Hypothesis & Theory ARTICLE

Front. Psychol., 12 November 2012 | doi: 10.3389/fpsyg.2012.00494

Too good to be true? Ideomotor theory from a computational perspective

  • 1Department of Computer Science, Faculty of Science, Eberhard Karls Universität Tübingen, Tübingen, Germany
  • 2Department of Psychology, Faculty of Science, Eberhard Karls Universität Tübingen, Tübingen, Germany

In recent years, Ideomotor Theory has regained widespread attention and sparked the development of a number of theories on goal-directed behavior and learning. However, there are two issues with previous studies’ use of Ideomotor Theory. Although Ideomotor Theory is seen as very general, it is often studied in settings that are considerably more simplistic than most natural situations. Moreover, Ideomotor Theory’s claim that effect anticipations directly trigger actions and that action-effect learning is based on the formation of direct action-effect associations is hard to address empirically. We address these points from a computational perspective. A simple computational model of Ideomotor Theory was tested in tasks with different degrees of complexity. The model evaluation showed that Ideomotor Theory is a computationally feasible approach for understanding efficient action-effect learning for goal-directed behavior if the following preconditions are met: (1) The range of potential actions and effects has to be restricted. (2) Effects have to follow actions within a short time window. (3) Actions have to be simple and may not require sequencing. The first two preconditions also limit human performance and thus support Ideomotor Theory. The last precondition can be circumvented by extending the model with more complex, indirect action generation processes. In conclusion, we suggest that Ideomotor Theory offers a comprehensive framework to understand action-effect learning. However, we also suggest that additional processes may mediate the conversion of effect anticipations into actions in many situations.

Keywords: ideomotor theory, associative learning, computational model, planning, consolidation

Citation: Herbort O and Butz MV (2012) Too good to be true? Ideomotor theory from a computational perspective. Front. Psychology 3:494. doi: 10.3389/fpsyg.2012.00494

Received: 31 July 2012; Accepted: 24 October 2012;
Published online: 12 November 2012.

Edited by:

Wilfried Kunde, Julius-Maximilians-Universität Würzburg, Germany

Reviewed by:

Digby Elliott, Liverpool John Moores University, UK
Markus Paulus, Ludwig Maximilian University Munich, Germany

Copyright: © 2012 Herbort and Butz. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in other forums, provided the original authors and source are credited and subject to any copyright notices concerning any third-party graphics etc.

*Correspondence: Oliver Herbort, Department of Psychology, Julius-Maximilians-Universität Würzburg, Röntgenring 11, 97070 Würzburg, Germany. e-mail: oliver.herbort@psychologie.uni-wuerzburg.de

Present address: Oliver Herbort, Department of Psychology, University of Würzburg, Würzburg, Germany.

Back to top