Motion Illusions Generated Using Predictive Neural Networks Also Fool Humans

TL;DR

This paper proposes that visual motion illusions are a result of the brain's predictive processes, demonstrated by a neural network model that creates illusions which also deceive human observers.

Contribution

The introduction of the EIGen generative model that produces visual illusions based on predictive neural networks, supporting the hypothesis about the origin of motion illusions.

Findings

Illusions generated by EIGen are effective on humans.

Illusory motion may stem from the brain's predictive mechanisms.

The model demonstrates a link between neural prediction and perception of motion.

Abstract

Why do we sometimes perceive static images as if they were moving? Visual motion illusions enjoy a sustained popularity, yet there is no definitive answer to the question of why they work. Here we present evidence in favor of the hypothesis that illusory motion is a side effect of the predictive abilities of the brain. We present a generative model, the Evolutionary Illusion GENerator (EIGen), that creates new visual motion illusions based on a video predictive neural network. We confirm that the constructed illusions are effective on human participants through a psychometric survey. Our results support the hypothesis that illusory motion might be the consequence of perceiving the brain's own predictions rather than perceiving raw visual input from the eyes. The philosophical motivation of this paper is to call attention to the untapped potential of "motivated failures", ways for artificial systems to fail as biological systems fail, as a worthy outlet for Artificial Intelligence and Artificial Life research.