Humans can perceive constant lightness, even when the illumination environment changes. Lightness perception has been the subject of vision research for many years, and luminance contrast with surroundings was traditionally believed to determine lightness perception. However, several studies have reported reverse contrast phenomena, which demonstrate the shortcomings of theories solely based on immediate contrast. To explain wide lightness/brightness phenomena, two theories have been proposed and are widely accepted to date: namely, the anchoring theory and spatial filtering models. Both can explain reverse contrasts and many lightness illusions, but they continue to exhibit certain weaknesses in the rigor and interpretability of the models. Recent studies suggest novel pathways in this research field. For example, Markov illuminance and reflectance (MIR) is a new computational model that is intuitively understandable and can rigorously account for lightness illusions. The framework of MIR also flexibly welcomes model modifications, which demonstrates its high potential for future developments. Although lightness perception is a classic topic of vision research, the introduction of novel computational methods put forward the possibility of new developments in this research field.
