Facilitating the Composition of Real-Time Cognitive State Inference Applications

Stottler Henke proposes to develop the Adept system, a novel “plug-and-play” (i.e., low-/no-code) cognitive state assessment software toolkit comprising a highly flexible and extensible system architecture with intuitive graphical user interfaces to readily integrate neural, physiological, and behavioral data stream analysis across a wide range of operational settings.  First, we will pursue a multi-pronged approach to facilitate interoperability with external devices/software, including automated plugin code generation and configuration wizards to minimize the level of required technical expertise.  Second, we will establish a no-code programming environment that guides a user through the process of composing a cognitive state assessment pipeline.  Finally, our approach includes out-of-the-box, device-agnostic support for common signal processing functions used in human state assessment, along with a toolbox of common components for use when enacting new cognitive state assessment applications.Construction of proof-of-concept prototypes during Phase I has answered research questions, reduced technical risk, and established a strong technical foundation for the continued development of Adept.  By applying the lessons learned and software developed during Phase I, we will demonstrate a fully functional, integrated Adept prototype by the end of Phase II, paving the way to transition the technology into one or more civilian or DoD programs.