There has been much research investigating team cognition, naturalistic decision making, and collaborative technology as it relates to real world, complex domains of practice. However, there has been limited work in incorporating naturalistic decision making models for supporting distributed team decision making. The aim of this research is to support human decision making teams using cognitive agents empowered by a collaborative Recognition-Primed Decision model. In this paper, we first describe the architecture of RPD-enabled agent (RPD-agent), in which we have implemented an internal mechanism of decision-making adaptation based on collaborative expectancy monitoring, and an information exchange mechanism driven by relevant cue analysis. We have evaluated RPD-agents in a real-time simulation environment, feeding teams with frequent decision-making tasks under different tempo situations. While the result conforms to psychological findings that human team members are extremely sensitive to their workload in high-tempo situations, it clearly indicates that human teams, when supported by RPD-agents, can perform better in the sense that they can maintain team performance at acceptable levels in high time pressure situations.

One research focus of human-centered teamwork is on advanced decision architectures that can help people make effective and timely decisions. This requires distributed team members to effectively establish shared situation awareness and to collaboratively develop explanations on how an unfamiliar situation might have been emerging. One key to achieve this goal is the ability to anticipate others' future information needs and to offer help proactively. In this paper we investigate a novel approach to anticipating teammates' information needs based on stepwise conversation pattern recognition, leveraging the idea of multi-party communication. This approach can be further extended to build a computational model for collaborative story building as needed in Recognition-Primed, naturalistic decision architectures.

Helping behavior in effective teams is achieved via some overlapping “shared mental models ” that are developed and maintained by members of the team. In this paper, we take the perspective that multiparty “proactive ” communication is critical for establishing and maintaining such a shared mental model among teammates, which is the basis for agents to offer proactive help and to achieve coherent teamwork. We first provide formal semantics for multiparty proactive performatives within a team setting. We then examine how such performatives result in updates to mental model of teammates, and how such updates can trigger helpful behaviors from other teammates. We also provide conversation policies for multiparty proactive performatives. 1.

We manipulated two variables predicted by models of team performance to affect team processes and performance: team organization and task load. In one team organization the team leader served as a manager only, without responsibility for prosecuting hostile tracks. In the other organization the team leader served as a player-manager. Task load was established by varying the number of hostile tracks. As hypothesized, teams performed at a higher level, that is took less time to act, when task load was low than high. Counter to predictions no performance differences were found between the two team organization conditions, but as anticipated, performance differed among team members by task load. Additionally, teams in which the team leader was a manager only were more proactive and transferred more information to other teammates without requests to do so, a communication pattern that has been shown to be indicative of high-performing teams.

This simulation experiment is the latest in a series conducted by the Adaptive Architecture for Command and Control (A2C2) research team. The focus was to evaluate the relative performance to two organizational structures on tasks that varied in terms of complexity and predictability. One structure represented a more traditional, functional form with interdependent nodes. The second structure was derived from computer-modeling to reduce the need for coordination by creating fairly autonomous divisional units. Results from a previous A2C2 experiment suggested that the more autonomous, divisional structure, while outperforming the functional structure in planned mission tasks, could be less effective with complex unpredictable tasks. Organization theory argues that coordination capability is an important factor in an organization’s ability to respond to an uncertain and complex environment. The question examined in this research was whether the different degrees of coordination capability developed by these two structures would influence the performance and process outcomes for both predictable and unpredictable tasks. The results show only limited differences in the results for the two structures, though these are in the direction predicted above. However, a more consistent

In this paper, we present techniques for modeling the decision-making processes of a team of synthetic agents operating in task selection and resource allocation settings within the third generation distributed dynamic decision-making (DDD-III) paradigm. The DDD-III simulator provides a controllable, multi-player, multi-platform organizational environment. The paper provides two major contributions. First, motivated by the need for a network of intelligent agents within C2 experimental settings, a brief overview of modeling techniques for the design of a network of collaborating agents is provided. Second, techniques for modeling the decision-making processes of synthetic agents in task selection and resource allocation settings are presented. In the proposed framework, the decision-making processes of a network of intelligent agents are addressed via limited look-ahead, auctionbased scheduling and resource allocation algorithms from the phase I of the three-phase organizational design process. Preliminary results of operationalizing the DDD-based multi-agent-network paradigm are presented in two different mission scenarios. A coordination-free scenario illustrates the basic structures of the intelligent agent design, in terms of stimulus-hypothesis-option-response (SHOR) model-based three-stage decision-making process. The second example, derived from the A2C2 Experiment 8, highlights the potential of utilizing the agent framework in C2 experiments. I.