An architecture for emotional decision-making agents

Abstract

Our research focuses on complex agents that are capable of interacting with their environments in ways that are increasingly similar to individual humans. In this article we describe a cognitive architecture for an interactive decisionmaking agent with emotions. The primary goal of this work is to make the decision-making process of complex agents more realistic with regard to the behavior moderators, including emotional factors that affect humans. Instead of uniform agents that rely entirely on a deterministic body of expertise to make their decisions, the decision making process of our agents will vary according to select emotional factors affecting the agent as well as the agent's parameterized emotional profile. The premise of this model is that emotions serve as a kind of automatic assessment system that can guide or otherwise influence the more deliberative decision making process. The primary components of this emotional system are pleasure/pain and clarity/confusion subsystems that differentiate between positive and negative states. These, in turn, feed into an arousal system that interfaces with the decision-making system. We are testing our model using synthetic specialforces agents in a reconnaissance simulation. Categories and Subject Descriptors General TermsHuman Factors. KeywordsEmotion, arousal, pleasure, pain, clarity, confusion. EMOTIONS AND DECISION-MAKINGThis paper describes a framework for modeling emotions in an interactive, decision-making agent intended to provide a nearly human level of competence in a focused task domain. We regard emotions essentially as subconscious signals and evaluations that inform, modify, and receive feedback from a variety of sources including higher cognitive processes and the sensorimotor system. Because our project focuses on decision making, our model emphasizes those aspects of emotion that influence higher cognition and not those that affect, for example, the immune system. We are integrating a connectionist model of emotions from Chown [1] with Rosenbloom, Laird, and Newell's [4] Soar architecture. The application area incorporates emotions and individual differences into the behavior models of synthetic special-forces agents in an army reconnaissance simulation.In our framework, symbolic assessments of a small set of "emotional attributes" reside in a working memory. Portions of working memory are accessible by the deliberate cognitive process, and portions are accessible by the emotion mechanisms, so memory serves as the interface between the two. These working memory elements combine with background knowledge to generate strategies, reasoning, and external behavior. At the same time, the cognitive model creates working interpretations of the environment and status of internal goals (situational awareness). Some of these interpretations and assessments feed into the connectionist model, which in turn continuously computes new values for each emotional attribute. Because the theory underpinning our model assumes that these responses were ultimate...

An approach for emotions and behavior modeling in a crowd in the presence of rare events

“…Research has showed that emotions have an important impact on the performance of most of our daily activities such as communication, planning, behavior and decision making (Scherer, 2005). It is assumed that emotions are the catalyst of the brain activity that directs our attention and behavior; also, they inform us of the importance of events that occur around us (Chown, Jones, & Henninger, 2002). In this section, we present our approach for behavior modeling.…”

Cognitive Modeling

“…This is a clear indicator that people do not treat information neutrally. As Kaplan and others [Chown, et al 2002] have argued, the human emotional system can be understood in these terms. The argument is based upon the idea that human emotions address the need our ancestors had to make very fast decisions in encounters with other dangerous predators.…”