Abstract
Visual processing is constrained by bottlenecks of space and attention. Interestingly, both of these bottlenecks can occur at a large scale—the visual hemifield. For example, object-selective neurons have large receptive fields, often encompassing the entire left- or right-visual field, and attention within each hemifield is supported by independent resources. Notable accounts of object representation (e.g., OSM) propose that iterative processing is engaged to overcome these bottlenecks, continuing until ambiguity about an object's identity and location is resolved. According to this framework, visual awareness of an object should also depend on the completion of this process, and should be gated by processing constraints within the left- and right-visual hemifields. Here, we tested and confirmed this hypothesis. We presented quartets of objects scattered among a variety of peripheral locations. Quartets appeared for 20-msec, with spatial arrangements either within a hemifield or spread across the left- and right-hemifields. Objects included gabors, inverted mooney-faces, and upright mooney-faces, each with a leftward- or rightward-tilt. We used OSM to mask (and cue) one of the peripheral locations, which may or may not have contained an object. On each trial, observers indicated (1) whether an object was present at the masked location, and (2) whether it was tilted to the left or right. Not surprisingly, orientation discrimination was worse when objects were presented within a hemifield compared to across hemifields. More important, within-hemifield arrangements profoundly disrupted detection of an object's presence. These effects occurred for each object type, presumably reflecting the operation of a processing bottleneck that applies to multiple levels of stimulus complexity and/or stages of visual analysis. Our findings cannot be accounted for by differences in crowding or visual acuity. Rather, we have demonstrated a more general phenomenon in which the hemifield-based architecture of visual processing gates access to visual awareness.
Meeting abstract presented at VSS 2018