CCNY Lecture Series
Overview of the Performance and Learning Models (PALM) Research Program
Jerry T. Ball
Warfighter Readiness Research Division
Human Effectiveness Directorate
Air Force Research Laboratory
Challenge and Vision
Much of the long-range planning for warfighter training technologies is built on the premise that an increasing emphasis on modeling and simulation will result in training improvements and cost savings. As our experience with these technologies has matured, it has become apparent that there remain a variety of scientific and technical challenges that must be addressed before maximum realization of these benefits will be achieved. One of these challenges is improving on the relatively poor state of human behavior representation in military simulations. In response to these challenges, the Air Force Research Laboratory’s Warfighter Readiness Research Division has established the Performance and Learning Models (PALM) research program for conducting basic and applied research in human behavior representation. The vision motivating creation of this research program is a revolution in training – more effective and more efficient training leading to a higher level of readiness, made possible through basic and applied cognitive science. We intend to achieve that vision through a dual-emphasis approach involving empirical research and computational cognitive modeling.
Current Research
The PALM team has active research efforts underway in the following
areas:
Cognitive Orientation and Navigation
Language-Enabled Synthetic Entities
Mathematical Models of Learning and Forgetting
Modeling the Effects of Cognitive Moderators on Human Performance
Unmanned Aerial Vehical (UAV) Pilot Modeling
Visuospatial Working Memory
All computational cognitive models developed by the PALM team are implemented using the Atomic Components of Thought – Rational (ACT-R) cognitive architecture and modeling environment running on top of Allegro Common Lisp.
Language-Enabled Synthetic Entities
The VERBOSE (VERbalization Between Operators and Synthetic Entities) project is an applied research effort aimed at the development of language-enabled synthetic entities for use in training simulation environments. The current plan is to merge a cognitive model of a UAV Predator Pilot performing a reconnaissance task with an extended version of a language comprehension model—called Double R Model (Referential and Relational Model). The combined model will be integrated into the Cognitive Engineering Research on Team Tasks (CERTT) testbed developed by Professor Nancy Cooke and associates of Arizona State University—Polytechnic, where the model will perform the role of the UAV pilot as part of a three-man UAV team performing a reconnaissance mission. The research contract awarded to Esther Levin of the City College of CUNY will evaluate the potential for using Latent Semantic Analysis (LSA) to determine word sense frequencies for use as a key part of the word sense disambiguation (WSD) component of the system.
Bio
Jerry T. Ball is a Senior Research Psychologist working for the Air Force Research Laboratory, Human Effectiveness Directorate, Warfighter Readiness Research Division in Mesa, AZ. Dr. Ball earned a PhD in Cognitive Psychology from New Mexico State University and a Masters Degree in Computer and Information Science from the University of Florida. His research interests include the development of cognitively plausible, yet functional language comprehension systems and the modeling of human behavior in complex environments.