, 2000). There are to date few studies of the role GDC-0199 purchase of V4 in figural completion behind occluders. However, one recent study compared responses of V4 neurons to real and “accidental”

contours (contours produced by the occluder which do not provide information about the true shape of the object) ( Bushnell et al., 2011a). This study found that responses to accidental contours were suppressed relative to real object contours, a suppression that disappeared with introduction of small gaps between the occluder and occluded objects. This suggests that V4 is an important stage in image segmentation. Cue Invariant Shapes ( Figure 6D). As objects typically can be defined by multiple features (e.g., color, motion, depth, contour), another important step in figure-ground segregation involves border-surface associations across multiple cues. As shown in Figure 6D, a square shape can be defined by luminance contrast, color, depth, or motion contrast

cues. Whether such invariance at mid-level processing stages is established by integration across multiple feature-specific input maps from V2 or via intra-V4 circuitry is unknown. Although the number of studies examining invariance in V4 is still limited, recent reports do support cue invariant shape coding in V4. Mysore et al. (2008) have described invariant V4 responses to shapes defined by either static or moving cues. In a study Sunitinib mouse by Handa et al. (2010), monkeys were trained on a cue dependent shape discrimination task (dependent on either a motion cue or luminance cue). About a third of the neurons in V4 responded selectively to a shape under ADP ribosylation factor both the motion and the luminance

cue conditions. Further studies are needed to support V4′s role in cue invariant shape recognition. Context Dependency ( Figure 6E). Central to the task of figure-ground segregation is the ability to modify what is perceived as figure and ground depending on situational cues such as stimulus context and attention. Indeed, there are numerous demonstrations of the ability of the visual system to modify the interpretation of what is figure and what is ground (e.g., the classic vase/face example where the figure is perceived as either a vase or as a pair of face profiles). That neuronal response in V4 is highly adaptable and modifiable will become particularly evident in the following section on attentional modulation. In particular, the role of top-down and bottom-up attentional influences on V4 activity has been a topic of intense investigation in the last two decades. However, only recently has the relationship between object representation and attention come into sharper focus. In the sections above, we have summarized studies on V4′s role in processing object features. There is also a vast literature on attentional effects in V4 (for reviews, Desimone and Duncan, 1995 and Chelazzi et al., 2011). Our purpose here is to try and draw ties between these two disparate bodies of literature.