, 2004 and Buia and Tiesinga, 2006). Anderson et al. (2011a) found attention affected firing rate differently for bursty versus nonbursty pyramidal cells. How the effects on gamma synchrony relate to neuronal firing enhancement during attention is not clear. Synchrony has also been proposed to underlie binding of object features, thereby enabling perceptual unity

(e.g., Singer and Gray, 1995). Neuronal oscillations of cells in different cortical columns in cat visual cortex may or may not synchronize depending on stimulus geometry (such as spatial separation and feature orientation) (Gray et al., 1989). Enhanced neural synchrony has also been demonstrated when contours are perceived to be part of the same surface but not when Sirolimus interpreted as belonging to different surfaces (Castelo-Branco et al., 2000). Thus, synchrony is a potential way to temporally bind different stimulus features in a cell assembly and provide coherent global percepts. Although our current understanding of the role of synchrony is still evolving (indeed synchrony has been implicated in many mental processes), perhaps it can be viewed as a mechanism for establishing relations (Singer, 1999), whether it be relations within a shape, within an attentional focus, or within a memory trace (e.g., Harris et al., 2003).

One hint comes from the association of gamma band oscillation with hemodynamic signals. Hemodynamic signals are thought to be more closely related to local field potentials (LFPs) than to action potentials (Logothetis et al., 2001). In fact, Niessing et al. (2005) reported that optically AZD5363 chemical structure imaged hemodynamic response Idoxuridine strength correlated better with the power of high-frequency LFPs than with spiking activity. Optical imaging of attentional signals in V4 in monkeys has shown enhancement of the hemodynamic response during spatial attention tasks (Tanigawa

and A.W.R., unpublished data). This is consistent with reported enhancements in gamma band synchrony (Fries et al., 2001) and predicts that spatial attention acts by elevating response magnitude in all functional domains within the attended locale (Figure 8A). This study also showed that feature-based attention (e.g., attention to color) may be mediated, not via enhancement of imaged domain response, but rather via enhanced correlations between task-relevant functional domains (e.g., color domains) in V4. Thus, feature attention may be mediated via correlation change across the visual field, but only within domains encoding the attended feature (Figure 8B). These differential effects of spatial and feature attention suggest that domain-based networks are dynamically configured in V4. We briefly give some consideration to how attentionally mediated reconfiguration of networks in V4 might be directed by top-down influences. V4 receives feedback influences from temporal (DeYoe et al., 1994 and Felleman et al.