Abstract
A number of psychology and neuroscience studies have used “simple versus complex” objects as a manipulation, and brain areas such as the lateral occipital complex (LOC) have shown greater activation in response to more complex objects. However, although everyone has certain intuitions about visual complexity, it is a concept that is difficult to formally define or empirically measure. As object complexity increases, so does the potential for low-level confounds such as the number of lines that define its shape, average luminance, spatial frequency, etc. In this study, we sought to examine what object properties might contribute to complexity and how that complexity affects responses in visual brain regions, while controlling for low-level differences between objects as much as possible. In order to do so, we presented a unique type of block-design fMRI task in which “simple” and “complex” shapes were formed entirely from illusory contours (ICs). All shapes appeared inside circles of a rotating, randomly generated plaid, a region of whose interior rotated in the opposite direction, with the object edges defined by the boundaries of the different rotation directions. Critically, these plaids were designed to saturate the neural and BOLD response to low-level visual properties, so that any differences in activation would be due solely to the shape of the IC. In addition to the base plaid (no IC shape), seven conditions of IC shapes were presented, varying in symmetry, regularity, and the number of edges and corners. Randomly generated irregular decagons were the most “complex” shapes; simpler shapes included circles, squares, regular decagons, and regular ten-sided stars, equated with the irregular decagons in diameter and/or area. We found that basic shapes elicited a weak response; however, as shape complexity increased, we observed increasing activation in extrastriate cortex, especially lateral occipital areas.
Acknowledgement: Supported by NSF/EPSCoR grant #1632849 to MRJ and colleagues