Novelty recognition models

Abstract

In the following chapter a theoretical model of novelty recognition mechanism is described. In the first part the insights as well as shortcomings of classic approaches (e.g. mere exposure, feeling of knowing, tip of the tongue) are discussed. Those models are mostly descriptive and do not provide the explanatory mechanism for the observed effects. Such a mechanism is proposed in the second part of the chapter. First a metacognitive mechanism of fluency is described and identified as an empirical counterpart of a neural network measure of coherence. The application of a coherence/fluency-based feedback loop allows for building a precise and parsimonious model of novelty recognition process. © 2013 Springer-Verlag Berlin Heidelberg.