Abstract
Can one perceive the shapes of multiple objects in parallel? Or must one "read out" the shapes one-by-one, as if reading words? Our recent experiments support the latter notion. We tested two extreme benchmark models of object shape perception under divided attention: an unlimited-capacity and a fixed-capacity model. In an unlimited-capacity model, shapes are analyzed independently and in parallel. In a fixed-capacity model, shapes are processed at a fixed rate (e.g., serial processing).
METHODS. To measure the capacity limits, we used a variant of the simultaneous-sequential paradigm (Scharff, Palmer, & Moore, 2011a, 2011b). The stimuli were photographs of novel physical objects: foam blocks, lego constructions, and crumply papers. The use of novel objects minimized the role of semantic categorization in the task. Observers searched for a particular exemplar among similar objects from the same set, insuring that non-shape stimulus properties (such as color and texture) could not be used to complete the task. Variable viewpoints were used to preclude image-matching strategies.
RESULTS. The results reject an unlimited-capacity model for shape perception, and are consistent with the predictions of a fixed-capacity model. A similar fixed-capacity limit has been previously observed in tasks that required reading words and making semantic categorizations of images (Scharff, Palmer, & Moore, 2011a; 2011b). Further experiments show that capacity limits are independent of task difficulty and are not mediated by mental rotation of objects.
DISCUSSION. While this study with 3D objects yielded limited capacity results, comparable studies on simple shapes and surface completion yielded results consistent with unlimited capacity. Together, the results indicate a narrow attentional bottleneck in some aspect of processing between elementary surface perception and object shape perception.
Meeting abstract presented at VSS 2012