Visual scenes are cluttered and complex, making it necessary to prioritize which objects and regions in an image should be processed. But how does the visual system decide where and what to attend to? We propose that the entire scene, the visual context, guides spatial attention towards objects which are behaviorally relevant to that context, a process that we call contextual cueing. This guidance is driven by implicit memory representations which are acquired incidentally.
Contextual information is useful because regularities in the visual environment are presented to observers in the form of visual context. Thus, a core theme of our lab is to characterize how regularities in the visual environment are encoded. As one example, we are exploring the neural substrate of implicit contextual learning. Initial work suggests that contextual learning is subserved by the hippocampus and medial temporal lobe.
The Dark Side of Visual Attention
People are astoundingly blind to unattended information. We employ a rich variety of tasks to examine the costs of inattention and the neural substrates of such attentional deficits.
Perceptual Learning and Neural Plasticity
he ability to recognize an object can be understood as the process of matching a perceptual image to a representation in memory. We employ both fMRI and psychophysics to identify how the brain develops representations for the many objects that we can see and recognize.
Visual Short-term Memory
Many perceptual processes require the observer to retain information briefly in working memory. How is such information organized and represented? In a project headed by Yuhong Jiang, we propose that the organization of visual short-term memory is based on spatial configurations. In a project initiated by Daeyeol Lee, we tested the units of visual working memory. René Marois is leading a neuroimaging project that studies the neural substrates of visual short-term memory capacity.
Our lab members pursue a broad range of topics in visual attention research.