Imaging Conscious and Subliminal Word Processing.

Abstract

The purpose of this chapter is to revisit the organization of the visual word recognition system, focusing on a single issue that my colleagues and I have addressed in several experiments: what patterns of activity characterize nonconscious and conscious words? Using pattern masking, words can be presented on the retina for several tens of milliseconds, and yet fail to be consciously seen by the participant. This raises two obvious questions. First, to what extent are such subliminal words processed in the brain? Second, what cerebral events distinguish subliminal words from words that are consciously reportable? Focusing on those well-defined questions narrows down the problem of consciousness to a more tractable level. This chapter is divided in three parts. First, I review the empirical data on the neuroimaging of masked words. I then describe a putative integrative theory based on the hypothesis of a "conscious neuronal workspace" (Dehaene & Changeux, 2000; Dehaene, Kerszberg, & Changeux, 1998; Dehaene & Naccache, 2001; Dehaene, Sergent, & Changeux, 2003). This model emphasizes the role of distributed neurons with long-distance connections, particularly dense in prefrontal, cingulate, and parietal regions. These neurons are capable of interconnecting multiple specialized processors and can broadcast signals at the brain scale in a spontaneous and sudden manner, thus creating a global availability that, according to our hypothesis, is experienced as consciousness (see also Baars, 1989). In the third part, I examine recent experiments that tested some predictions of the workspace model. (PsycINFO Database Record (c) 2008 APA, all rights reserved)