Research report Blindsight after hemidecortication: Visual stimuli in blind hemifield influence anti-saccades directed there Olga Savina, Andre Bergeron and Daniel Guitton* Montreal Neurological Institute, McGill University, Canada article info Article history: Received 28 July 2011 Reviewed 16 November 2011 Revised 5 February 2012 Accepted 1 May 2012 Action editor Carmen Cavada Published online 18 May 2012 Keywords: Hemidecortication Blindsight Anti-saccades Retino-tectal pathway Saccadic eye movements Plasticity abstract Patients missing a cortical hemisphere, removed surgically at adulthood, cannot consciously see a visual probe stimulus (P) flashed in their blind contralesional, hemifield. Nevertheless, they have a low-level form of blindsight wherein P can affect the reaction time of a manual response to the appearance of a visual target in their seeing hemifield. This ability is thought to require the pathway from retina-to-ipsilesional superior colliculus (SC) to cortex of the remaining hemisphere (Leh et al., 2006a, 2006b, 2009). Apart from emitting ascending signals, the SC normally sends saccade commands to the brainstem, a function seemingly conserved after hemidecortication because such patients can generate voluntary and accurate saccades bilaterally (Herter and Guitton, 2004). However, they cannot generate goal-directed saccades to P in their blind hemifield. We hypothesized that, in hemidecorticate patients, P might influence anti-saccades directed to the blind hemifield, to the mirror location of a visual cue presented in the seeing hemifield. We used anti-saccades because our patients could scale their anti-saccade amplitudes approxi- mately according to different cue locations, thereby permitting us to control the end point of their anti-saccades to the blind hemifield. We identified in these patients a new form of blindsight wherein unseen P, if properly timed at the anti-saccade goal location in the blind hemifield, reduced the reaction time and improved the accuracy of anti-saccades directed to that general location. We hypothesize that P in the blind hemifield produced low-level signals in the ipsilesional SC that, if appropriately located and timed relative to anti- saccade goal and onset, interacted with anti-saccade motor preparatory activity e produced by descending commands to SC from the remaining hemisphere e so as to modify both anti-saccade reaction time and end point. Our results support normally encoded and functionally useful, but subliminal, signals in the retina-to-ipsilesional SC-to- reticular pathway of hemidecorticate patients. ª 2012 Elsevier Ltd. All rights reserved. * Corresponding author. Montre´ al Neurological Institute, 3801 University Street, Room 786, Montre´al, Que´ bec, Canada H3A 2B4. E-mail address: daniel.guitton@mcgill.ca (D. Guitton). Available online at www.sciencedirect.com Journal homepage: www.elsevier.com/locate/cortex cortex 49 (2013) 861 e876 0010-9452/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.cortex.2012.05.001