Skip to main content
. 2019 Jun 5;6(6):181904. doi: 10.1098/rsos.181904

Figure 3.

Figure 3.

Grand-averaged EEG spectra in SNR over the occipito-temporal ROI (averaged across left and right occipito-temporal channels, P9 & 10, PO11 & 12 respectively). A SNR of 3 corresponds to 200% of amplitude increase relative to EEG noise level. Responses at 0.86 Hz and subsequent harmonics (e.g. 1.71 Hz, 2.57 Hz, 3.42 Hz and so forth) shown separately for (a) upright and (b) inverted face stimuli. Three-dimensional scalp maps are shown for significant harmonics, indicating the largest responses over left and right occipito-temporal channels. On grand-averaged data, there were no significant individual harmonics for inverted faces in this paradigm. Note that the relatively small SNR at 0.86 Hz for upright faces relative to higher harmonics does not imply that amplitude at that frequency is low, but to a larger EEG noise in this low frequency range (see also e.g. [60]). The colour scale is showing the scale from 1 to the maximum SNR of each harmonic. Note that the scale of y-axis is cut to 5 in order to have a better visualization of the SNR of each significant harmonic of the individual face recognition responses. The response SNR at the presentation rate of 6 Hz is much higher than 5 (figure 2).