Page 180 - Demo

Chapter 61783.4%; zygomaticus: 363 trials, 4.4%). After exporting the EMG signals for each trial, which were downsampled to 10 Hz, a baseline correction was performed. Here, the mean signal of the first 500ms of stimulus presentation, in which the expression was still neutral, served as baseline and was subtracted from each datapoint of the following 1.5s belonging to the same trial, in which the emotional expression developed. The data during the last second of the stimulus presentation in which the full emotional expression was shown (ten datapoints) was then averaged for each participant and each muscle region across the two presentations of each of the 40 distinct stimuli.Skin ConductanceTwo electrodes were attached to the index finger and the ring finger of the participants%u2019 non-dominant hands to measure changes in skin conductance (SC). The signal was transmitted using the same Biopac MP160 system as for the fEMG recordings with an EDA100C module (settings: 2000 Hz sampling rate, Gain: 10%u00b5V, 1 Hz low-pass filter) and all physiological recordings were stored in the same AcqKnowledge datafile. The SC data was also preprocessed in the PhysioData Toolbox v.0.6.3 (Sjak-Shie, 2019), by applying the same visual inspection procedure as for the fEMG data. Due to poor data quality, six participants had to be entirely excluded, as well as a total of 313 trials (3.9%) of the remaining participants. Similar to the fEMG data, the downsampled SC data (10Hz) was then baseline-corrected for each trial. Based on previous research (Tsunoda et al., 2008), we decided to consider the maximum positive deflection in the baseline-corrected signal up until four seconds after emotion onset (i.e., a 5.5s time window) as peak response to the stimulus, and eventually log-transformed the distribution after adding 1 as a constant to each data point. QuestionnairesDescriptions of the completed questionnaires which were not outcome measures in our analyses are summarized in the Supplemental Material, including descriptive statistics by group and group comparisons (see Table S3). In the following, more detailed information about the questionnaires of interest for the current study is provided. Interoception measures. Two self-report questionnaires were used in order to measure subjective perceptions of interoceptive accuracy and attention: