U.S. Air Force Summer Faculty Fellowship Program

U.S. Air Force Summer Faculty Fellowship Program

U.S. Air Force Summer Faculty Fellowship Program

U.S. Air Force Summer Faculty Fellowship Program

AFRL/RH 711 HPW Wright-Patterson Air Force Base, Ohio

SF.15.07.B0078: Auditory Perception and Speech Communication

Iyer, N.

(937) 255-4381

The goal of the research project is to understand the mechanisms and principles by which effective communication and speech perception occur in acoustically-challenging environments (“cocktail-party” scenarios) and explicitly apply the knowledge gained to develop effective, robust and intuitive interfaces for not just human-human communication, but also human-machine communication. Areas of research includes: 1) characterizing and modeling the sensory and cognitive constraints in complex acoustic environments with multi-sensory inputs, 2) Bi-directional listener-talker interactions and adaptations, 3) Characterizing the impacts on communication efficiency and effectiveness with decreasing speech intelligibility, 4) Developing intuitive, next-generation, and robust speech-based displays/speech output systems that will enhance speech and improve communication in operation environments not only for humans but also human-machine communication.

SF.15.07.B5700: Increasing Information Transfer in Audio Display Systems

Simpson, B.

(937) 255-4463

Human audition is an amazingly complex modality capable of extracting spatial, spectral, and temporal information from multiple simultaneous sound sources even in adverse listening environments. However, most real-world audio display systems rely on relatively simple stimuli that fail to take full advantage of the inherent capabilities of human listeners. The goal of this research is to find ways to increase the amount of information transferred to listeners through audio display systems. The effort involves two broad areas of research. The first area focuses on the generation of robust and intuitive azimuth, elevation, and distance cues that maximize the transfer of spatial information in audio displays, especially in noisy environments that involve more than one virtual sound source. The second area focuses on improving the segregation of competing sound sources in complex listening environments, especially those that involve more than one simultaneous speech signal. A major component of this research is a study of the role that non-energetic "informational" masking plays in the perception of multiple speech signals.

SF.15.08.B7532: Mathematical Modeling for Performance Prediction: Mechanism Development to Account for Effects of Cognitive Moderation

Jastrzembski, T.

(937) 938-4046

Researchers at the Air Force Research Laboratory’s Airman Systems Directorate have developed, matured, and made more robust a mathematical model for performance prediction, known as the Predictive Performance Equation (PPE) (see Jastrzembski, Gluck, & Gunzelmann, 2006; Jastrzembski, Gluck, & Rodgers, 2009). This model has been carefully validated across a variety of domains and contexts – scaling from laboratory experimental data available in the psychological literature to increasingly complex and militarily relevant team and pilot data measured in the Air Force Research Laboratory‘s Distributed Missions Operations testbed (Schreiber, Stock, & Bennett, 2006). The predictive model functions by capturing learning signatures and mathematical regularities from the human memory system through calibration of learning and decay parameters using historical performance data, and extrapolates those unique learning signatures to make predictions of performance at specific later dates in time. The model critically extends that previous research by additionally accounting for the effects of temporal distribution of training on learning – a well-documented phenomenon known as the spacing effect – which reveals that given two training regimens of equal length and equal amounts of training opportunities, learning is more stable when practice events are spaced further apart in time. This research seeks to extend the model even further, by incorporating mechanisms that explicitly attenuate performance through effects of cognitive moderation (i.e., enhancement of performance from brain stimulation or caffeine; decrement of performance from fatigue or excess workload), so that a more complete picture may be gleaned regarding the range of likely performance effectiveness under known conditions.

Anderson, J. R., & Schunn, C. D. (2000). Implications of the ACT-R learning theory: No magic bullets. In R. Glaser (Ed.), Advances in instructional psychology: Educational design and cognitive science, Vol. 5. Mahwah, NJ: Erlbaum.

Jastrzembski, T. S., Gluck, K. A., & Gunzelmann, G. (2006). Knowledge tracing and prediction of future trainee performance. I/ITSEC annual meetings, Orlando.

Jastrzembski, T. S., Gluck, K. A., & Rodgers, S. (2009). Improving military readiness: A state-of-the-art cognitive tool to predict performance optimize training effectiveness. I/ITSEC annual meetings, Orlando.


Schreiber, B. T., Stock, W. A., & Bennett, W. (2006). Distributed mission operations within-simulator training effectiveness baseline study: Metric development and objectively quantifying the degree of learning. AFRL-HE-AZ-TR-2006-0015-Vol II. Available online at: www.dtic.mil.

SF.15.08.B7534: Visual Analytics

Havig, P.

(937) 255-3951

The field of information visualization is wide ranging and runs the gamut from aesthetically pleasing visualizations to those that give much needed information to a user on demand. Air Force applications for cyberspace need to rely heavily on information visualization techniques to provide this “on demand” capability in such a way that users do not need to be experts in the field to understand a visual display. Further, visual analytics looks at how to optimize the interaction with the visualization so the user spends more time exploring and understanding the visualization and less time trying to figure out how to navigate the environment. We are interested in this cross road between optimal user interface and visualization of large, complex data sets. 

SF.15.13.B0918: Human Morphology, Modeling, and Discrimination

Camp, J.


Modern defense applications have become increasingly human-centric, focusing, for example, on identification of individual or group characteristics and behavior. Our current human-centric applications include: simulating realistic human size, shape, and motion for biofidelic computer animations; discriminating among individuals or groups of individuals from a distance (soft biometrics), and expanding understanding of the relationship between human structure and movement. The purpose of this research project is to merge anthropometric and morphological measurement with human modeling and movement analysis. Research goals include statistical analysis of 3-D human scans, variable reduction and identification of key anthropometric variables and shape descriptors, prediction and simulation of human size and shape, and correlation of structural measures with movement parameters. The ultimate goal of the research is to create a 3-D human model that adapts size and shape according to parameters such as weight, gender, age, etc.

Our Human Signatures Laboratory is equipped with whole body scanners, motion capture cameras, video cameras, and other advanced sensor technologies. We have various software tools for human shape and motion analysis. We have built strong capabilities in human modeling, and have large existing databases of human size, shape, and movement patterns. Candidates with demonstrated knowledge and experience in biology or anthropology, computer science, and statistical techniques are desired. Selected applicants will be expected to work with USAF staff, collaborating university faculty, and contract support staff to develop the methodologies and conduct experiments to validate them if necessary.

SF.15.09.B0919: Human Characterization and Activity Recognition in Full Motion Video (FMV) from Fielded Systems

Camp, J


Modern defense applications have become increasingly human-centric, focusing, for example, on identification of individual or group characteristics and behavior. The purpose of this research project is to explore human characterization and activity recognition from full motion video using realistic resolutions and observation geometry from fielded systems. Research goals include identification and development of techniques to characterize humans observed in full motion video (FMV) using qualitative and quantitative features in order to determine the likelihood that that particular person was observed in another, seemingly unrelated, video stream. Extraction of anthropometric, biomechanics, and soft biometric methods are desired as well as pattern recognition and machine intelligence techniques to determine match confidences.


Candidates with demonstrated knowledge and experience in machine intelligence, pattern recognition, and computer vision are desired. Selected applicants will be expected to work with USAF staff, collaborating university faculty, and contract support staff to develop the methodologies and conduct experiments to validate them if necessary.

SF.15.09.B1133: Trust Research in the Air Force: Implications for Human-Machine Trust

Lyons, J.

(937) 255-8734

This research will investigate the psychological (e.g., heuristics/biases), technological (e.g., transparency design features), or contextual predictors of trust calibration between humans and machines. Special emphasis is placed on empirical evaluation of how one or more of the facets described above shapes the trust and subsequent performance of human-machine systems. We would like to understand how to predict when an individual will engage in misuse or disuse when aided by intelligent automation and how to augment the human-machine system to reconcile the misalignment to promote better human machine system performance. 

SF.15.11.B0086: Predictive Toxicology

Mattie, D.

(937) 904-9569

Mathematical modeling provides quantitative prediction of the dose- and time course- responses of mammalian systems to toxic chemicals. Simple and complex in vitro toxicological models can provide novel findings uncovering chemical-response mechanisms. This research program focuses on developing methods and computational models in application across in vitro/single cell to complex whole animal/human systems for more rapid assessment and analysis of chemical and material toxicity. Research areas support the overall advancement of human toxicokinetic and toxicodynamic modeling of exposure to single and mixtures of operational chemicals and materials. Approach sought will aid in development of biologically-based kinetic (BBK) models that incorporate the limiting rate factors for the absorption, distribution, and elimination of chemicals and nanomaterials in human. We seek to generate, interpret, code and integrate data for primary toxicity mechanisms starting from membrane effects and transport after exposure to those differentially modulating critical cell pathway functions prior to final elimination of the insult. Sub-objectives include: 1) developing, building and assessing in silico and in vitro models (cell lines, co- and multi-cell systems) for quantifying toxicity effects. These studies include novel approaches to computational and cell-based validation of key toxicity control mechanisms such as induction or loss of specific pathways or proteins and those incorporating broader rate limiting processes for providing quantitative evaluation of chemical safety. 2) Coding empirical toxicology data from rapidly-acquired, high-throughput and high-content in vitro toxicity studies to aid in developing rate-limiting and mechanistically-based BBK toxicity models, which aid in in vivo toxicity kinetics and response prediction for mammalian systems (an in vitro-to-in vivo extrapolation). 3) Integrating current toxicology modeling with ~omic data sets obtained from current or emerging technologies involving genomic expression through induction, activation and function of cellular proteins. 4) Developing computational approaches to couple in vitro gene expression patterns to protein induction and activation with incorporation of chemical structural activity measures and protein/pathway activity. 

SF.15.12.B0911: Efficient Constraint-Based Search Mechanisms in a Cognitive Domain Ontology

Douglass, S.

(937) 938-4057

Air Force Research Laboratory’s Airman Systems Directorate cognitive scientists are researching ways to increase the autonomy of cognitive models and agents. One approach to increasing autonomy involves specifying agents with formal representations of themselves, their knowledge, and the affordances of the situations in which they are acting. Researchers leading this approach are developing these representations or Cognitive Domain Ontologies (CDO) using System Entity Structure (SES) theory. SESs are founded on set theory. CDOs are used by autonomous agents to generate effective actions according to the contingencies and affordances presented by the environments they are situated in. These contingencies and affordances are made available as 'constraints' in a CDO. CDO also contains an agent's behavior repertoirre that gets soft assembled per these dynamic constraints. A CDO is the knowledge-base of the agent and contains representations of elements such as the environment, resources, goals, behaviors, etc. The elements of a CDO are linked together by these constaints. Airman Systems Directorate researchers are looking for efficient constraint-based search mechanisms to limit the combinatorics of CDO search. The search algorithms will be grounded in AI-based methodologies & set theory and must be executable on parallel/distributed high performance systems. A key feature of the proposed algorithms must be scalability. The algorithms must be able to complete the search process within 0.3-1.0 sec wall-clock time. Performance analysis of algorithms will therefore be a critical aspect of the research. The successful execution of the algorithm will result in a set of cognitive behaviors within the CDO which will prescribe effective action in the situated environment.

Airman Systems Directorate researchers are interested in collaborating with academic partners that can contribute to the research and development of contraint-based search algorithms used to process a cognitive domain ontology. Collaborators would design, develop, and analyze knowledge and constraint representation schemes. Collborators would also develop algorithms, implement them in high level programming languages, and execute/evaluate them in high performance parallel/distibuted architectures.



Zeigler, B., & Hammonds, P. (2007). Modeling & Simulation-Based Data Engineering: Introducing Pragmatics into Ontologies for Net-centric Information Exchange. Academic Press.

SF.15.12.B0912: Developing and Validating Quantitative Theories of Human Cognitive Processing

Gunzelmann, G.

(937) 938-3554

General theories of cognition have been successful in accounting for many important aspects of cognitive processing. At the same time, there are many components of cognition where well-validated theories are lacking. A critical area that has received relatively little attention in the cognitive science community is fatigue. Within the Air Force Research Laboratory’s Cognitive Models and Agents Branch scientists are conducting research to develop quantitative theories of how factors like sleep loss and time on task impact cognitive processing. The research focuses on detailed laboratory studies to expose important phenomena combined with the development of computational models to account for the empirical results. We are interested in collaborations with university faculty with expertise in sleep, vigilance, and related areas to develop empirical studies and computational models to expand our understanding of these areas of cognitive functioning.


Keywords: Vigilance; Sleep deprivation; Fatigue; Computational modeling; Cognitive architecture 

SF.15.12.B0913: Competency-based Needs Assessment, Training Design and Delivery, and Performance Assessment Research in Adaptive Environments for Continuous Learning

Bennett, W.

(937) 938-2550

The US Air Force has invested heavily in a training and rehearsal concept called Distributed Mission Operations (DMO) and is working on augmenting and integrating DMO in a new construct called Live, Virtual, and Constructive (LVC) operations. DMO provides the virtual and constructive elements of the new construct while the Air Force is interested in integrating live, operational systems into a managed realistic, adaptive, and affordable environment enterprise for training, rehearsal, test and evaluation. The environment allows local and wide area connection of virtual simulators, constructive models, gaming environments, and relevant live operational systems, such as actual aircraft.

The success of DMO and LVC hinges on several critical research needs that must be addressed. (1) Mission needs and critical knowledge, skills, and experiences must be specified and represented at appropriate levels of analysis. (2) Training objectives and scenarios must be designed to meet these specifications using a principled instructional approach. (3) Construct-oriented, systematic methods to predict, diagnose, monitor, and assess the performance of trainees within environments must be developed and validated. These methods will assist in the prescription of content and remediation within DMO and LVC ops to address knowledge and skill deficits, and to help develop a new class of human performance and machine learning based models. (4) Methods, models, and metrics that are linked to training objectives and that permit routine and longitudinal assessments of individual and team performance and proficiency in synthetic environments and operational settings must be developed and validated. (5) The appropriate identification, integration, and validation of a "mix" or "family" of complimentary trainers and environments that promote, maintain, and/or accelerate learning and performance must be developed. There are also significant research opportunities to develop, implement, and evaluate innovative methods to link and represent core knowledge, skills, and experiences in a way that helps define the environments for learning, and which facilitates training development, delivery, evaluation, and transfer.

We are also interested in developing and validating criterion measures related to the impact of DMO and LVC on learning, proficiency, and readiness and that help to quantify intervals necessary for refresher training. There is considerable latitude for research that explores how best to manage some, most, or all of the learning enterprise. Research can include (1) improving the quality and precision of needs assessment, gap, and trade space analyses (2) training/scenario design, delivery, and management tools (3) integrating diverse approaches to training such as game-based systems and environments, intelligent and adaptive training environments, and part task trainers (4) developing methods to improve the credibility and security of learning, data exchanges, and interoperability among the systems (5) rapid prototyping of novel approaches to human performance monitoring, modeling, assessment, and feedback (6) developing more precise and generalizable performance measurement and proficiency tracking data  (7) improving ways to visualize and package feedback data for after action reviews. It may also include application of different approaches/strategies to learning and assessment in a variety of integrated and adaptive environments and contexts.


Arthur, W., Jr., Day, E.A., Bennett, W., Jr., & Portrey, A. (Eds.). (2013). Individual and team skill decay: The science and implications for practice.  Mahwah, NJ: Taylor Francis

Bennett, W. Jr., Schreiber, B, Portrey, A.M., & Bell, H.H. (Eds.) (2013).  Challenges in transforming military training: Research and application of advanced simulation and training technologies and methods.  Military Psychology (Special Issue).

Pavlova, E., Coovert, M. D., & Bennett, W., Jr. (2012, April). Trust development in computer-mediated teams. Society for Industrial & Organizational Psychology. San Diego, CA.

Alliger, G.M., Beard, R., Bennett, W., Jr., & Colegrove, C.M. (2012). Mission essential competencies: An integrative approach to job and work analysis. In M.J. Wilson, W. Bennett, Jr., S.G Gibson, & Alliger, G.M. Alliger (Eds.). The handbook of work analysis in organizations: The methods, systems, applications, & science of work measurement in organizations. Mahwah, NJ: Taylor Francis.

Arthur, W.E., & Bennett, W., Jr. (2012). Innovations in team task analysis: Identifying team–based task elements, tasks, and jobs. In M.J. Wilson, W. Bennett, Jr., S.G Gibson, & Alliger, G.M. Alliger (Eds.). The handbook of work analysis in organizations: The methods, systems, applications, & science of work measurement in organizations. Mahwah, NJ: Taylor & Francis.

Schreiber, B.T., Bennett, W., Jr., Colegrove, C.M., Portrey, A.M., Greschke, D.A., & Bell, H.H. (2009). Evaluating pilot performance. In Ericsson, K. A. (Ed.), The development of professional expertise: Approaches to objective measurement and designed learning environments. New York: Cambridge University Press.


Day, E. A., Arthur, W. Jr., Bell, S. T., Edwards, B. D., Bennett, W. Jr., Mendoza, J. L., & Tubre, T. C. (2005). Ability-based pairing strategies in the team-based training of a complex skill: Does the intelligence of your training partner matter? Intelligence, 33, 39-65.

SF.15.12.B0916: Non-Invasive Brain Stimulation to Enhance Cognitive Performance in Air Force Operators

McKinley, A.

(937) 938-3598

The purpose of this project is to evaluate non-invasive brain stimulation techniques and technologies to enhance and optimize human performance. Specifically, the aim is (1) perform basic research into the neurobiological mechanisms of non-invasive brain stimulation responsible for changes in behavioral performance and (2) to conduct applied research in the efficacy of non-invasive brain stimulation techniques, such as transcranial direct current stimulation, as a means to facilitate cognitive skills such as visual search, learning/memory, and attention. The goal is to improve performance through direct augmentation of cortical excitability or activation. All research will be conducted within the cognitive performance laboratory suite, located at Wright-Patterson AFB, OH.

SF.15.13.B0915: Sensor Platform Development for Rapid to Real-Time Detection in Biofluids

Hagen, J.

(937) 938-2576

The ultimate goal of performance monitoring is to build sensors capable of continuous, real-time analysis of biomarkers for targets indicating stress, fatigue, vigilance, and overall other physiological conditions. Biomarkers found in biofluids can be extremely indicative of physiological state. Traditionally, these biomarkers are assessed with labor intensive biofluid (blood, saliva, urine) sampling and analysis with complex equipment and assays (HPLC, ELISA etc.). To make biomarker tracking a feasible monitoring system, sensor platforms must be developed for rapid to real time analysis. These can be in handheld form factors such as a lateral flow assays or in a wearable form factor such as a transdermal patch.

The objectives of this research are to develop sensor platforms that are amenable to either rapid or real-time analysis of biofluids. Of particular interest are blood and sweat. Sensor platforms should have a small/portable form factor for handheld assays or flexible/wireless capability for wearable form factors. Platforms should be capable of detecting a wide range of molecule types from small <300 Dalton to proteins >3000 Dalton. Additional interest lies in pre-processing of biofluids to increase sensitivity/selectivity of the sensor platform.

SF.15.14.B0837: Strategic Adaptation in Dynamic Non-stationary Environments

Myers, C.

(937) 938-4044

Dr. Chris Myers is conducting research that uses machine learning and control theory techniques to model human performance and decision making within dynamic, non-stationary environments. The work involves both human experiments and computational modeling, with the aim of understanding how humans adapt to changes in complex, dynamic, and collaborative task environments. We are looking for university faculty interested in developing optimal models of visual search, multitasking, decision-making, and dyadic collaboration. The ideal candidate will have a background in machine learning, cognitive science/cognitive psychology/mathematical psychology, and/or control theory, and will have programming experience with R and/or Python.


Chen, X., Howes, A., Lewis, R. L., Myers, C. W., & Houpt, J. W. (2013). Discovering computationally rational eye movements in the distractor ratio task. First Annual Multidisciplinary Conference on Reinforcement Learning and Decision Making, Princeton University, Princeton, NJ, USA.

Myers, C. W., Lewis, R. L., & Howes, A. (2013). Boundedly Optimal Adaptation During Visual Search. 35th Annual Meeting of the Cognitive Science Society. Berlin, Germany.


Veksler, V. D., Myers, C. W., & Gluck, K. A. (2012). An Integrated Model of Associative and Reinforcement Learning. 34th Annual Meeting of the Cognitive Science Society. 

SF.15.14.B0840: Neurobiology of Cognitive Performance

Jankord, R.

(937) 938-3144

Performance under stress determines mission effectiveness. The goal of the Neurobiology of Cognitive Performance Team is to understand the biological mechanisms that affect performance. Our work involves physiological and behavioral (attention, anxiety, spatial memory and emotional memory) testing in rodents and examination of the neurobiological changes that occur following treatment. Current projects in our laboratory include a study on neural modulation via transcranial direct current stimulation (tDCS) and a behavioral genetics study on stress resiliency. Our tDCS study seeks to understand the biological mechanisms (gene expression and cell signaling pathways) by which transcranial direct current affects neuronal activity, providing insight into how this methodology affects cognitive function. The behavioral genetics study involves the use of BXD mice, an established genetics reference population, where behavioral outcomes of stress exposure are mapped (QTL mapping) onto defined chromosomal sequences. The goal of the behavioral genetics study is to identify novel genetic factors that modulate performance in a stressful environment.

SF.15.14.B2414: Adaptive Training Methodologies for Analyst Teams

Tripp, L.


This research topic focuses on developing adaptive training methodologies for individuals and teams performing intelligence analyst’s tasks for Air Force mission areas. Providing an optimal level of difficulty of training content has been shown not only to increase the efficiency of training (i.e., not wasting time with content that is significantly below or above an individual’s current competency level) but also increases retention and transfer (e.g., McDaniel & Einstien, 2005; Yue, Bjork, & Bjork, 2013). The first step in building adaptive training system is understanding trainees’ current level of competency for a set of training objectives (i.e., identifying current user state in terms of competencies). The next step is to provide recommendations which optimize training efficiency and accelerate the development of expertise. The objective of this research is to optimize real-time micro and macro level scenario adaption to individuals and teams in realistic training environments. Micro, in this case, referring to adding/deleting events within a scenario to appropriately modulate the level of difficulty. Macro, in this case, referring to identifying the most effective training event for the next training opportunity. The ideal candidate will have a background in cognitive science/cognitive psychology/mathematical psychology, operations research, computer science, and/or artificial intelligence and will have programming experience.


Yue, C. L., Bjork, E. L., & Bjork, R. A. (2013). Reducing verbal redundancy in multimedia learning: An undesired desirable difficulty?. Journal of Educational Psychology, 105(2), 266.

McDaniel, M. A., & Einstein, G. O. (2005). Material Appropriate Difficulty: A Framework for Determining When Difficulty Is Desirable for Improving Learning.


SF.15.14.B0843: Engineering In Vitro Multicellular Organ System For High Throughput Screening of Aerospace Toxicology Targets

Hussain, S.

(937) 904-9517

In order to protect the warfighter in operational environments, novel technologies are being developed to enhance warfighter capabilities. However, prior to implementation of these advancements, there is a need to rapidly and systematically evaluate the potential risk that these advanced nanoscale materials and aerosol chemicals might pose to the warfighter. At this time, due to the large number of chemical substances and the emergence of new materials, in vivo testing would be too lengthy and costly to pursue, yet traditional in vitro methods are lacking. For example, the use of unicellular in vitro models to represent complex tissues, such as the lungs, has received sharp criticism as these models cannot accurately depict how a multi-cellular tissue will respond. Therefore, one of the critical challenges faced in predictive toxicity is the lack of flexible in vitro model systems, which – i) allow for rapid, quantitative, and systematic testing; and ii) mimic the in vivo tissue microenvironment to accurately predict in vivo transport behaviors of chemicals and nanomaterials. However, in vitro cultures are useful for providing a preliminary foundation for studies to assess dosing ranges, probable mechanisms of toxicity, and allow for the refinement of techniques before progressing with costly in vivo studies. In addition, high throughput screening (HTS) using in vitro models can be performed to down select chemical targets for future in vivo studies. The development of microfluidics chambers in conjunction with co-cultures can be used to mimic blood flow and chemical transport. The goal is to develop complex in vitro cell model systems (3D models, organs on chips, airways on chips, etc) that better represent the organ systems found in vivo to further extrapolate in order to adequately assess human health effects of these nanoscale materials and aerosol chemicals.

SF.15.14.B0844: Biolmolecular Interactions of Nanomaterials: From Safety to Applications

Hussain, S.

(937) 904-9517

Our current research focus is to perform basic studies in order to establish a foundation for understanding the biomolecular interactions of nanomaterials (NMs). The unique quantum properties of NMs strongly influence their physico-chemical properties, conferring electrical, optical and magnetic properties not present in the corresponding bulk materials at a larger scale, making them ideal candidates for novel technologies to aid the AF mission. However, before the full potential of such a technology can be reached, a basic foundation evaluating the biomolecular interactions on NMs in a controlled system must be established. We have developed a characterization paradigm to identify the physical and chemical behavior of engineered NMs in order to link cell responses to specific NM parameters. In addition, multi-cell model approaches are being applied for low-level acute and chronic NM studies to elucidate the subsequent impact of NMs on cell signaling and gene expression. As NMs possess unique properties, it is highly probable that when they encounter an external field, such as radio-frequency, electromagnetic, or laser, that their enhanced surface energy and reactivity will produce a distinct effect. One major research thrust is to evaluate if simultaneous cellular exposure to NMs and an external field would produce synergistic cellular outcomes, assessing endpoints on an entire cellular population, protein, and genetic level. In order to evaluate this effect, we have developed a new prototype of non-traditional assays to evaluate the effects of engineered NMs on cellular systems under the influence of non-invasive incidental electromagnetic fields found in every day environments. Since not all NM display toxicological concerns, our focus has also been on utilizing biocompatible NMs for the development of nano-devices. Using gold nanomaterials, graphene, graphene oxide, and novel combinations of these materials, we will establish the groundwork for the biological interactions of NMs which can in turn be utilized to predict the implication of NM exposure, and exploit the benefits of NMs to the AF’s full advantage.

SF.15.16.B0001: Recognizing Malicious Intent in Firewall Traffic

Thomas, G


Typically, cyber attackers make multiple attempts to intrude upon a network prior to success. While it might be possible to interrupt their efforts, network defenders generally do not concern themselves with failed attempts for multiple reasons. First, the traffic that gets rejected is not commonly captured. Second, if it is captured, the data set is extremely large. Finally, there is an enormous amount of data that needs analysis from traffic that cleared the firewall, which monopolizes the resources of network defense and leaves little time for concern for traffic that did not enter the network. In order to determine if traffic rejected by network firewalls could be used to predict and prevent attacks, research is needed to 1) develop data sets that replicate attempted intrusions, 2) determine if there are patterns of activity that can be detected through automation, and 3) conduct research to determine the best methods for presenting that activity to human cyber defenders.

SF.15.16.B0002: Advanced Human Language Technologies for Multilingual Multimedia Information Extraction and Retrieval

Slyh, R.

(937) 255-9248

The objective of this research is to develop advanced human language technologies (HLTs) such as automatic speech recognition, machine translation, named entity tagging, part-of-speech tagging, and morphological analysis for use in multilingual multimedia information extraction and retrieval applications. Of particular interest are (1) algorithms and techniques to incorporate broader context (i.e., across sentences and utterances) as opposed to (or in addition to) current techniques that generally process inputs one sentence or utterance at a time without regard to prior sentences or utterances; (2) deep neural networks, long short-term memory networks, and other advanced algorithms for HLTs; (3) active on-line learning of possible translations and/or transliterations of out-of-vocabulary words encountered in machine translation; (4) methods for segmenting multi-story videos (e.g., news broadcasts) by jointly using speech and video frame information; and (5) improved methods for processing languages with little labeled training data.

SF.15.16.B0003: Social Dynamics in Human-Machine Teaming

Stokes, C.

(937) 919-3778

The (Human-Machine Social Systems Lab) HMSS Lab investigates core social cognitive factors that underlie human interactions and impact decision making and performance. These factors are examined in various human-machine teaming (HMT) contexts in order to generate empirical evidence and derive a novel theory of human-machine social systems. The explosion in advanced technology areas such as artificial intelligence and biometrics, and the potential for gestalt gains from human-machine collaboration necessitates a greater emphasis on social cognitive approaches in order to achieve the full capability of HMT.

The HMSS Lab is defined by a multidisciplinary, collaborative approach and leverages multimodal assessment capabilities. Core disciplines include social psychology, psychophysiology, neuroscience, human-robotic interaction, computer science, and organizational and human factors psychology with an emphasis on team effectiveness. Multimodal capabilities include eye-tracking, EEG, EMG, GSR, and facial recognition to correlate with standard self-report and behavioral measures.

The lab operates as an RHX resource housed at Yale University and directed by an AFRL researcher (Dr. Stokes) under an AFRL-Yale agreement. As such, unique collaborative opportunities abound, bringing together leading academic minds fused with a DoD perspective. Candidates with demonstrated knowledge and experience in any of the aforementioned disciplines are desired, however, we currently have particular interest in candidates with expertise in machine learning techniques applied to human-robotic/agent interaction. Selected applicants will be expected to work with AFRL and Yale University researchers to develop methodologies and contribute to human-robot/agent experiments.

SF.15.17.B0001: Perceptual & Cognitive Factors in Real-Life Information Seeking: Theories, Models, & Methods

Warren, R.

(937) 255-9943

People actively seek information. They may want the information just out of curiosity and for entertainment (e.g., watching TV); or to immediately act on it (e.g., checking traffic to change lanes); or for long term planning and decision making (e.g. gathering data on used cars). The search may be basically sensorial (sniffing an aroma, visually scanning a crime scene, listening for a sound in the night); or social (asking for directions; or by reading books and on-line internet pages; or by using sophisticated technology. Searches may be efficient or inefficient, successful or unsuccessful, or truly informative or riddled with errors and wrong conclusions. Search errors can be due to misperception or misinterpretation (false alarms) or misses (failure to find what is there). Many factors can influence the search itself and its success or failure such as attention, prior knowledge, training, biases, cultural factors, social factors (individual versus team search), time, resources, and technology. Indeed, two people may view the same event but attend to different information, or may react differently to the same information. Since curiosity and search behavior is so central to humans, we need to better understand basic perceptual, cognitive, and affective factors in information seeking. By understanding, we do not mean a collection of anecdotes and rules of thumb. Rather we seek an ecologically-relevant general theory based on real-world empirical facts and expressed in mathematical and computational models. We need metrics for quantifying available information and for assessing search performance. Ultimately, we seek methods to augment humans searching for information and to increase performance in real-life ecologically-valid situations.

SF.15.17.B0002: Molecular Tools for Biosignature Tracking

Chavez Benavides, J.

(937) 938-2575

Working in the Molecular Signatures Branch, our group focuses on the design of sensing platforms that allow selective and fast biomarker detection. Importantly, in order to track biosignatures relevant to different scenarios, including stress, fatigue or cognitive function, multiple biomarkers need to be monitored simultaneously. Therefore, different technical approaches for multiplex assays are currently being developed in our group including the use of single analyte assays in parallel and the use of cross reactive arrays. A key component of these systems is the biorecognition element (BRE) that provides the selectivity to the sensing scheme proposed. Our group is interest in developing novel selection techniques for DNA aptamers (and other nucleic acid-based BREs). In this AFRL Summer Faculty Fellowship Program, we seek for novel ideas for fast and robust aptamer selection techniques for small molecule analytes. Typical analytes of interest include neurotransmitters, steroids and others. The selected BREs will be integrated in nanoparticle-based plasmonic sensors, and complex fluorescence probes being developed to monitor neurochemicals in the brain and peripheral tissue to aid understanding of brain function.

SF.15.17.B0003: Advancing Biosensor Development using Synthetic Biology Approaches

Kelley-Loughnane, N.

(937) 255-3784

Our AFRL team is investigating biological engineering as a tools, specifically we are developing biologically-based sensors for stand-off detection and human performance modulators. Nature offers extraordinary examples of specificity, controlled response, and exceptional signal amplification that enable nature to detect and track items of interest. Molecular switches are exquisite examples of how nature can bind molecules with specificity and then produce a signal output. Engineering natural systems as novel sensors is an important technical capability for the United States Air Force. The goal of this research effort is to exploit the cell’s ability to sense small molecules in complex backgrounds, amplify reporting signals, and self-replicate, in order to develop cell-based sensors for detection of chemical and biological targets for environmental and human performance applications. One approach has been to develop cell-based sensors using riboswitches which are regulatory RNAs located in the 5'-untranslated region of messenger RNA sequences. They serve as molecular switches composed of two structural domains: an aptamer domain that binds to a small molecule with specificity and an expression platform that controls the expression of a downstream gene via conformational changes that are induced by small molecule ligand binding to the aptamer domain. The process for selecting a sensing component and the subsequent coupling of this sensing element with a functional reporter (synthesis pathway) is critical for applications to human performance and novel material interfaces. This research will apply the synthetic biology paradigm of the “design, build, and test”, in order to control and coordinate cellular functions. One application of these engineered “sense and respond” cells is in the integration of human microbiome which affects diet metabolism, weight, locomotor function, oxygen utilization, immune function, cognition and even radiation protection; all important attributes for human performance. Bringing this collaboration and expertise to AFRL via the Summer Faculty Fellowship Program will help to shape growing capabilities in this area for the larger DoD Synthetic Biology team.

SF.15.17.B0004: Novel Measures of Human-Machine Trust

Brill, J.

(937) 656-4158

The US Air Force is investing heavily into the development of autonomous systems. As such, there is considerable interested in studying trust in automation/autonomy. Effective human-machine teaming requires appropriate (calibrated) levels of trust. Without calibrated trust, systems may be underused due to operator distrust or overused due to overreliance. Presently, trust is primarily measured through self-report questionnaires, or it is inferred through behavioral measures (such as response time) or physiological indices. The goal of this research is to explore new measures of human-system trues. Novel approaches are appreciated, and may include any combination of task-related surrogates for inferring trust, questionnaires, or physiological indices. A visiting faculty member, if selected, will work alongside the Human Insight and Trust (HIT) Team, a 24-person team dedicated to studying human-machine trust. Team members have numerous basic and applied research projects, and our facilities include an F-16 simulator, a UH-60 Blackhawk simulator, a robotics laboratory, and numerous standalone computer stations.

AFRL/Airman Systems

Dr. Rajesh Naik
711th Human Performance Wing (711 HPW/CL)
2610 Seventh St. Bldg 441, Rm 2-101
WPAFB, OH 45433
Telephone: (937) 255-8222
E-mail: 711.HPW.ChiefScientist@us.af.mil