This provides a functional account of the physiological finding that the response of a V1 neuron to its preferred input within its receptive field is higher when this input pops out from a background than when the same input is just part of a homogeneous texture (Allman et al., 1985, Knierim and Van Essen, 1992 and Marcus and Van Essen, 2002). Lateral connections (Gilbert

and Wiesel, 1983 and Rockland and Lund, 1983) between JAK drugs V1 neurons, leading to mutual suppression between neurons tuned to similar input features, have been suggested as mediating such contextual dependencies of V1 responses. For example, V1 neurons preferring the same or similar orientations are more likely to suppress each other. This iso-orientation suppression reduces V1 neural responses to a homogeneous texture. Meanwhile, V1 neurons preferring, and thus responding to, the pop-out foreground region escape this iso-orientation suppression, more so when the orientation contrast is higher between the

foreground and background bars, making the foreground region more salient according to the V1 saliency hypothesis (Li, 1999 and Li, 2002). This contextual influence on V1 responses is present whether the animal is Selleck Autophagy inhibitor awake (Knierim and van Essen, 1992) or under anesthesia (Nothdurft et al., 1999), regardless of feedback from V2 (Hupé et al., 2001). This bottom-up nature of saliency is in line with the dissociation between attentional attraction and the awareness of the cue in our psychophysical data. Our study succeeded in linking V1 activities directly with saliency

(in terms of cueing effects). In particular, as the orientation contrast between the foreground bars and the background bars increased, V1 neurons responded more vigorously to foreground found bars. This was seen in our data in the form of a larger C1 amplitude, a stronger BOLD signal, and a stronger attentional cueing effect. Until now, only the behavioral predictions of the V1 saliency hypothesis had been tested. These tests have provided various confirmations of the theory including (1), the attentional attraction of an eye of origin singleton (Zhaoping, 2008), whose unique feature is not represented in any visual cortical area other than V1; (2), the close relationship between the reaction times for finding visual search targets and the properties of feature selectivities of the neurons in V1 (and not in extra-striate cortices) (Koene and Zhaoping, 2007); and (3), the alignment between the reaction times in visual search/segmentation tasks and the saliency predicted by the V1 saliency hypothesis (Zhaoping and May, 2007), but not by traditional saliency models (reviewed by Itti and Koch [2001]).