Abbreviation: ECG: Electrocardiogram; E-Field: Electric Field; HRV: Heart Rate Variability; MRI: Magnetic resonance Imaging; tDCS: transcranial direct current stimulation.
Heart rate variability
Cardiac activity was measured continuously and then separated in four time-points: 5-min of baseline (baseline), initial 5 minutes of the tDCS session (5-min tDCS), 10-min during the 0-back task (during 0-back) and the last 5-min of the tDCS session (last 5-min tDCS). Cardiac activity was acquired using a Biopac ECG100C amplifier (Biopac Systems Inc., USA) and Biopac Acqknowledge software 4.3. For the amplifier, the gain was set at 5000, with a high pass filter of 0.05 Hz and a low pass filter of 35 Hz. The sampling rate was set at 1000 Hz. The electrocardiogram was set-up using the lead I configuration (Francis, 2016), with two Ag/Agcl electrodes attached below the right and left clavicle and a third reference electrode placed under the left ribs. The electrocardiogram data were analyzed with PhysioData Toolbox (version 0.6.3) which allows for automated R-peak detection and inter-beat-interval (IBI) extraction. Misidentified R-peaks were manually corrected after visual inspection of the data.
HRV (in milliseconds) was assessed by calculating the Root Mean Squared of Successive Differences (RMSSD) of the detrended IBI data (Tarvainen et al., 2002 )and High-Frequency-HRV (HF-HRV, 0.15 to 0.40 Hz), using the absolute power of the HF band, as calculated using the Lomb-Scargle method. Importantly, although there are several ways to investigate HRV, only RMSSD and HF-HRV were analyzed here because they present robust effects when using NIBS intervention over the PFC and they are markers of the parasympathetic system (Makovac et al., 2017).
RMSSD and HF-HRV were calculated for time epochs of 5 min, as recommended elsewhere (Malik et al., 1996).Therefore, for the time-point in which we recorded the HRV during the 0-back task (10min), the average of two epochs of 5 min was obtained.
tDCS
TDCS was performed using a NeuroConn device (DC-Stimulator Plus, NeuroConn, Germany). The session consisted of a 20-minute direct current, delivered using two rubber electrodes of 25cm² (5x5cm) applied to the scalp with a conductive gel. The anode and cathode were placed over the left and right neuronavigated DLPFC, respectively ( left: x = -38, y = +44, y = +26; right: x = 38, y = +44, y = +26) (Blumberger et al., 2018), pointing towards Cz. Stimulation at these coordinates has been found to achieve optimal anticorrelation with the ACC in functional brain studies (Fox et al., 2012)(Razza et al., 2022). The active tDCS sessions were performed using currents of 1.5mA and 3mA. The sham protocol was identical but consisted of a brief active period of 30 seconds fade-in and 30 seconds fade-out at the beginning and end of the session to a current intensity of 3mA with the device set to 0mA for the remainder. The ‘study mode’ of the tDCS device was used to deliver active or sham current based on a randomized imputed code, allowing for double-blinding of participants and study personnel. The codes were randomized viahttps://www.randomizer.organd were managed by an independent party not directly involved in data acquisition.
Self-reported measures
The State-Trait-Anxiety Inventory (STAI) state inventory was administered at the beginning of each session (Spielberger et al., 1970). The STAI state measures how participants are feeling at that moment using a 4-point Likert scale (1 = not at all, 2 = a little, 3 = moderately, 4 = very much). This measurement was collected to identify the baseline psychological state of each participant before the tDCS session. Moreover, the Visual Analogue Scale (VAS) was collected two times (at baseline and after the tDCS session). In VAS participants are asked to indicate from ‘zero’ to ‘one-hundred’ (“I do not experience this at all”, “I experience this very much”) how much they were feeling the following moods: ‘angry’, ‘tense’, ‘sad’ ‘happy’, ‘stressed’ and ‘anxious’. Higher scores indicate higher levels of perceived stress or negative affect.