Information Technology Design for the Intelligence Analyst
Human-centered design principles should guide information presentation.
Increasing intelligence requirements and skillful collection technology have flooded the intelligence community with raw information. To address this problem, big data and artificial intelligence technologies aid the initial processing and exploitation of the information. But as technology continues to grow in capability, consumers must temper expectations regarding its impact on intelligence analysis.
Ensuring the human analyst is paramount in the end-user information design specifications must be a primary focus along with human-information interaction—the brain’s relationship to the information ecosystem provided by the information technology system and the cognitive art and science of intelligence analysis. Fundamental principles of human-centered information design should guide information technology design, and it is imperative that they be addressed early in system development. And end-user intelligence analytic corps must participate and influence the information design.
Given the exponential growth in U.S. technology capabilities, big data solutions have been seen as immediate and effective ways to deal with the overwhelming amount of raw intelligence collected. Big data is not just about data capture, storage and transfer—it includes the use of predictive analytics, behavioral analytics and other methods that extract value from large data sets of disparate structures and content. To augment big data solutions, artificial intelligence (AI) was developed. As Barry Zulauf, chief of the Solutions Division in the Office of the Director of National Intelligence (ODNI), noted, “. . . humans cannot possibly process all the information that is available. Artificial intelligence solutions help think through the data and display for analysts what they need to focus on, so humans can do what they do best—provide insight” (SIGNAL Media, October 2, 2020, “Technologies, Virus Wreak Changes…”). AI technologies help speed the processing of raw information and unburden the intelligence analyst. This is achieved through capabilities such as natural language processing, enhanced predictive analysis, indications and warning alerts, fusion of multiple intelligence (multi-INT) sensor data and fusing open-source information with classified information. Mark Lowenthal, former assistant director of Central Intelligence for Analysis and Production, cautions that AI is “very important in terms of pattern recognition and working through sets of data, but again we need a person to interpret it for us” (SIGNAL Media, July 27, 2020, “Intelligence Analysis Needs Course…”).
Information design, or how information is presented to the analyst, is critical for intelligence analysis. While casting a large collection net can increase the odds of obtaining critical pieces of information, the information collected is often incomplete and comes from sources of varying reliability. AI cannot replace or replicate an individual analyst’s years of experience and deep-layered knowledge needed to ascertain adversary intent. AI cannot replace the human cognitive power to detect patterns and anomalies—the proverbial “connecting the dots”—by filling in information gaps referencing that same human repository of experience. But innovative information design can support the human ability of sensemaking—understanding a situation in ways that can be conveyed to the intelligence consumer.
The concept of ergonomics offers a strategic and philosophical approach. Ergonomics is commonly associated with body comfort—office chairs, desk height, computer screen distance and angle. But fundamentally, ergonomics is about designing for people. Ergonomics consulting firm Humanscale says the term refers to the science of fitting a workplace to the user’s needs to increase efficiency and productivity and reduce discomfort.
And the International Ergonomics Association states that ergonomics helps harmonize things that interact with people in terms of their needs. This is not about human-machine interface, human-system interaction, human-computer interaction or user interface. In this case, the workplace or “thing” is not an office, furniture or a system; it is about the information itself. And the interface is not the eye or physical appendages but the brain. In that respect, this is cognitive ergonomics. The established discipline of cognitive ergonomics is concerned with mental processes, such as perception, memory, reasoning and motor response, as they affect interactions between humans and a system. In this context, the relevant components of cognitive ergonomics include mental workload, decision-making and skilled performance. This level of information design is more than conventional human-computer interaction and user interface like windows, mouse, cursor and pull-down menus but about the cognitive relationship with the information itself.
The Society for Experiential Graphic Design considers information design as the practice of presenting material in a way that makes it accessible and easily understood by users. In its most sophisticated forms, it helps users understand complex data by organizing and simplifying information in ways they can quickly grasp. Information design is closely associated with effective and functional display techniques versus pure aesthetics.
The most commonly known solutions include data visualization (visualization of raw data) and information visualization (visualization of processed or synthesized data) solutions. These have tended to be variations on the same themes: scatter plots, graphs, geospatial time-space maps, word maps, network association plots, information dashboards, etc. These are well-known, established, effective methods easily integrated into an information system. Fully understanding cognitive ergonomics can influence which visualization techniques can be used and how to arrange the information presentation as well as the development of new information presentation techniques.
Consideration should also be given to exploring new solutions and revisiting the fundamental human factors that influence information design. An example is text visualization. This is not just making specific words of varying sizes and colors, as in a word map, but words and phrases displayed in context. The presentation may include not only the source of the text but surrounding passages. Another emerging field is knowledge visualization—the presentation of knowledge created by human assessments and judgments. This solution aims to transfer knowledge to another person leveraging the same cognitive attributes that make data and information visualization efficiently palatable for humans.
Information design must also take into account the unique aspects of each intelligence discipline. For example, political-military analysis is different from measurement and signals intelligence analysis, and imagery analysis is different from cyber intelligence. The design of intelligence-related IT needs to be guided by an understanding of the end-user mission and business process. One methodology to achieve this is cognitive task analysis (CTA), which can help capture end-user information design requirements for not only the display of material but the structure of textual, document-based information. Per the User Experience Professionals Association, CTA aims to understand tasks that require a lot of cognitive activity from the user, such as decision-making, problem-solving, memory, attention and judgment. It is the study of what people know, how they think, how they organize and structure information, and how they learn when pursuing an outcome they are trying to achieve. Some of the steps of cognitive task analysis are mapping the task, identifying the critical decision points, clustering, linking and prioritizing them, and characterizing analytic and decision strategies. Processes such as CTA can help define how to structure information and achieve a balance of human usability and system functionality.
The information design stage should be an integral part of the initial design and development. It helps define the system information end state—what effort the system is supporting and to whom the information is presented. The ideal influencers in this effort are the actual end users, the intelligence analysts. In reality, their involvement with private industry system development is difficult at best and more often problematic. Even in ideal situations, it is hard to get the end user to participate. They have work to perform, and organizations do not want to lose their time on the job. Organizations may also feel this is what they are paying for—someone who knows their business and can design, develop, build and deliver an information system without such help.
Even when they participate, end users can have difficulty providing input. Many times they don’t know what they want because they don’t know what technology can provide. End users may find it difficult to actually put into words (sufficient to guide software coding) what they’re seeking because analysis includes components of intuition, conjecture and supposition. Even when provided, it can be difficult to translate user information requirements into design and software coding language or protocols. One option to improve user support is to provide nontechnical end users one-day training on techniques to help them define and express their information structure and presentation requirements.
Ultimately, human-centered design principles should guide the information presentation end state. As this is still an evolving industry and academic field, it lacks the definitiveness of conventional software design processes. Even still, there is a robust body of knowledge that is accessible and understandable to the lay person that can provide a philosophical information design strategy to ensure technology is designed for the human analyst.
Gary Gomez has over 20 years of government and industry experience in the intelligence community. He teaches intelligence studies at Old Dominion University in Norfolk, Virginia, and has also taught information visualization at George Mason University in Fairfax, Virginia. He is on the board of the Hampton Roads AFCEA Chapter and is chair of the chapter Intelligence Committee.