Investigating Local and Configural Shape Processing with Steady-State Visual Evoked Potentials

Abstract

Object recognition relies on shape, comprised of both local and configural shape information. Local shape information involves elements such as line curvature and edge angles confined to specific regions and perceived independently, while configural shape information arises from the spatial arrangement of local shape features. Previous research suggests that humans primarily use edges and contours for object recognition and are sensitive to disruptions in configural shape. However, it remains unclear how the human visual system separately encodes local and configural shape information, and whether distinct neural mechanisms underlie these processes. We presented stimuli manipulating local and configural shape information independently to participants while recording EEG responses using a Steady-State Visual Evoked Potentials paradigm. Our results reveal that object recognition depends on two mechanisms: one more transient that is sensitive to local shape localized in occipital regions, and one sustained and sensitive to configural shape manipulations localized in right temporal cortex.