guidance should be adjusted to optimize learning and performance levels. This paper explores the use of adaptive techniques to scaffold user interaction and presents a taxonomy of techniques for adaptive scaffolds within complex software systems. The techniques identified in the proposed taxonomy can help software scaffolds select appropriate adaptations in response to the user's learning and operating needs. A scaffold called nAble was implemented to explore the application of adaptive techniques from the taxonomy to support an analysis methodology called the Analysis of Competing Hypotheses (ACH).

This chapter explores the potential of synthetic worlds to help individuals learn and practice higher order thinking skills including critical and creative thinking. Such skills are difficult to teach in traditional educational settings and are rarely addressed in common simulation-based training programs. The potential is explored first through a fictional account of a day in the life of Tara Defarge, an eco-detective in the year 2025 and her interactions with the Cognitive Playground. Present day understandings about teaching, learning and virtual worlds are used to show that synthetic worlds naturally consist of appropriate metaphors and capabilities to foster critical thinking. An assessment of the state of intelligent interfaces and adaptive systems informs a discussion of the technological challenges that must be met before a Cognitive Playground can be realized.