Face processing emerges from object-trained convolutional neural networks

TL;DR

This study investigates whether face recognition abilities can develop in convolutional neural networks trained only on general objects, testing the domain-general versus domain-specific theories of face processing.

Contribution

It demonstrates that face recognition can emerge in object-trained CNNs without face-specific training, challenging the idea that specialized mechanisms are necessary.

Findings

Object-trained CNNs can recognize faces and face-like stimuli.

Face processing emerges without face-specific training.

Results support domain-general mechanisms in face recognition.

Abstract

Whether face processing depends on unique, domain-specific neurocognitive mechanisms or domain-general object recognition mechanisms has long been debated. Directly testing these competing hypotheses in humans has proven challenging due to extensive exposure to both faces and objects. Here, we systematically test these hypotheses by capitalizing on recent progress in convolutional neural networks (CNNs) that can be trained without face exposure (i.e., pre-trained weights). Domain-general mechanism accounts posit that face processing can emerge from a neural network without specialized pre-training on faces. Consequently, we trained CNNs solely on objects and tested their ability to recognize and represent faces as well as objects that look like faces (face pareidolia stimuli).... Due to the character limits, for more details see in attached pdf