ISR Decision Support Tools that More Naturally Support Both Intuitive and Analytical Modes of Thought

Recent technological innovations have converged to allow for the production of huge (and ever growing) volumes of ISR data. To address this challenge, we have witnessed steady investment into research that improves data fusion and sensor data analysis, but these efforts continue to be system-centric solutions, and do not substantially address the human element. We propose a focused research effort that builds on decades of experience in ISR data analysis, human-machine interaction design, decision support systems, and real-time data analytics. We will pursue three innovations. First, we will apply machine learning to the arduous task of automatically correlating events in a manner that reduces overall report volume, presents key data in context, and reduces the effort required to discover meaning. Second, we will utilize a cognitive systems engineering process to identify visual display and interaction designs that intelligently support multiple objectives including: facilitating pattern matching, representing interconnected data in natural semantic units, and providing tools that offload working memory. Third, we will develop a pragmatic approach to accumulating knowledge through work-serving interactions in order to adapt and improve underlying automation and human-machine interaction. Phase I will result in a proof-of-concept demonstration and evaluation, establishing a strong foundation for Phase II.