The Power of mind: Blocking visual perception by hypnosis
by Dr. Barbara Schmidt
Biological and Clinical Psychology, University of Jena, Jena, Germany
In our recently published study (www.nature.com/articles/s41598-017-05195-2), we show that a blocking suggestion given under hypnosis has significant impact on visual perception and underlying brain responses measured via EEG.
Past research has shown that suggestions during hypnosis can have a deep impact on behavior and perception1-3. In our study, we used the suggestion of a perceptual blockade (Figure 1) to block visual perception. We predicted that the suggestion of a perceptual blockade leads to deficient counting performance of visual stimuli. Further, we hypothesized that the P3b response to visual stimuli should be significantly reduced during the perceptual blockade compared to a control condition without blockade suggestion4. The amount of the P3b reduction should be associated with deficient counting performance and a more realistic experience of the blockade.
Participants’ ability to follow suggestions varies. Therefore, we measured the individual level of suggestibility with a standardized procedure by a group test5 before the main experiment. We included 60 participants with different levels of suggestibility: 20 low suggestible, 20 middle suggestible and 20 high suggestible participants.
Figure 1: Illustration of the suggested perceptual blockade.
We recruited healthy participants at the University of Jena. All participants signed informed consent statements. The study was approved by the ethics committee of the Faculty of Social and Behavioral Sciences of the Friedrich Schiller University of Jena and followed the ethics declaration of Helsinki.
The assessment of suggestibility of participants was organized in groups of about 8 participants using the Harvard Group Scale of Hypnotic Susceptibility5.
Our sample consisted of 60 participants (30 female, 30 male) with 20 low suggestible, 20 middle suggestible and 20 high suggestible participants. Mean age of participants was 23.1 years (range 18-44 years).
Visual Oddball task
The three types of stimuli in the visual Oddball paradigm included triangles as frequent stimuli, appearing in 80% of trials. Circles and squares served as rare stimuli, each appearing in 10% of trials. Each stimulus was displayed for 500 ms at the center of a monitor screen, separated by an inter-stimulus interval of 1000 to 2000 ms. Each Oddball task consisted of 500 stimuli and lasted about 10 minutes.
EEG was recorded from 64 electrodes. The impedance of each electrode was kept below 10 kOhm. The initial sampling rate was 1000 Hz, down-sampled to 250 Hz offline. Initially, the reference electrode was FCz, but we re-referenced the data to linked mastoids (TP9 and TP10) offline. The EEG signal was amplified by BrainAmp amplifiers and recorded with the BrainVision Recorder software (both Brain Products, Gilching, Germany). Data analysis was realized with the EEGlab software6. Artifacts related to eye movements were removed with the ICA procedure of EEGlab7. Epochs were selected between -200 ms and 800 ms for each stimulus and condition. In order to obtain event-related brain potentials for each participant in response to stimuli, artifact-free epochs were averaged for each participant, electrode, and experimental condition. The event-related brain potentials (ERP) were baseline-corrected by the average EEG activity between -200 and 0 ms for each electrode. N1 and P2 amplitudes were assessed at electrode Fz and the P3b amplitude was evaluated at electrode Pz. Time windows for amplitudes were: 80-168 ms for the N1, 168-272 ms for the P2 and 320-472 ms for the P3b. Statistical analyses were computed with R8. For between-group t-tests, we used the Welch unequal variances t-test implemented in R that corrects the degrees of freedom in case of unequal variances. For ANOVA effects, we report generalized eta squared values as effect size. All reported correlations are Spearman correlations.
Each participant completed the hypnosis condition and the control condition in one experimental session. The order of conditions was counterbalanced across participants. Prior to the hypnosis condition, participants were handed a real wooden board providing a model for the imagined blockade. Hypnosis was induced by a series of verbal instructions via an in-ear microphone by a trained hypnotist from outside the experimental chamber.
Then, the suggestion of the perceptual blockade followed. Participants were told that the board would block their perception of the screen. While they imagined the blockade, the visual Oddball paradigm was presented. In the control condition, participants were presented the same visual stimuli but without the suggestion of a wooden board in front of their eyes. During both conditions participant’s EEG was recorded. To make sure that participants’ eyes were open during the presentation of the visual stimuli, we used an eye tracker (SensoMotoric Instruments, Berlin) and observed a video showing participants’ eyes.
After each condition, participants were asked how many squares they counted. Based on their answers, we computed the counting performance as the percentage of correctly counted squares in each condition.
Counting performance was significantly worse in the hypnosis condition compared to the control condition (F(1,57) = 33.6, p < .001, η2 = .24). Counting performance was reduced for all participants, but the performance of high suggestible participants was the most deficient, see Figure 2.
Figure 2: Counting performance in both conditions. Participants were very good at counting in the control condition, but significantly worse in the hypnosis condition in all suggestibility groups, especially in the high suggestible group. Error bars indicate standard errors.
A typical comment after the hypnosis condition, especially from highly suggestible participants, was “I concentrated on the wooden board and saw a board that was similar to the one you showed me in the beginning of the experiment. The board was sometimes more transparent like a cloud, so I could see the stimuli behind, sometimes more solid, so the stimuli were hidden behind it.”
Event-related brain potentials
For N1 and P2 amplitudes, only the main effect of stimulus type reached significance: N1: F(2,114) = 20.5, p < .001, η2 = .06; P2: F(2,114) = 20.7, p < .001, η2 = .05. There were no significant differences between the hypnosis and control conditions.
For P3b amplitudes, the main effect of stimulus was significant: F(2,114) = 211.8, p < .001, η2 = .49. P3b amplitudes were largest for the to-be-counted stimuli, followed by the to-be-ignored rare stimuli and the to-be-ignored frequent stimuli, see Figure 3. This is the typical oddball effect9,10. In the hypnosis condition, the amplitudes of the P3b were significantly reduced compared to the control condition, F(1,57) = 22.4, p < .001, η2 = .06.
Figure 3: P3b amplitudes for the three stimulus types of the Oddball task in the hypnosis and control condition pooled for all participants. The Oddball effect indicated by higher P3b amplitudes for to-be-counted stimuli is significantly reduced in the hypnosis condition as compared to the control condition. Thin lines indicate standard errors, grey areas indicate the P3b time window.
As we were especially interested in the P3b response to the rare to-be-counted stimuli, we conducted separate analyses only including the squares. As expected, P3b amplitudes were maximal over the parietal brain areas where P3b magnitude is commonly observed maximally. Topographical effects of the P3b reduction for the to-be-counted stimulus are displayed in Figure 4.
Figure 4: Topographical maps for the P3b response to the to-be-counted stimuli pooled for all participants, visualizing P3b magnitudes that were significantly reduced in the hypnosis condition compared to the control condition.
We also conducted an ANOVA with the factors condition and group including only the rare to-be-counted stimuli. The P3b amplitudes were significantly reduced in the hypnosis condition compared to the control condition F(1,57) = 44.6, p < .001, η2 = .15. The interaction of condition and group was also significant: F(2,57) = 4.9, p = .01, η2 = .04. The reduction of the P3b magnitude under hypnosis was observed in all suggestibility groups, but highly suggestible participants showed the greatest reduction, see Figure 5.
Importantly, the amount of the P3b reduction under hypnosis was significantly associated with counting performance (r = .56, p < .001). The smaller the P3b amplitudes under hypnosis, the worse was the counting performance under hypnosis.
Figure 5: Brain responses to the to-be-counted stimuli, showing smaller P3b magnitudes during the hypnosis condition as compared to the control condition for all suggestibility groups. The grey areas indicate the P3b time window and thin lines indicate standard errors.
The present study revealed that a suggested perceptual blockade significantly impairs visual perception. Participants had to count visual stimuli and made significantly more counting errors in the hypnosis condition when they were suggested a wooden board in front of their eyes compared to a control condition without such visual blockade suggestion. While participants counted the visual stimuli, we also measured early and late ERP responses in response to the visual stimuli. Earlier responses occurring about 100-200 ms post stimulus onset reflect basic sensory processing11, whereas later responses at about 300-400 ms post stimulus onset indicate mainly higher-order cognitive processing of stimuli9. Early brain responses were not affected by the blockade suggestion or suggestibility of participants. But the amplitude of the later P3b brain response was significantly reduced in the hypnosis condition compared to the control condition. This effect was significant for all participants, but most pronounced for highly suggestible participants.
The presence of equivalent early brain responses to the visual stimuli in both the hypnosis and the control condition indicates that participants saw the stimuli equally well in a physical sense. But later processes that are important for cognitive processes like counting visual target stimuli are affected by the suggested blockade. This is in line with the idea that suggestions under hypnosis cause a dissociation between sensory and higher perceptual processing areas in the brain. In other words, when we tell participants under hypnosis that a wooden board in front of their eyes blocks their vision, they still see the stimuli physically, but further processing of these stimuli is impaired, leading to deficient counting performance. This shows that our mind is very powerful and able to significantly modify the processing of stimuli in response to a few words of suggestion.
 Kosslyn, S. M., Thompson, W. L., Constantini-Ferrando, M. F., Alpert, N. M. & Spiegel, D.
Hypnotic visual illusion alters color processing in the brain.
American Journal of Psychiatry 157, 1279-1284 (2000).
 Raz, A., Fan, J. & Posner, M. I.
Hypnotic suggestion reduces conflict in the human brain.
PNAS 102, 9978-9983 (2005).
 Raz, A., Kirsch, I., Pollard, J. & Nitkin-Kaner, Y.
Suggestion reduces the Stroop effect.
Psychological Science 17, 91-95 (2006).
 Spiegel, D., Cutcomb, S., Ren, C. & Pribram, K.
Hypnotic hallucination alters evoked potentials.
Journal of Abnormal Psychology 94, 249-255 (1985).
 Shor, R. E. & Orne, E. C.
Norms on the Harvard Group Scale of Hypnotic Suggestibility, Form A.
International Journal of Clinical and Experimental Hypnosis 1, 39-47 (1963).
 Delorme, A., & Makeig, S.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis.
Journal of Neuroscience Methods 134, 9–21 (2004).
 Debener, S., Thorne, J., Schneider, T. R., & Viola, F. C.
Using ICA for the analysis of multi-channel EEG data in Simultaneous EEG and fMRI: recording, analysis and application
(ed. Ullsperger, M. & Debener, S.) 121-134 (Oxford University Press, 2010).
 R Development Core Team. R: A language and environment for statistical computing http://www.R-project.org/ (2016).
 Polich, J.
Updating P300: an integrative theory of P3a and P3b.
Clinical Neurophysiology 118, 2128-2148 (2007).
 Sutton, S., Braren, M., Zubin, J. & John, E. R.
Evoked-potential correlates of stimulus uncertainty.
Science 150, 1187-1188 (1965).
 Thorpe, S., Fize, D. & Marlot, C.
Speed of processing in the human visual system.
Nature 381, 520-522 (1996).