Functional Magnetic Resonance Imaging and Evoked Potential Correlates of Conscious and Unconscious Vision in Parietal Extinction Patients Jon Driver,* ,1 Patrik Vuilleumier,* Martin Eimer,† and Geraint Rees* , ‡ *Institute of Cognitive Neuroscience, University College London; †Department of Psychology, Birkbeck College; and ‡Wellcome Department of Cognitive Neurology, Institute of Neurology, London, United Kingdom Received March 7, 2001 We describe recent functional magnetic resonance imaging (fMRI) and event-related potential (ERP) studies of visual extinction in patients with right pa- rietal damage who can detect isolated visual stimuli on either side, yet often miss contralesional (left) stim- uli during bilateral stimulation. We consider the neu- ral fate of such extinguished visual stimuli and how neural responses differ for consciously detected ver- sus extinguished stimuli. fMRI findings indicate that extinguished stimuli evoke activity in striate and ven- tral extrastriate visual cortex, despite escaping aware- ness. Activations for extinguished stimuli can be found even in category-specific (face-responsive) ar- eas of the fusiform gyrus. On the other hand, activa- tions in visual cortex are stronger for consciously de- tected versus extinguished stimuli, with parietal and frontal areas of the intact left hemisphere also impli- cated in this comparison. Recent ERP data likewise suggest differential neural responses for consciously detected versus extinguished stimuli. We discuss these findings in relation to current speculations about the neural basis of conscious and unconscious perception. © 2001 Academic Press Visual extinction is a relatively common neurological sign after unilateral brain injury, particularly when right posterior-parietal cortex is damaged (Bisiach and Vallar, 1988; Driver et al., 1997; Driver and Vuil- leumier, 2001). It frequently manifests as one compo- nent of the spatial neglect syndrome (Bisiach and Val- lar, 1988), though in some cases it may dissociate from other aspects of neglect (Cocchini et al., 1999). In visual extinction, patients can detect isolated visual stimuli on either side (i.e., visual fields are intact, consistent with sparing of posterior occipital cortex); yet they often miss contralesional (typically, left) stimuli during bilateral stimulation. This has commonly been attrib- uted to a pathological bias in spatial attention follow- ing the lesion (Posner et al., 1984; Desimone and Dun- can, 1995; Driver et al., 1997), that disadvantages contralesional stimuli in situations where multiple stimuli compete to attract attention, as on bilateral but not unilateral trials. Some evidence already exists, from purely behav- ioral experiments, to suggest that extinguished visual stimuli may undergo residual unconscious processing. Reaction times to detect or locate stimuli on the ipsile- sional side can be affected by the presence versus ab- sence of concurrent undetected stimuli on the contrale- sional side (Marzi et al. 1996; Vuilleumier and Rafal, 2000). More complex properties of an extinguished stimuli, such as its color, shape, or even its identity and semantics, can also influence performance in an im- plicit manner (for reviews, see Driver, 1996; Driver and Vuilleumier, 2001). Finally, the rate of extinction for bilateral trials can be modulated by the relationship between concurrent contralesional and ipsilesional stimuli. Typically extinction is reduced if these stimuli group visually to form a single Gestalt (Mattingley et al., 1997), but is increased if two ungrouped stimuli are identical in the visual property to be reported (Baylis et al., 1993; Vuilleumier and Rafal, 2000). On the basis of such behavioral findings in parietal extinction patients, Driver (1996; Driver et al., 1997; Driver and Mattingley, 1998) and others (e.g., Heilman et al., 1997; Robertson et al., 1997) have suggested that unconscious residual processing of extinguished stim- uli might arise in posterior visual cortex and/or more anterior areas along the ventral visual stream. These areas are often structurally intact in neglect and ex- tinction patients with parietal damage. However, the function of such areas could in principle be compro- 1 To whom correspondence and reprint requests should be ad- dressed at Institute of Cognitive Neuroscience, University College London, 17 Queen Square, London WC1N 3AR, UK. Fax: UK-(0)207- 813-2835. E-mail: j.driver@ucl.ac.uk. NeuroImage 14, S68 –S75 (2001) doi:10.1006/nimg.2001.0842, available online at http://www.idealibrary.com on S68 1053-8119/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.