Releasing crowding prior to a saccade requires more than “attention”: response to van Koningsbruggen and Buonocore


We thank van Koningsbruggen and Buonocore (2013) for their interest in our recent study, in which we investigated the relationship between eye movements and visual crowding, the phenomenon whereby a target object in peripheral vision is made difficult to recognize when closely flanked by distractor objects (Pelli and Tillman, 2008). We found that, just prior to a saccadic eye movement, the deleterious effects of crowding are reliably diminished at the saccade goal (Harrison et al., 2013a). As plausible explanations for such a pre-saccadic release from crowding, we discussed changes in the gain of visual neurons (e.g. Moore and Armstrong, 2003), a reduction in probabilistic positional uncertainty (Greenwood et al., 2009; van den Berg et al., 2012), and receptive-field shifts (Tolias et al., 2001) brought about by the known neurophysiological links between oculomotor and visual areas (for reviews, see Schall, 2002; Moore et al., 2003). Van Koningsbruggen and Buonocore have proposed two interesting alternative explanations for our results, and we respond to these here.