The U.S. military has made a significant investment in fielding a wide variety of airborne and ground Full-motion Video (FMV) electro-optical and infrared sensors to provide superior situational awareness and persistent surveillance of the battlefield. These sensors collect an increasingly unmanageable amount of data, up to terabytes per hour from a single wide area motion imagery sensor. Even with conventional FMV sensors, the data being produced far exceed the number of intelligence analysts available to manually exploit the data. Together, the U.S. Army Communications-Electronics Research, Development and Engineering Center, U.S. Army Intelligence and Information Warfare Directorate (I2WD), and the Joint Improvised Explosive Device Defeat Organization (JIEDDO) are working to address this operational need. The project to provide an initial material capability to meet these requirements is named Advanced Video Activity Analytics (AVAA). The AVAA is maturing a video processing exploitation framework (VPEF), a video data model (VDM), a video annotation web service (VAWS), and integrating computer vision analytic algorithms as plug-ins. The framework provides standardization, integration, and parallelization of computer vision algorithms (CVAs), making them interoperable and testable. The system processes large-scale data and manages the results using a video data model. This paper describes the formulation for testing and evaluation conducted at the Army Intelligence Center of Excellence at Fort Huachuca, AZ, to measure AVAA's ability to improve video data processing and to reduce the cognitive load on analysts while providing the building blocks for improved knowledge discovery across Intelligence domains. The techniques can be applied to understand and refine cognitive load on training. Quickly processed full-motion imagery data can also facilitate population of simulation data for an experimentation or training event.