Shaping and incentive learning: a reply to Marco Dorigo

The interaction of learning and motivational mechanisms lies at the heart of many organisms´ adaptive responses to a complex and changing world. Our present understanding of this interaction is far from satisfactory, in large measure due to our ignorance of how events acquire motivational significance and how these motivational factors then interact with various forms of learning. A realisation of this weakness in relation to animal psychology is reflected in recent experimental work, e.g. the influence of innate behaviour systems on learning and the renewed interest in incentive motivation. Modelling such interactions with simulated animals and environments represent another more recent means of developing performance theories of behaviour. This course of action is recognised by Dorigo and Colombetti (1997) when they refer to running an animat computer program with a specific set of parameters as an `experimental subject´. Experimental work with real animals has the virtue of biological credibility; however the simulation studies approach represents a new and significant testing ground for performance theories of behaviour