Search PNNL

• PNNL Home
• About
• Research
• Publications
• Jobs
• News
• Contacts

Research Applications: Human-System Integration/Human-Information Interaction

Pacific Northwest National Laboratory (PNNL) Cognitive Informatics research and development in human information interaction spans several areas of application aimed at improving human-system performance, enhancing human-computer interaction, and developing solutions to information analysis and visualization to effectively raise human-computer bandwidth.

Representative projects are described below with points-of-contact email addresses.

Motivation and Capability Modeling for Threat Anticipation

This project will identify motivation parameters that influence adversaries behavior and develop a model that reflects the relationships among these parameters. It provides a likelihood estimate of scenarios based on Socio-cultural parameters, which are defined according to specified life narrative indicators that act as proxies for underlying motivation constructs. It takes knowledge, skills and abilities (human capability information) into account to identify possible mappings between actors and their roles within threat scenarios. The innovative approach of integrating socio-cultural factors into the anticipatory analysis of threat scenarios enables analysts to gain a perspective on events and their implications that transcend situational awareness. This approach promotes what we have called cultural sensemaking allowing the analyst to see the world from the perspective of the adversary, with insight into motivational factors that address the question of who and why in addition to the what, where, and when questions that have historically guided intelligence analysis processes. Contact: Frank.Greitzer@pnl.gov

Distributed Grid Operator Training in the Electricity Infrastructure Operations Center (EIOC)

Investigations of large-scale outages in the North American interconnected electric system often attribute the causes to three T's: Trees, Training and Tools. To document and understand the mental processes used by expert operators when making critical decisions, a naturalistic decision making (NDM) model was developed. Transcripts of conversations were analyzed to reveal and assess NDM-based performance criteria. An item analysis indicated that the operators' Situation Awareness Levels, mental models, and mental simulations can be mapped at different points in the training scenario. This may identify improved training methods or analytical/ visualization tools. This study applies for the first time, the concepts of Recognition Primed Decision Making, Situation Awareness Levels and Cognitive Task Analysis to training of electric power system operators. The NDM approach provides a viable framework for systematic training management to accelerate learning in simulator-based training scenarios for power system operators and teams. Contact: Frank.Greitzer@pnl.gov

Normal and Emergency Power Grid Operations Visualizations

We are working closely with power grid operators to document current practice and displays/processes used in normal and emergency grid operations. Cognitive task analyses were used to document the process and applicable displays and visualizations. The analysis revealed not only display design deficiencies but also inefficiencies in the process that could be traced to the difficulty or number of steps needed to access relevant information. The results of the analysis informed the design of new displays and visualizations that are aimed to streamline and improve the decision process, where more rapid access to relevant information and improved analysis support saves time and allows the dispatcher/decision maker more time to focus more on problem solving. Contact: Frank.Greitzer@pnl.gov

Human Factors Analysis of Advanced Tools and Visualizations for Power Grid Operations

The Pacific Northwest National Laboratory (PNNL) has an ongoing research program aimed at developing advanced analyses and visualizations to enhance operator situation awareness and decision making in normal and emergency power grid operations. Acknowledging the need for validation studies, PNNL has initiated a human factors test and evaluation program within its Electricity Infrastructure Operations Center (EIOC) to conduct appropriate validation studies using internally-developed visualization tools. The expectation is that the framework and experimental methods specified and employed in this project will be useful for identifying performance impacts and utility of power grid analysis and visualization tools developed not only by PNNL but also by other R&D groups, with evaluations that may be conducted on behalf of industry and DOE stakeholders. Ultimately, it is hoped that the human factors and test/evaluation methodology, applied within the PNNL EIOC, will lead to the deployment of more effective tools and visualizations that will ultimately improve the performance of power grid operators, supervisory personnel, and regulatory policy makers through enhanced situation awareness and cognitive decision support, and ultimately create more reliable and secure electric power grid interconnection and operations.

2009 EMS Users Conference Presentation     Contact: Frank.Greitzer@pnl.gov

Predictive Modeling for Insider Threat Mitigation

This Laboratory-Directed Research and Development project is being conducted under the Predictive Defense focus area of PNNL's Information & Infrastructure Integrity Initiative. Current practice in addressing the insider cyber threat is to monitor the network and individual systems in order to identify when someone is not following established policy or is abusing their authorized level of access in a way that is harmful to the organization. This involves the use of tools such as Firewall logs or IDS systems on networks or host systems that produce records of activity are later reviewed. Thus, current practice is reactive (post-hoc). A major research challenge is to effectively reduce the time between defection and detection, even to the point where detection of "threat indicators" can help to predict such exploits before they are completed. Major accomplishments include a review of prior research and practice in insider threat detection tool development (establishing the technical/scientific basis for the work), development of conceptual design for the envisioned predictive/adaptive functions of the model—illustrated in the diagram below showing the processing of incoming sensor data to infer observations, processing of observations to infer indicators, and analysis of indicators to gauge extent of threat (malicious behaviors). The research also described threat indicators that address cyber and social/organizational factors as precursors to malicious exploits. Current focus is on implementing selected classification algorithms and reasoning components of the predictive model. Contact: Frank.Greitzer@pnl.gov

Scalable Reasoning Systems

Technology to support knowledge transfer and cooperative inquiry must offer its users the ability to effectively interpret knowledge structures produced by collaborators. Communicating the reasoning processes that underlie a finding is a method for enhancing interpretation that can result in more effective evaluation and application of shared knowledge. In knowledge management tools, interpretation is aided by creating knowledge artifacts that can expose their provenance to scrutiny and that can be transformed into diverse representations that suit their consumers? perspectives and preferences. This National Visualization and Analytics Center (NVAC) research on Scalable Reasoning is advancing tools and methods for teams of collaborating analysts to capture, share, and reuse analysis processes and reasoning strategies through a combination of desktop and mobile environments for information capture and synthesis. Analytic products can be visually composed and disseminated to consumers who in turn can "unwrap" these products to evaluate their associated evidence, assumptions, and reasoning. This research is being carried out by PNNL for the U.S. Department of Homeland Security, which established NVAC as a national and international resource for visual analytics. Contact: Bill.Pike@pnl.gov

Intelligent Multi-Agent System for Knowledge Discovery

Researchers at PNNL are working on the design and development of systems that enhance human-information interaction in information analysis and discovery for diverse applications, such as intelligence analysis and bio-informatics. The goal is to enable a decision maker to ask a complicated question and receive fused information to support knowledge discovery. Our approach integrates a cognitive model and a cooperative community of intelligent software agents into a computer-based data analysis architecture that supports information analysis, synthesis, and discovery from massive, complex, and heterogeneous data sets to aid in research. The cognitive modeling effort supports the system's capability to adapt to changes in the analyst's strategies for data gathering or analysis. The research on intelligent systems is focused on developing innovative algorithms for agent management, techniques for managing explicit and implicit knowledge, and novel methods for transforming retrieved information into a common ontological form to enable knowledge insertion. Finally, we are developing methods for presenting this fused information in an appropriate manner to the domain expert. Contact: Olga.Kuchar@pnl.gov

Testbed for Intelligence Analysis Tools

The Advanced Research and Development Activity, through its Novel Intelligence from Massive Data (NIMD) program, has undertaken a research program to assist intelligence analysis. As part of this program, PNNL developed a Glass Box Analysis Environment (GBAE) to provide data to increase researchers' understanding of cognitive foundations and to provide an integration/test environment for NIMD-developed tools. A continuing challenge was to define requirements for automated data collection functions that are unobtrusive, yet robust and complete enough not only to capture lower-level data on human-computer transactions but also to shed light on the analyst's higher-level cognitive processes. The GBAE instrumentation software has produced an extensive repository of automated human-computer interaction data augmented by user- initiated annotations.

Glass Box Flier     Contact: Laura.Curtis@pnl.gov

Methodology and Metrics for Intelligence Analysis Tool Evaluation

Intelligence analysis (IA) professionals are confronted each day with high demands for rapid, yet accurate assessments that require discovery and marshalling of evidence, integration and synthesis of data from disparate sources, interpreting and evaluating data and information that are constantly changing, and making recommendations or predictions in the face of inconsistent and incomplete data. Recognizing the difficulty of the IA task, stakeholders and the research community have been seeking technology-based solutions to reduce the analyst's workload and improve the throughput and quality of IA products. A challenge for the research community is to develop useful and valid metrics and measures that may be used to assess the impact of such products. Some fundamental issues are: (a) how to employ rigorous methodologies in evaluating tools, small-sample size limitations and the need to control for task difficulty and possible effects of time or learning; (b) how to measure the difficulty/complexity of IA tasks; and (c) development of more rigorous (summative), performance-based measures of human performance beyond the more traditional reliance on formative assessments (e.g., subjective ratings). This critical challenge must be addressed to ensure that tools and techniques introduced into the IA process are effective. Research on these challenges is currently being conducted by PNNL in support of intelligence community stakeholders. Contact: Frank.Greitzer@pnl.gov

Enabled by the Office of Science

Security & Privacy | Site Index | Contact PNNL