Presented at the International Conference on Applied Human Factors and Ergonomics (AHFE), Las Vegas, NV (July 2015).

Research in narrative science across academia and industry continually uncovers new insights that have relevance to understanding how narratives affect behavior change in regions around the world. These insights are valuable to military decision-makers and analysts who seek to understand how adversaries create and use narratives to alter population behaviors. Unfortunately, the transfer of findings from research institutions to operational settings is often slow, indirect, and fraught with challenges (Pfautz & Koelle, 2014). Research results are most frequently expressed in scientific papers that tend to circulate among researchers, but not with tactical communities. We are developing a technique that facilitates the creation of models that represent the relationships among key research findings by leveraging concepts from user-centered modeling (Pfautz, Carlson, Koelle et al., 2009) and informed by prior work in model composability (Davis & Anderson, 2004). This technique supports the creation of models by researchers as well as operational users. We have uncovered several key insights in our development of this modeling technique. First, qualitative abstract-level summaries of research results are often useful on their own, as they provide descriptive information that can explain key variables and their interactions. Second, the use of well-defined qualitative terms enables a form of composability for summary models, which has the potential to enrich the larger body of research by connecting related findings. Third, the explicit capture of context within the summary models provides a means for recognizing or specifying the conditions under which a research finding may be applicable (e.g., with respect to culture, geographic location, mission objective, or intended audience). Fourth, enriching summary models with the ability to support higher model fidelity (i.e., deeper levels of detail about relationships in the model) allows for the representation of conflicting evidence and conflicting summary models, as well as connection to quantitative data and data-driven models. These summary models, represented as heterogeneous networks, can be used by different types of tactical tools to satisfy a variety of goals, such as analyzing an adversary’s narratives, decomposing a specific adversary message, and assessing the biophysiological response to a narrative. Finally, results from applying research in operational environments can be fed back to the model context and could potentially provide researchers with data on the real-world implementation of concepts from the academic community.