Modeling infant object perception as program induction

TL;DR

This paper shows that a domain-general learning system can induce core object properties like rigidity and persistence from limited data, challenging the idea that specialized modules are necessary for infant object perception.

Contribution

It demonstrates that a general inductive learning model can replicate infant-like object perception without specialized core systems, using minimal environmental data.

Findings

Model induces core object properties from few examples

Replicates infant object perception phenomena

Challenges the necessity of specialized object recognition modules

Abstract

Infants expect physical objects to be rigid and persist through space and time and in spite of occlusion. Developmentists frequently attribute these expectations to a "core system" for object recognition. However, it is unclear if this move is necessary. If object representations emerge reliably from general inductive learning mechanisms exposed to small amounts of environment data, it could be that infants simply induce these assumptions very early. Here, we demonstrate that a domain general learning system, previously used to model concept learning and language learning, can also induce models of these distinctive "core" properties of objects after exposure to a small number of examples. Across eight micro-worlds inspired by experiments from the developmental literature, our model generates concepts that capture core object properties, including rigidity and object persistence. Our findings suggest infant object perception may rely on a general cognitive process that creates models to maximize the likelihood of observations