Imaging of neural oscillations with embedded inferential and group prevalence statistics
Fig 4
Effects in a subgroup of participants: Mapping induced oscillations in the beta band (13-30 Hz) during visual stimulation.
The data are that of Fig 3, and the present figure layout is identical. a) Subspace computation, example subject: in this participant, only one significant spatial dimension was retained for the signal subspace contributing to stronger power in the beta band. b) Subspace computation, group level: γ = .22 was the highest population prevalence that could be rejected at a p = 0.05, thus the majority null hypothesis could not be rejected. The analysis was pursued with the subgroup (n = 8) of participants that showed the hypothesized effect. The purpose was to appreciate the spatial concordance across subjects and compare iES to standard source imaging approaches. c) Spatially filtered signals, example subject: induced power changes in the band of interest (beta, but also in alpha band) are clearly visible in 3 example trials. d) Spatially filtered signals, group level: induced power changes in the band of interest were found in the participant subgroup (n = 8). e) Subspace correlation maps, example subject: the hypothesized effect localized to the right post-central/parietal cortex. f) Subspace correlation maps, subgroup level: the effect localized to the right post-central gyrus. Note that this effect cannot be generalized to the majority of the population that the subjects were drawn from (see b) but only to a subset, which may present interesting capacity for identifying subtypes in participants.