A recurrent model for perceptual learning

Perceptual learning has received enhanced interest during the last years both from theoreticians and experimentalists. Recent experimental results reveal that mechanisms underlying perceptual learning are more complex than previously expected, thereby ruling out any explanations based on simple neural network models. These findings do not represent an insignificant exception to the rule but are evidence that present models fail to reflect some important characteristics of the learning process. A new model is introduced that is able to overcome problems of classical neural networks and that might be viewed as a hybrid between supervised and unsupervised learning