 |
About the theory
The PSI theory is a theory of human action regulation by psychologist Dietrich Doerner [1, 2]. It is an attempt to create a body-mind link for virtual agents. It aims at the integration of cognitive processes, emotions and motivation. This theory is unusual in cognitive psychology in that emotions are not explicitly defined but emerge from modulation of perception, action-selection, planning and memory access. Emotions become apparent when the agents interact with the environment and display expressive behavior, resulting in a configuration that resembles emotional episodes in biological agents. The agents react to the environment by forming memories, expectations and immediate evaluations. This short presentation is a good overview.
PSI agents possess a number of modulators and built-in motivators that lie within a range of intensities. These parameters combined to produce complex behavior that can be interpreted as being emotional. Additionally, by regulating these parameters, the agent is able to adapt itself to the different circumstances in the environment.This theory has been applied to different virtual agent simulations in different types of environments [3, 4, 5, 6, 7, 8] and has proven to be a promising theory for creation of biologically plausible agents.
More information can be found at http://web.uni-bamberg.de/ppp/insttheopsy/ (In German but there are some papers in English). Some simulation software is also available for download from this site.
References
[1] Doerner, D. The mathematics of emotions. In Frank
Detje, D.D., Schaub, H., eds. Proceedings of the Fifth InternationalConference on Cognitive Modeling, Bamberg, Germany,(2003) 7579
[2] Katrin Hille, A Theory on Emotion, http://web.uni-bamberg.de/ppp/insttheopsy/dokumente/Hille_A_theory_of_emotion.pdf
[3] D. Doerner and K. Hille. Artificial souls: Motivated emotional
robots. In Proceedings of the International Conference on Systems, Man and
Cybernetics, pages 38283832, 1995.
[4] Bartl, C., Doerner, D. Comparing the behavior of psi
with human behavior in the biolab game. In Ritter, F.E., Young, R.M., eds.:
Proceedings of the Second International Conference on Cognitive Modeling, Nottingham, Nottingham University Press (1998) http://web.uni-bamberg.de/ppp/insttheopsy/dokumente/BartlDoerner_Comparing_the_behaviour_of_PSI_with_human_behaviour_in_the_BioLab_game_(Memo32).pdf
[5] Mei Yii Lim, Ruth Aylett, Christian Martyn Jones, Emergent
Affective and Personality Model,The 5th International Working
Conference on Intelligent Virtual Agents (IVA05), LNAI 3661, Kos, Greece,
September 12-14, 2005 http://www.macs.hw.ac.uk/~ruth/Papers/agents-affect/IVA2005mei.pdf
[6] M.Y.
Lim, R. Aylett and C.M. Jones Affective Guide with Attitude,
Proceedings of the 1st International Conference on Affective Computing and
Intelligent Interaction, Beijing, China, 22-24 October, 2005. http://www.macs.hw.ac.uk/~ruth/Papers/agents-affect/ACII05-mei.pdf
[7] Harald Schaub, Introduction and Delineation of the
Investigation, June 2, 2004 http://web.uni-bamberg.de/ppp/insttheopsy/dokumente/Schaub_The_year_of_the_gardener._Behaviour_modelling_in_a_complex_situation._(Memo-MPG09).pdf
[8] Joscha Bach, The MicroPsi Agent Architecture. Proceedings of ICCM-5, Universitäts-Verlag Bamberg, 2003. http://www.macs.hw.ac.uk/~ruth/Papers/psi-papers/bach03micropsi.pdf
[9] Joscha Bach, Enhancing Perception and Planning of Software Agents with Emotion and Acquired Hierarchical Categories. Proceedings of MASHO 02, German Conference on AI KI 2002, Karlsruhe, Germany 2002. http://www.macs.hw.ac.uk/~ruth/Papers/psi-papers/BachMasho2002.pdf
|
