Figure 10.

Effects of selective attention to one of the surfaces of the plaids. Monkey 3 was trained to direct attention to the first grating (cue) appearing on the screen. After 1000 ms of cue onset, the second component appeared in front of or behind the first component. The monkey was required to respond with a lever release only when the cued grating changed luminance, ignoring changes on the noncued surface. (A) Sliding window autocorrelation of the MUA for the same recording site studied in Figure 1A. After the addition of component 2, but with attention directed to component 1, oscillatory activity was still disrupted, similar as to when no attention was payed to the stimulus (Fig. 1). (B) LFP spectra for the gratings (green curves) and for the plaids. For the latter case, orange traces represent attention directed to the surface in the foreground, whereas the red traces indicate attention directed to the surface in the background. Continuous and dotted vertical lines indicate the frequency induced by the grating and plaid stimuli, respectively, when attention was directed to the fixation point. Single-trial traces are shown to the right. (C) No significant differences in firing rate or spike-field coherence were found depending on whether the monkey was paying attention to the fixation point, to component 1 or to component 2, as shown in (C) and (D). The * symbol in 1* + 2 or 1 + 2* indicates to which component attention was directed to. No symbol (1 + 2) indicates that attention was directed to the fixation point. However, the oscillation frequency for the population of sites systematically shifted depending on to which surface the monkey directed its attention to. Thin traces in (B), left panel enclose the 95% confidence interval of the mean.