Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster F in Poster Session F: Thursday, August 6, 1:45 – 3:30 pm, Kimmel Center, Shorin & Rosenthal Rooms

Spatial Attention Enhances Figure-Ground Modulation in Human Early Visual Cortex

Emil Olsson¹, Michael Epstein², Juneau Wang³, Karen Tian³, Brian Maniscalco¹, Jennifer Motzer³, Minh Nguyen¹, Meera Sriram¹, Olenka Graham Castaneda¹, Angela Shen¹, Sydney Liu¹, Yuzheng Wu¹, Maggie Zhang¹, Diana Gamboa³, Richard Brown⁴, Victor A.F. Lamme⁵, Hakwan Lau⁶, Biyu J. He⁷, Jan W Brascamp⁸, Ned Block², David Chalmers², Rachel N. Denison³, Megan AK Peters¹; ¹University of California, Irvine, ²New York University, ³Boston University, ⁴City University of New York, ⁵University of Amsterdam, ⁶Sungkyunkwan University, ⁷NYU Langone Health, ⁸Michigan State University

Presenter: Emil Olsson

Figure-ground segregation requires the visual system to assign local image elements to a figure or to its background based on the broader context of the scene. In primate visual cortex, figure-ground modulation (FGM) has been linked to recurrent processing, and human fMRI studies have shown FGM in V1--V3. However, it remains unclear how attention interacts with FGM, with mixed empirical results. We analyzed fMRI data from 30 adults viewing texture-defined figure-ground stimuli. Participants maintained central fixation, while texture-defined targets appeared in four peripheral quadrants. Central cues directed covert spatial attention to one of the four target locations, leaving the remaining three unattended. Using population receptive field mapping and an independent spatial localizer, we isolated the parts of V1, V2, and V3 responsive to the target locations. We then quantified FGM using a figure-present minus figure-absent contrast, and examined how attention changed FGM across V1-V3. FGM occurred for both attended and unattended targets, but attention increased FGM, with the same qualitative pattern present across V1-V3. These results indicate that attention enhances the separation between figure-present and figure-absent responses, improving figure-ground segregation at early stages of visual processing.

Topic Area: Computational Models of Vision & Visual Cortex