Facenet: A connectionist model of face identification in context

Publisher Website
Google Scholar
Abstract
The role of contexts in face identification constitutes a weak point of existing cognitive models of face recognition. A connectionist system (Face-net) based on a layered network has been specified and implemented to investigate the processes underlying identification. The architecture of the Face net system takes contextual information explicitly into account in the construction of identity representations, and is provided with a reinjection mechanism which gives it dynamic properties. The model proposes that three indicators are extracted in parallel in person identification from a face: familiarity feeling (feeling of déjà-vu of the face stimulus), identity feeling (feeling that we know the person) and identity content (information about the person resulting from the integration of the contexts). Face net underwent an experimental procedure to study the structuring of identity representations in various learning conditions defined by the specificity and the variability of the encoding context. The simulation results showed a significant interaction in identification performance between the variability and specificity factors. Identification of faces learned in variable contexts was not affected by a contextual change during recognition, whereas non-variable faces were affected, and all the more so when their encoding contexts were non-specific. These results are discussed in terms of generalisation (through a “semantisation” process) and categorisation in a contextual distributed memory. Of course, this kind of result has no ecological validity, but the model offers new predictions for further experiments on real subjects.

This publication has 22 references indexed in Scilit: