Supplementary Material
Supplementary Methods & Results
Additional details on EEG recording and processing. EEG data from 30 electrodes (Fp1, Fpz, Fp2, F7, F3, Fz, F4, F8, FC5, FC1, FC2, FC6, T7, C3, Cz, C4, T8, CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, O1, Oz, O2) on standardized scalp sites covering the entire scalp were collected by means of a 136-channel EEG system with active signal shielding technology (Refa_Ext-system, TMS International B.V., www.tmsi.com). Additionally, electrodes on mastoids (TP9 and TP10) were recorded. For the measurement of ocular artifacts, vertical and horizontal eye movements were recorded both under the right eye and at the outer canthi of both eyes. These data were recorded by IMAGO (pfitec®, Endingen, Germany) with a sampling rate of 512 Hz and an online band-pass filter of 0.1–80 Hz. All channels were amplified against the average of all connected inputs. Impedances were kept below 5 kW.
BrainVision Analyzer 2.0 was used for offline data analysis. EEG data was first re-sampled to 256 Hz and then re-referenced to an average reference. After manual artifact inspection, an offline 0.5–49 Hz bandpass filter (time constant 0.3183 s, 24 dB/octave) and a 50 Hz notch filter were applied. To avoid potential noise caused by the study set-up at the beginning and end of a measurement, we used only data between 10 and 310 seconds in each condition and person (total of 5 Min/condition/person) for subsequent analysis. Following segmentation into 2-second epochs (10% overlap allowed), ocular movement artifacts were corrected using the Gratton & Coles algorithm and artifactual segments (gradient 50 < µV; maximum and minimum amplitude between -100 µV and 100 µV) were rejected and excluded from further analysis (i.e., < 5 %). For all remaining artifact-free 2-second EEG epochs (on average, 157.6 ± 11.0 epochs per subject), Fast Fourier Transformation (FFT) with 0.5 Hz resolution (Hanning Window: length 10 %) was applied. Due to potential functional differences between lower and upper alpha activity (Kubicki, Herrmann, Fichte, & Freund, 1979), analyses were conducted for both alpha-1 (7.5–10 Hz) and alpha-2 (10–12.5 Hz). After averaging the power spectra across all epochs, we exported the mean alpha-1 and alpha-2- powers (µV2) for each electrode and participant. For statistical analyses, we calculated global (average of all 30 electrodes) and region-specific measures (frontal = Fp1, Fpz, Fp2, F7, F3, Fz, F4, F8, FC5, FC1, FC2, FC6; temporal = T7, T8; central-parietal = C3, Cz, C4, CP5, CP1, CP2, CP6, P7, P3, Pz, P4, P8, POz, occipital = O1, Oz, O2) on the basis of log-transformed alpha-1 and alpha-2 powers of the electrodes.
Additional results on frequency bands other than alpha. We employed several ANOVAs for repeated measures to explore exercise-induced changes in frequency bands other than alpha (see Table S3). Moreover, for the frequency bands that significantly changed due to exercise, we calculated hierarchical regression models to examine whether habitual or induced expectations also predict changes in other frequency bands (see Table S4 for the prediction of frequency spectra at POST30; for POST2 and POST15, neither habitual, all ΔR2 = .014, ps ≥ .204, nor induced expectations, all ΔR2 ≤ .027, ps ≥ .091, predicted any of these frequency bands).
1
Supplementary Material
Table S1
Detailed content of multimedia film clips for manipulating expectations
Condition / Enhanced Expectation / Expectation / Control 1No Expectation / Control 2
No-Effect Expectation
Effects of exercise / Current research shows that exercise has considerable positive effects on mind & body / No information about effects of exercise
Information about scientific study of brain activity during exercise, the EEG method, and the EEG application in this study / Current research casts doubt on the overarching benefits of exercise
Duration and intensity of the exercise in this study are particularly suited to maximizing health benefits / Only intensive and sustained exercise results in health benefits
Moderate, short exercises as in this study barely result in benefits
Exercise will augment the secretion of happy hormones, which results in a considerable mood increase / Exercise will barely augment the secretion of happy hormones and therefore not affect mood
Exercise will result in a relaxation of the central nervous system and a related effect on inner balance and calmness / Such an exercise is not suited to relaxing the central nervous system
Exercise will reduce stress hormones, which results in improved tolerance of daily stressors / Only longer and more intense exercises will reduce stress hormones and result in more inner balance and calmness
Although not perceivable, exercise will reduce blood pressure and decrease augmented blood sugar and fat / Bodily processes will rarely be affected by such an exercise
Effects of compression shirt / Due to its compression properties, the shirt supports the cardio-vascular system, thereby easing breathing during the exercise / The shirt supports removal of heat and sweat from the body for accurate EEG measurements
Thus, the shirt considerably intensifies the already described health effects of exercise
Table S2
Bivariate correlations, means, and standard deviations for all variables included in the regression analyses (N = 76)
Variable / 1 / 2 / 3 / 4 / 5 / 6 / 7 / 8 / 9 / 10 / 11 / 12 / 13 / 14 / 15 / 16 / 17 / 18 / 191. / Induced Expectation C1 / —
2. / Induced Expectation C2 / .78* / —
3. / Induced Expectation C3 / –.05 / –.05 / —
4. / Habitual Expectation / –.12 / –.14 / –.10 / —
5. / PACES – Enjoyment / .12 / .15 / –.01 / .39* / —
6. / STAI – Anxiety Pre / –.15 / –.23* / .13 / –.18 / –.18 / —
7. / STAI – Anxiety Post / –.03 / –.16 / .19 / –.29* / –.33* / .69* / —
8. / POMS – Vigor Pre / .11 / .11 / –.06 / .27* / .34* / –.45* / –.36* / —
9. / POMS – Vigor Post / .03 / .04 / –.12 / .31* / .61* / –.31* / –.46* / .58* / —
10. / POMS – Depr/Anx Pre / –.22 / –.27* / .09 / .08 / –.10 / .63* / .35* / –.17 / –.03 / —
11. / POMS – Depr/Anx Post / –.15 / –.18 / .13 / –.03 / –.12 / .55* / .39* / –.06 / –.11 / .68* / —
12. / POMS – Fatigue Pre / –.41* / –.39* / –.14 / –.04 / –.26* / .37* / .13 / –.47* / –.18 / .47* / .32* / —
13. / POMS – Fatigue Post / –.30* / –.32* / .01 / –.17 / –.49* / .46* / .37* / –.26* / –.40* / .42* / .54* / .52* / —
14. / POMS – Hostility Pre / –.15 / –.25* / .07 / –.01 / –.11 / .54* / .39* / –.09 / –.02 / .68* / .66* / .37* / .46* / —
15. / POMS – Hostility Post / .11 / .13 / .15 / –.15 / .02 / .40* / .39* / –.05 / .07 / .48* / .59* / .13 / .32* / .60* / —
16. / Alpha-2 Power Pre / –.08 / .09 / .00 / .18 / .05 / –.03 / –.22 / .02 / .19 / .08 / –.03 / .12 / .07 / .04 / -.02 / —
17. / Alpha-2 Power Post2 / –.07 / .09 / .02 / .16 / .06 / .01 / –.19 / .02 / .16 / .15 / .07 / .16 / .12 / .10 / .07 / .95* / —
18. / Alpha-2 Power Post15 / –.06 / .11 / .03 / .17 / .05 / .01 / –.22 / .02 / .18 / .14 / .03 / .13 / .10 / .07 / .06 / .97* / .98* / —
19. / Alpha-2 Power Post30 / –.05 / .14 / .03 / .23* / .11 / –.03 / –.25* / .08 / .22 / .10 / .00 / .09 / .07 / .04 / .02 / .97* / .96* / .98* / —
Mean / 35.00 / 3.34 / 36.41 / 33.49 / 1.87 / 2.00 / 0.35 / 0.18 / 1.07 / 0.98 / 0.29 / 0.13 / 0.45 / 0.57 / 0.55 / 0.53
Standard deviation / 6.79 / 0.65 / 7.44 / 6.57 / 0.64 / 0.87 / 0.53 / 0.39 / 0.81 / 0.75 / 0.53 / 0.37 / 0.43 / 0.42 / 0.43 / 0.43
Note. Bivariate correlations represent Spearman’s rank correlation due to non-normality of some of the data. Induced Expectation C1: Contrast Enhanced Expectation / Expectation vs. Control 1 (No Expectation); Induced Expectation C2: Contrast Enhanced Expectation / Expectation vs. Control 2 (No-Effect Expectation); Induced Expectation C3: Contrast Enhanced Expectation vs. Expectation. Alpha-2 power measures represent the average of all electrodes (global alpha-2 power).
* p < .05.
3
Supplementary Material
Table S3
Delta, theta, and beta frequencies: Descriptive statistics and ANOVA effects of measurement time (N = 76)
After exercise / ANOVA resultsMeasure / Before exercise / POST2 / POST15 / POST30 / Test statistic / p /
Delta / 0.82 ± 0.16 / 0.86 ± 0.18 / 0.84 ± 0.17 / 0.84 ± 0.16 / F(2.661, 199.601) = 2.131 / .105 / .03
Theta / 0.40 ± 0.21 / 0.43 ± 0.21 / 0.44 ± 0.22 / 0.44 ± 0.22 / F(3, 225) = 8.30 / < .001 / .10
Beta-1 / 0.11 ± 0.25 / 0.17 ± 0.25 / 0.18 ± 0.25 / 0.17 ± 0.25 / F(3, 225) = 27.18 / < .001 / .27
Beta-2 / 0.17 ± 0.22 / 0.23 ± 0.21 / 0.24 ± 0.21 / 0.23 ± 0.20 / F(2.709, 203.262) = 16.691 / < .001 / .18
Note. Data are presented as mean ± standard deviation. POST2 (as an example): measurement 2 min after exercise.
1 Greenhouse-Geisser correction due to violation of sphericity.
Table S4
Summary of hierarchical regression analyses of habitual and induced exercise expectations predicting neuronal exercise effects (power spectra other than alpha; N = 76)
Variable / Theta / Beta-1 / Beta-2β / 95 % CI1 / β / 95 % CI / β / 95 % CI
Step 1
Baseline / .930** / [.853, 1.000] / .944** / [.846, 1.000] / .880** / [.767, .988]
Step 2
Baseline / .923** / [.846, 1.000] / .940** / [.842, .992] / .881** / [.768, .995]
Habitual Expectation / .063† / [-.004, .145] / .046 / [-.024, .117] / -.012 / [-.103, .108]
Step 3
Baseline / .928** / [.842, 1.000] / .939** / [.838, 1.000] / .884** / [.772, .999]
Habitual Expectation / .057 / [-.013, .135] / .053 / [-.017, .139] / -.014 / [-.111, .110]
Induced Expectation C1 / .003 / [-.112, .121] / .020 / [-.066, .109] / .004 / [-.103, .125]
Induced Expectation C2 / .035 / [.030, .158] / .048 / [-.026, .121] / .027 / [-.112, .145]
Induced Expectation C3 / -.051 / [-.124, .019] / .003 / [-.076, .093] / .050 / [-.127, .071]
Step 1. R2 / .866*** / .892*** / .774***
Step 2. R2 / .870*** / .894*** / .775***
Step 2. ∆ R2 / .004 / .002 / .000
Step 2. ∆ Cohen’s f2 / 0.03 / 0.02 / 0.00
Step 3. R2 / .874*** / .897*** / .776***
Step 3. ∆ R2 / .004 / .003 / .001
Step 3. ∆ Cohen’s f2 / 0.03 / 0.03 / 0.00
Note. Baseline represents the variable that corresponds to the outcome variable (e.g., Theta). Induced Expectation C1: Contrast Enhanced Expectation / Expectation vs. Control 1 (No Expectation); Induced Expectation C2: Contrast Enhanced Expectation / Expectation vs. Control 2 (No-Effect Expectation); Induced Expectation C3: Contrast Enhanced Expectation vs. Expectation. ∆ Cohen’s f2 represents the individual contribution of the additional predictor/set of predictors. According to Cohen (1988), effect sizes f2 of 0.02, 0.15, and 0.35 are considered small, medium, and large, respectively.
1 Bias corrected and accelerated bootstrap 95% CIs (1000 samples) are reported in square brackets.
† p < .10. * p < .05. ** p < .01.*** p < .001.
5