One of the goals of the Defense Modeling and Simulation Office (DMSO) has been to promote the development and assessment of computational human behavior representations (HBRs) that potentially provide synthetic forces -- both Red and Blue -- for live, virtual, and constructive military simulations. This paper reviews the domain of HBRs that could be integrated into military simulations. The intent is to provide the modeling and simulation (M & S) community an understanding of specific HBR models and to identify specific interoperability problems. The study identified 19 different HBRs that have at least some applicability to military simulations. The following aspects of each model were reviewed: Model Purpose and History of Development, Principal Metaphors and Assumptions, Cognitive/Behavioral Functions Represented, Applications, Technical Considerations, and Evaluation. Each model was evaluated on whether it supported the following cognitive and behavioral functions: perception, learning, psychomotor performance, decision making, attention, problem solving, situation awareness, cognitive workload, short-term memory, emotional behavior, long-term memory, and social behavior. Analyses of these models suggested the following generalizations concerning the current state of the art in human behavior modeling: (1) decision making is a universal function of all models; (2) all models can represent some form of memory storage and retrieval functions; (3) both the "front-end" of cognition (perception and attention) and cognitive output (psychomotor action) are represented in most models; (4) because most models do not include learning functions, they may not react appropriately to novel situations; (5) the capability to emulate situational awareness is explicitly represented in only a few models; and (6) very few of the models have the capability to simulate emotional or social behaviors. (5 tables, 204 refs.).