The grounding of motivation in artificial animals: Indices of motivational behavior

  • Authors:

  • Tony Savage

  • Affiliations:

  • School of Psychology, The Queen's University of Belfast, Belfast BT7 1NN, UK

  • Venue:
  • Cognitive Systems Research
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

The grounding issue in AI research is concerned with how best to integrate an artificial creature, such as a robot, with the environment in which it functions. In many investigations these interactions are defined and imposed on the agent by an external source, e.g. the designer of the robot. In the debate on the grounding issue little or no attention has been paid to the motivational aspects of grounding. The subject area of motivation is central to understanding how an organism is capable of autonomous functioning, i.e. deciding what is important to its continued existence and determining the means of achieving the relevant goals. This paper sets out to examine this aspect of the grounding debate and identifies some of the features which characterize authentic motivational structures in animals and discuss the possibility of implementing similar structures in artificial creatures. It will be argued that an appreciation of the interactive nature of motivation may be the key to understanding motivational grounding.