Fig 1.
Counting accuracy in hypnosis (HYP, red) and control (CON, grey) depicted separately for the total group of participants (all) and the three subgroups of low, middle, and highly suggestible participants.
Counting accuracy (%) was defined as the number of silently counted targets relative to the total number of targets presented. Black circles and lines mark the median and the interquartile range (difference between 75th and 25th percentiles), respectively. Pink and grey areas indicate the probability density function of the corresponding data points.
Fig 2.
(A) Grandaverage waveforms (pooled across all participants, n = 60) and 95% C.I. in hypnosis (HYP, red) and control condition (CON, blue) for target, distractor, and standard stimulus type at electrode Pz. The grey rectangle marks the P3b time window (320–470 ms) used for statistical analysis of data at sensor- and source-level. Sensor-level topographies at the peak latency of the target P3b (410 ms) depending on condition and stimulus type. (B) For each stimulus type a one-sample t-Test was performed and the results were assessed with a t-contrast to test for differences between experimental conditions, i.e., HYP < CON, within the P3b time window as shown in (A). There were no significant amplitude differences for the distractor between HYP and CON. The summary statistic scalp-time images were thresholded at p < .001 (uncorrected) with FWE correction at cluster-level, p < .05, based on random field theory. The statistical parametric maps (SPMs) are displayed as Maximum Intensity Projection (MIP) of the 3D (scalp x time) summary statistic image. Blue dots mark the electrode sites. post = posterior; ant = anterior (C) Effect size, partial eta square (), SPMs display the temporal evolution of the significant t-Contrasts HYP < CON for the target and standard shown in (B). Grey dots mark the electrode sites and the white lines’ crossing the peak effect size of the space x space or space x time map.
Fig 3.
Source-level effects of the three-stimulus-oddball paradigm during the P3b window (320–470 ms) in the control (CON) and hypnosis (HYP) condition.
(A) Effects of interest (F-contrast) showing sources whose activation was significantly different from zero. r = right; l = left. (B) Statistical comparison (t-contrast) of target > standard for source activities within the P3b window in the CON and HYP. The cortical images show sources that were significantly more activated following processing of target as compared to standard stimulus. The labeled brain structures refer to the cluster peaks. PFC = prefrontal cortex; FLC = frontolateral cortex; SPL = superior parietal lobule; LOC = lateral occipital cortex; Hipp = Hippocampus; OFG = occipital fusiform gyrus; CC = posterior cingulate cortex; PCG = precentral gyrus. The summary statistic images of the cortical mesh were thresholded at uncorrected p = .001 with FWE correction at cluster-level, p = .05, based on random field theory. (C) Parameter estimates (β-value, 90% C.I.) of target, distractor and standard in CON and HYP for sources located within cluster peaks of the left and right hemisphere (see Fig 3B and S1-10 Table in S1 File). (D) Grandaverage cluster peak source waveforms and 95% C.I. Since polarity of source waveforms is meaningless, and depends, inter alia, on the orientation of the cortical surface where the sources in question are located, values of source strength are expressed as root mean square (RMS, in arbitrary units). Grey rectangle marks the P3b window.
Fig 4.
Effect of hypnosis at source-level.
(A) Effect size () of statistical comparison (t-contrast) of hypnosis (HYP) vs. control (CON) for the processing of the target stimulus. Left panel: The cortical mesh images show sources that were significantly less activated in HYP compared to CON following processing of target. SPL = superior parietal lobule; OFG = occipital fusiform gyrus. The summary statistic images of the cortical mesh were thresholded at uncorrected p = .001 with FWE correction at cluster-level, p = .05, based on random field theory. r = right, l = left. Right panel: The bar plots summarize the contrast estimates (90% C.I.) for three stimulus types in HYP and CON at the respective peaks of the cluster. (B) The grandaverage cluster source waveforms (across participants, n = 60) and 95% C.I. are based on the average of source waveforms within the respective clusters found in Fig 3A (see S1-11 Table in S1 File) for the three stimulus types in HYP (red) and CON (black). Source strength is expressed as root mean square (RMS, in arbitrary units). Grey rectangle marks the P3b window (320–470 ms). (C) Comparison between modeled (dotted lines) grandaverage data, i.e., all back-projected sources, and measured (solid line) grandaverage data in sensor space at electrode Pz for three stimulus types in HYP (red/pink) and CON (black/grey). Note that back-projected model data at sensor-level (modeled data) match very well the measured data profiles as indicated by grandaverages at Pz and the scalp topographies at 0.41 s.
Fig 5.
Spectrotemporal connectivity within the network of three electrodes (O1, CP1, F5) at the left hemisphere across all participants (n = 60).
(A) Evolution of the pairwise connectivity during the hypnosis (HYP, left) and control (CON, right) condition for the target stimulus. The connectivity values represent the time-variant partial directed coherence (PDC) from 1 to 30 Hz within the latency range of −800 to 700 ms relative to the target stimulus. Note, that directed information flow is mainly located in the low frequency range (8−12 Hz) both in HYP and CON. (B) Absolute difference of the connectivity matrices between HYP and CON. Warmer colours (red) indicate HYP > CON and cooler colours (blue) CON > HYP. The grey rectangle marks the spectrotemporal region of interest (ROI) that was used for statistical comparison of both conditions. (C) Effect size expressed as Cohen’s d for the statistically significant ROI difference in spectrotemporal connectivity between HYP vs. CON. White lines in (B) and (C) encircle the significant spectrotemporal region at p < .05 corrected for multiple comparisons in the time-frequency plane using cluster-based statistics. (D) Lead field magnitude (μV) of three source cluster peaks located in the left hemisphere: the occipital fusiform gyrus (l OFG), the superior parietal lobe (l SPL), and the frontolateral cortex (l FLC). The maximum lead field magnitude of these source cluster peaks defined three electrodes as suitable representatives of these structures that were selected for the connectivity analysis.