Home // HUSO 2015, The First International Conference on Human and Social Analytics // View article

Moral Behavior and Empathy Modeling through the Premise of Reciprocity

Authors:
Fernanda M. Eliott
Carlos H. C. Ribeiro

Keywords: Artificial moral machine; Empathy; Biologically inspired architecture; Evolutionary game; Assortativity in networks

Abstract:
We may get the opportunity of conceiving modeling artificial moral behavior and empathy if we renounce the perspective of an immaterial soul playing a role in the process of moral behavior. Philosophers such as Michel de Montaigne wrote that the laws of consciousness, supposed to emerge from nature, are essentially born from custom. Hence, we may provide a basis to that modeling if we pore over moral behavior as a form of cooperation built upon customs among emotions and feelings (as part of cognition). With this perspective in mind, we describe herein a bio-inspired computational multiagent architecture composed of artificial emotions, feelings and by an Empathy Module responsible for providing an action selection that rudimentary mimics moral behavior. The Empathy Module follows a reciprocity assumption as its main design concept. As relations between different subjects can be represented by networks, we explore different network topologies that can characterize the agent-agent interactions, by defining the moral agents neighborhood. For assessment of the proposed architecture, we use a version of an evolutionary game that applies the prisoner dilemma paradigm to establish changes over the network topology. Our results indicate the feasibility of artificial moral behavior leading to cooperative selection of action when applied in environments (networks) whose reciprocity assumption works in accordance with the environmental topology: networks with neutral assortativity w.r.t. node degree (i.e., agent neighborhood size) fit more closely with the leading premise of our Empathy Module than those with a disassortative degree correlation.

Pages: 37 to 44

Copyright: Copyright (c) IARIA, 2015

Publication date: October 11, 2015

Published in: conference

ISSN: 2519-8351

ISBN: 978-1-61208-447-3

Location: St. Julians, Malta

Dates: from October 11, 2015 to October 16, 2015