New Models for Old Questions: Evolutionary Robotics and the ‘A Not B’ Error

摘要: In psychology the 'A not B' error, whereby infants perseverate in reaching to location where a toy was previously hidden after it has been moved new location, subject of fifty years research since first identified by Piaget [1]. This paper describes novel implementation error paradigm which is used test notion that minimal systems evolutionary robotics modelling can be explore developmental process and generate hypotheses for natural experimental populations. The model demonstrates agents controlled plastic continuous time recurrent neural networks perform task homeostatic mediation plasticity produce perseverative patterns similar those observed human infants. addition, shows trend production errors reduce during development.