Simulation of emotions of agents in virtual environments using neural networks

A.-J. van Kesteren, A.J. van Kesteren, Hendrikus J.A. op den Akker, Mannes Poel

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    51 Downloads (Pure)


    A distributed architecture for a system simulating the emotional state of an agent acting in a virtual environment is presented. The system is an implementation of an event appraisal model of emotional behaviour and uses neural networks to learn how the emotional state should be influenced by the occurrence of environmental and internal stimuli. A part of the modular system is domain-independent. The system can easily be adapted for handling different events that influence the emotional state. A first prototype and a testbed for this architecture are presented.
    Original languageUndefined
    Title of host publicationLearning to Behave: Internalising Knowledge
    EditorsK. Jokinen, Dirk K.J. Heylen, Antinus Nijholt
    Place of PublicationEnschede
    PublisherUniversity of Twente
    Number of pages11
    Publication statusPublished - 20 Nov 2000
    EventLearning to Behave: Internalising Knowledge - Enschede, the Netherlands
    Duration: 22 Nov 200024 Nov 2000

    Publication series

    NameTwente Workshop on Language Technology
    PublisherUniversity of Twente
    ISSN (Print)0929-0672


    WorkshopLearning to Behave: Internalising Knowledge
    Other22-24 November 2000


    • HMI-IA: Intelligent Agents
    • METIS-119683
    • IR-67104
    • EWI-9819

    Cite this