PARTICIPANTS
Thirty one participants with normal or corrected-to-normal vision participated in this study. Nineteen were medical students (8 women and 11 men, mean age of 23.1) and twelve were surgeons (5 women and 7 men, mean age of 49.5), all right-handed. All participants were recruited by invitation and gave written informed consent before the recording session. The procedures were conducted in accordance with Protocol #46-2020, approved by the Ethics Committee (Comité Ético Científico) of the Universidad de Talca.
PROCEDURE
To evaluate the amplitude of the alpha oscillations in experts and novices surgeons, we performed EEG measurements while participants performed suture in both relaxed and stressed conditions, as well as in the (eyes-open) baseline condition. Novices were fourth-year medical students with 1 to 2 hours of standardized training in a suture workshop in the School of Medicine, University of Talca. Experts were established physicians with a minimum of 3 years and a maximum of 20 years, with regular suture procedures during their practice as surgeons.
In a quiet room, participants were seated in front of a small table containing the surgery pad, tools and suture, plus a lamp pointing at the table (Figure 1 A). All participants were instructed on how to perform the sutures in a simulation model 3/0 of 75 cm (Braun a video the surgical technique) in a wound closure pad (Jig Mk 3 skin pad, Limbs and things, GA, USA), standard surgical instruments and 75 cm nylon 3/0 sutures (Braun Hessen, Germany).
Scalp EEG recordings from participants were obtained in 2 conditions: baseline and suture. In the baseline condition, participants were resting for 3 min with their eyes open, and another 3 min with their eyes closed. The order of these trials was randomly defined and balanced across participants. In the suture condition, participants performed 6 suture trials of 5 minutes each (Figure 1 B). These trials were divided into 3 trials at their own pace, hereinafter referred as relaxed, and 3 trials in which they were subjected to stressful and distracting stimuli, from now on referred as stressed. The inter-trial interval was 2 minutes for all conditions. In the relaxed condition, participants were instructed to perform sutures, whereas in the stressed condition they were instructed to perform as many sutures as possible. Additionally, in the stressed condition only, participants received feedback about the remaining time, and information about performance of other participants, every minute. The sequence of the relaxed and stressed trials was randomized and balanced across participants.
EEG RECORDINGS
The electroencephalogram (EEG) was continuously recorded while participants completed all conditions, using a 37-channel BioSemi ActiveTwo system (BioSemi B.V., Amsterdam, Netherlands): 32 scalp sites (Fp1, Fp2, F7, F3, Fz, F4, F8, T7, C3, Cz, C4, T8, P9, P7, P5, P3, P1, Pz, P2, P4, P6, P8, P10, PO7, PO3, POz, PO4, PO8, O1, Oz, O2, and Iz) according to the modified 10–20 System (American Electroencephalographic Society, 1994), left and right mastoids, and 3 electro-oculogram (EOG) channels (at outer canthi of each eye, and below the right eye). All signals were recorded in single-ended mode. The EEG and EOG were low-pass filtered with a 5th-order sync filter (half-power cutoff at 208 Hz) and digitized at 1024 Hz.
DATA ANALYSIS
The data from this study is available in http://dx.doi.org/10.17632/xb8fzmrf8j.1.
Data analyses were conducted using a combination of EEGLAB [44] and ERPLAB [45], running on MATLAB 2019b (MathWorks, Natick, MA, USA). EEG signals were bandpass filtered offline using a non-causal Butterworth infinite impulse response filter, with half-power cutoffs at 0.1 and 40 Hz, and a roll-off of 12 dB/octave, and then down-sampled to 256 Hz. Eye-movement artifacts and eye blinks were corrected using independent component analysis (ICA). Subsequently, scalp channels were referenced offline to the average of the left and right mastoids, and the three EOG signals, plus Fp2, were used to create two new bipolar vertical and horizontal EOG derivations in order to explore remaining ocular artifacts.
For each EEG recording, the first 4 seconds of data were removed from all conditions to minimize the presence of artifacts. After that, data segments of 2.5 and 4.5 minutes were extracted from baseline and suture trials, respectively. Thus, the EEG trials for baseline were 2.5 minutes of data and the EEG trials for suture conditions were merged resulting in 13.5 minutes of data. EEG data were subjected to a Fast Fourier Transform through a 4-sec, 50% overlap, Hanning-taper, artifact-free moving window. Power spectra with a number of averaged windows of less than 180 were eliminated from further analysis. The grand average power spectra (µV2) were computed for each EEG channel for all recordings. Thus, based on the scalp distribution of the alpha power (8-12 Hz) in the occipital areas (see Figure 3), a ROI was defined by averaging the alpha power at three occipital electrodes (O1, O2 and Oz). Finally, mean values of the power in the alpha band, from each participant’s ROI, were compared across conditions. In addition, we obtained the number of sutures (stitches) from each participant, for each condition.
STATISTICAL ANALYSIS
Statistical differences were estimated by Bayesian analysis using the Bayes Factor Toolbox for Matlab (https://github.com/klabhub/bayesFactor in Matlab written by Bart Krekelberg) based on Rouder et al. 2012 [46]. Statistical differences in power values were evaluated by a two-factor analysis of variance (ANOVA), with between-subjects factor of expertise (two levels: expert and novice) and within-subject factor of condition (three levels: baseline, relaxed and stressed). Statistical differences between the numbers of sutures were evaluated using a two factor ANOVA. Main effects for the ANOVA were estimated as the ratio of the Bayes factors for the full model and the restricted model, obtained by excluding the factor for the main effect. Differences between means were assessed as the Bayes factor (BF) for paired or unpaired t-tests. Statistical significance for t-tests was set to a probability from data > 0.90 (BF > 10), and for correlations was set to probability for the alternative hypothesis (pH1) > 0.90 (BF >10). Again, a BF for the alternative hypothesis between 2 and 10 was considered as moderate non-conclusive evidence.
To evaluate the association between alpha power and the efficiency of suture movements, we computed the Bayes factor for the Pearson product-moment correlation coefficient. A strong correlation was defined when r2 values were equal or greater than 0.5, and a moderate correlation when r2 values were between 0.45 and 0.5. Unless otherwise specified, all values are reported as mean + SD in the main text and as mean + SEM in the figures.