Home // PATTERNS 2011, The Third International Conferences on Pervasive Patterns and Applications // View article

Patterns of Emotion Driven by Affect State and Environment

Authors:
Paul Joseph
Haim Levkowitz

Keywords: patterns; affect; emotions; Panksepp; autonomous agent; mammalian emotion

Abstract:
The neuroscientist Jaak Panksepp posits the existence of seven physical systems in the mammalian brain, that when simulated in the laboratory, result in emotions identical to known emotions. These seven systems are SEEKING, RAGE, PANIC, FEAR, LUST, CARE, and PLAY. From the perspective of mathematics, these emotions form the dimensions of a phase space in which every emotion is located and modeled as a dynamical system. Inputs from outside the system interact with internal values and inform the dynamical system, resulting in a simulated affect that in turn drives observed patterns of emotion. This paper discusses a framework built to explore this premise. As a first cut it uses a linear seven-variable dynamical system to drive a search heuristic for a utility-based reflex agent. The results indicate that even this simple linear model shows the basic patterns of emotion observed in mammals. These patterns of emotion facilitate discovery and decision. It provides an independent “always on, automatic” discovery and decision system capable of adaptive goal setting, which works without the need for significant additional cognitive analysis. It offers the basis of a framework that is extendable to include additional sensory information, learning, memory/persistence, and an independent cognitive system.

Pages: 47 to 52

Copyright: Copyright (c) IARIA, 2011

Publication date: September 25, 2011

Published in: conference

ISSN: 2308-3557

ISBN: 978-1-61208-158-8

Location: Rome, Italy

Dates: from September 25, 2011 to September 30, 2011