The present study focused on essentially the whole/part visual analysis taken at its most basic and asked whether a single part can be defined as local info rmation. Although some researchers have implied that a single part can be defined as local, none have directly inve stigat ed this issue.
A divided-attention procedure, based on Rensink's (1997) 'change blindness' paradigm, was developed to answer this (and related) questions. Changes were made to the global or local levels of letter and shape stim uli. Participants indicated if they had detected a change and whether the change was global or local.
The first set of experiments tested Fink et al's (1997) stimulus category theory, which argues that local information is processed in the hemisphere best suited for processing a particular stimulus category. Combinations of hierarchical letter and shape stimuli were used. All demonstrated a tendency for global changes to be detected with greater accuracy in the right visual field and local changes in the left visual field (LVF), providing evidence against a theory of stimulus
category.
The second set of experiments asked whether a single part could be defined as local information. A new stimulus was devised to specifically test this question, namely a 'shape in shape' stimulus. A large 'global' shape encompassed a smaller 'local' shape. The first experiment resulted in a LVF bias for global and local change detection. To investigate the possible cause of the LVF bias subsequent experiments manipulated various factors, including the number, the location, and the
size of the parts. The results suggest that the relative size of the smaller shape has a direct impact on the hemispheric asymmetry of 'global' and 'local' processing. The findings also suggest that a single part can be processed as local information.
The final experiment applied the idea of a single global and local shape to more ecologically valid stimuli, namely products. Results mirrored those of the first set of experiments.