Visual complexity exerts opposing effects on object categorization and identification

Abstract

Visual complexity is a quality of objects that can be understood in terms of information richness: The more complex an object is, the more information it contains. Visual complexity is therefore likely to have a considerable impact on visual object processing. Here we demonstrate that visual complexity does indeed affect visual object processing but that it exerts different effects depending on task requirements. In tasks that require little differentiation among objects (superordinate categorization), reaction times decrease linearly as a function of increasing visual complexity. In tasks that require more differentiation (difficult object decision), complexity exerts the opposite effect. We argue that this dynamic interplay between visual complexity and object individuation can be explained by assuming that complexity affects two operations in visual object recognition differentially: When perceptual representations are formed it exerts a negative effect because representations of complex objects are harder to assemble than representations of less complex objects, but once formed, it exerts a positive effect because complex objects—due to their information richness—activate fewer candidate representations in visual long-term memory that need to be differentiated in the recognition process.

Keywords:

• Categorization
• Object decision
• Object recognition
• Visual complexity
• Visual similarity

J. Frederico Marques was sponsored by a Foundation for Science and Technology of Portugal Project Grant PTDC/PSI-PCO/114822/2009. Christian Gerlach is grateful to the Friends of Fakutsi Foundation.

J. Frederico Marques was sponsored by a Foundation for Science and Technology of Portugal Project Grant PTDC/PSI-PCO/114822/2009. Christian Gerlach is grateful to the Friends of Fakutsi Foundation.