Extending the recognition-primed decision model to support human-agent collaboration

Abstract

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 an RPDenabled agent architecture (R-CAST), 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 R-CAST agents in a real-time simulation environment, feeding teams with frequent decisionmaking tasks under different tempo situations. While the result conforms to psychological findings that human team members are extremely sensitive to their workload in hightempo situations, it clearly indicates that human teams, when supported by R-CAST agents, can perform better in the sense that they can maintain team performance at acceptable levels in high time pressure situations.

Communications for Agent-Based Human Team Support

“…Agents aiding teams (Lenox et al, 1997(Lenox et al, , 1998, face a different set of problems: (1) identifying information that needs to be passed to other team members before being asked; (2) automatically prioritizing tasks for the human team members; (3) maintaining shared task information in a way that is useful for the human users. Agents assuming the role of equal team members (Traum et al, 2003;Fan et al, 2005Fan et al, , 2006) must additionally be able to: (1) competently execute their role in the team; (2) critique team errors; (3) independently suggest alternate courses of action. Perhaps because of these challenges, there are very few prior results on human-agent team aiding and teamwork.…”

“…Perhaps because of these challenges, there are very few prior results on human-agent team aiding and teamwork. Examples of tasks that were investigated include target identification (Lenox et al, 1997(Lenox et al, , 1998, achievement of a military rendezvous plan and delivery of supplies to troops (Fan et al, 2005(Fan et al, , 2006. All of this prior work has uniformly found that human-agent teams exhibited superior performance over human-only teams not only in achievement of task objectives but also in performance stability.…”

“…Fan et al (2005Fan et al ( , 2006 have evaluated the use of cognitive agents within a collaborative Recognition-Primed Decision model (RPD) for supporting human teams performing military decision tasks. In one task (Fan et al, 2005), the human teams had to maximize the amount of supplies delivered to the troops by successfully protecting an airport. The agents were able to participate in human teams and increase the amount of supplies delivered when the humans were under time-stress.…”

Crowd modeling and simulation technologies

“…RPD is believed to be a more realistic account of how human make decisions in everyday situations (particularly, time-critical situations), which emphasizes the role of experiences and knowledge in decision making. Recently, RPD based models have been used in human-agent collaboration [Fan et al 2005] and military simulations [Ting and Zhou 2008].…”

Activity Recognition for Agent Teams

“…Agents that support individual human team members face the following challenges: (1) modeling user preferences; (2) determining optimal transfer-of-control policies [84]; (3) considering the status of user's attention in timing services [45]. Agents aiding teams [60][61][62][63], face a different set of problems: (1) identifying information that needs to be passed to other team members before being asked; (2) [62,63], achievement of a military rendezvous plan [60,61] and delivery of supplies to troops [31,32]. All of this prior work has uniformly found that human-agent teams exhibited superior performance over human-only teams not only in achievement of task objectives but also in performance stability.…”