Prediction as Internal Simulation: Taking Chances in What to Do Next

Malte Schilling,Katharina Rohlfing,Holk Cruse

doi:10.3389/fpsyg.2012.00405

Abstract

Prediction as Internal Simulation: Taking Chances in What to Do Next

Highlights

A commentary on Whatever next? Predictive brains, situated agents, and the future of cognitive science by Clark, A
While we value Clark’s thoughts on the hierarchical prediction machine approach and its implications, one important requirement is not addressed: the connections between different levels of representations are not arbitrary, but it appears that these can be flexibly utilized and in this way different combinations of interconnected – predicting – layers can serve diverse purposes. We put this issue into the viewpoint of representations, central to which functional internal models are
We agree with Clark that these internal models have to be predictive

Summary

Introduction

A commentary on Whatever next? Predictive brains, situated agents, and the future of cognitive science by Clark, A. (in press). While we value Clark’s thoughts on the hierarchical prediction machine approach and its implications, one important requirement is not addressed: the connections between different levels of representations are not arbitrary, but it appears that these can be flexibly utilized and in this way different combinations of interconnected – predicting – layers can serve diverse purposes.

Results

Conclusion