Figure 2: (A) The spectral decomposition, (B) the
averaged 12 Hz and 15 Hz Fourier component topographies, and(C) the cortical sources for the 12 Hz and 15 Hz Fourier
components significantly different from noise are shown (red clusters)
for the attend to peripheral rings pre-cue condition. (D, E, F)The same information is shown for the attend to fixation cross pre-cue
condition.
In addition to the descriptive reporting of the spatial distribution of
significantly oscillating sources at the ssVEF driving frequencies of
both pre-cue conditions, ssVEF responses to both pre-cue baselines were
directly compared using cluster-based permutations statistics. A
significant positive cluster indicated increased ssVEF power partly in
right occipito-parietal, hMT+ and V2 areas when participants attended
the peripheral rings instead of the central fixation cross (summed
t-values = 82.23, p = 0.022, maximum parametric t(19) = 3.39, p = 0.003,
Cohen’s d = 0.76, see Figure 3A).
However, the absence of a left hemisphere effect does not automatically
imply an interaction between the pre-cue condition (attend rings vs.
attend cross) and hemisphere (left vs. right). Therefore, we extracted
the ssVEF power values for the two pre-cue conditions in the homologous
left hemisphere regions in order to test such an interaction by
comparing the mean relative power across the corresponding cortical
sources for each pre-cue condition in the left and right hemisphere. The
observed interaction (F(1, 19) = 7.78, p = 0.012, η2 =
0.29) corroborated the increased implication of the right
occipito-parietal, hMT+ and V2 areas when both peripheral rings were
attended (see Figure 3B).