Building life-like 'conscious' software agents

  • Authors:
  • Stan Franklin

  • Affiliations:
  • Institute for Intelligent Systems and Department of Mathematical Sciences, The University of Memphis, Memphis, TN 38152, USA E-mail: stan.franklin@memphis.edu URL: http://www.msci.memphis.edu/~fra ...

  • Venue:
  • AI Communications
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Here we will briefly describe action selection and languagegeneration mechanisms in two 'life-like' software agents, CMattieand IDA, and discuss issues that bear on the topics ofarchitectures for behavior control, interdependencies betweenemotions and goal‐based behaviors, and coordination ofscripted and improvised behaviors. These agents are life-like inthe sense of interacting with humans via email in natural language.They are 'conscious' only in the sense of implementing apsychological theory of consciousness [3,4]. At this writing we areexploring the transition from scripted language production to moreimprovised speech generation. We are also investigatingdeliberative behavior selection mechanisms whereby alternativescenarios are produced and evaluated, and one of them chosen andacted upon. (Supported in part by ONR grant N00014-98-1-0332. Thisresearch was done with essential contributions from the ConsciousSoftware Research Group including Art Graesser, Satish Ambati,Ashraf Anwar, Myles Bogner, Arpad Kelemen, Irina Makkaveeva, LeeMcCauley, Aregahegn Negatu, Uma Ramamurthy, Zhaohua Zhang)