On common factors in visual illusions: A review

Illusions have fascinated humans since ancient times, provoking deep questions about the relationship between perception and what is out there. How can we trust our senses if illusions can deceive us so easily? For these reasons, it seems that illusion research should be at the heart of vision research. However, whereas most scholars appreciate the use of, for example, spatial illusions for spatial vision research, many deny that illusions per se are a proper research topic. One argument is that illusions have very little in common conceptually. Here, we would like to argue that, in fact, it is the other way around. We review many studies, the majority of which show only weak correlations between illusion magnitudes, i.e., the shared variance between illusions is low. A participant can be strongly deceived by one illusion and only weakly by a similar one. Hence, investigating one specific spatial illusion may tell us little about spatial vision in general. Still, we will argue that illusions are a great tool for investigating vision in a principled manner because illusion research can help us to understand how the brain solves the ill-posed problems of vision. For this reason, illusion research in itself is a valuable research topic.