When subjects directed covert search to the right VF with the search array located 5° left, with eye-gaze at 10° left, the left IPS exhibited a strong BOLD response (Fig. 2H).

However, there was only a weak response when the search was directed to the left VF, with the search array being located at 5° right, and the eyes oriented 10° right relative to the head (Fig. 2G). Hence, the left IPS is much stronger activated for covert search to the right, contralateral VF, independent of the eye-gaze orientation, and the array location in screen coordinates. To quantitatively assess the effect of the FOR on the BOLD response, we calculated OSI-744 cell line the percentage signal change for the ROIs in the IPS in both hemispheres and for the ROI centred on the right FEF (Fig. 4A and B). These ROIs were defined by comparing eye-centred contralateral to ipsilateral conditions (see ‘Materials and methods’). As mentioned above, the comparison of non-eye-centred contralateral to ipsilateral conditions did not yield any significantly activated voxels. These ROIs were located in the posterior and anterior part of the left IPS, the posterior right IPS and the right FEF buy Ribociclib (Fig. 4A). These ROIs were included

in the fronto-parietal regions that were shown to be activated based on the contrast [sR(fC), sL(fC), sL(fR), sR(fL)] > [‘all control conditions’]. All four ROIs (left pIPS, left aIPS, right pIPS and right FEF) showed a significant main effect of search condition across all sessions, (Table 2). Further, to test our hypothesis that in these four regions the two eye-centred contralateral conditions

elicited a significantly higher activation, Rutecarpine independent of eye gaze or array location with respect to the head or body, we applied in each ROI four t-tests comparing each of the two eye-centred contralateral conditions with the two eye-centred ipsilateral conditions. Thus, we compared sL(fC) > sR(fC), sL(fC) > sR(fL), sL(fR) > sR(fC) and sL(fR) > sR(fL) with paired two-tailed t-tests. The left pIPS and aIPS and right FEF always revealed higher activation when the covert search was directed to the contralateral side in eye-centred FOR (Table 2), confirmed by t-tests corrected for multiple comparison with the Bonferroni–Holm method. The right pIPS showed a significant main effect for the four conditions, P = 0.0082 in the one-way anova, and t-tests revealed two conditions where search directed to the contralateral VF elicited a higher response than ipsilateral (Table 2). Thus, it is the array location or search direction in eye-centred FOR that determines the strength of the BOLD signal in the search-related fronto-parietal and visual cortex. Overall, the quantitative analysis summarized in Fig. 4B exhibited the presence of a spatially selective map of the current focus of visuospatial attention in the IPS and right FEF. Objects within these regions are represented in an eye-centred manner.