, 2007; Crespi et al., 2011). Three mechanisms may contribute to non-retinotopic processing. One involves an explicit spatiotopic map implemented by neurons whose receptive fields are head-centered or object-centered (referred to as the spatiotopic map). Such neurons have been found in the parietal cortex (Galletti et al., 1993; Duhamel et al., 1997; Chafee et al., 2007; Crowe et al., 2008), and are arguably implicated in other visual areas (d’Avossa et al.,

2007; McKyton & Zohary, 2007; Crespi et al., 2011). Another mechanism of non-retinotopic processing involves updating of stimulus representation on a retinotopic map associated with saccadic eye movements (referred to as peri-saccadic updating). Such updating phenomena have been observed in the parietal (Duhamel et al., 1992; Merriam et al., 2003), frontal (Sommer & Wurtz, 2006) and visual (Nakamura Oligomycin A mw & Colby, 2002; Merriam et al., 2007) cortical areas. The third possible mechanism entails predictive remapping of spatial attention in retinotopic coordinates to keep track of spatiotopic locations of attended objects around saccades (referred

to as attentional BKM120 molecular weight remapping) (Cavanagh et al., 2010; Rolfs et al., 2011). Our recent study has shown that perceptual learning in motion direction discrimination between two stimuli is specific to the relative positions of the two stimuli in a spatiotopic reference frame (Zhang & Li, 2010), suggesting that spatiotopic processing mechanisms in the

dorsal visual pathway can be altered in favor of the trained spatial relation of the stimuli. However, Interleukin-3 receptor it is unclear whether perceptual learning of stimulus attributes processed by the ventral pathway has similar spatiotopic specificity; it remains unknown which of the non-retinotopic mechanisms described above could account for the spatiotopic learning effect. Investigation of these issues can provide insights into spatiotopic processing and perceptual learning. A total of 51 naive subjects with normal or corrected-to-normal eyesight were recruited from undergraduate and graduate students. Each of the subjects was required to sign an informed consent form before the experiments. The current study conformed with the Declaration of Helsinki, and was approved by the ethics committee of Beijing Normal University. The stimuli were generated with a matlab-based psychophysics toolbox (Brainard, 1997; Pelli, 1997) on a computer monitor (Iiyama Vision Master Pro 514, 100-Hz frame rate at 1600 × 1200 pixels) at a viewing distance of 60 cm. A head-and-chin rest was used to stabilize the subject’s head, and a desktop-mounted tracking system (EyeLink 1000; SR Research Ltd., Mississauga, Ontario, Canada) was used to monitor the subject’s eye positions in each trial, ensuring the faithfulness of their gaze direction.