Emotion Processing and Regulation Task (Static Stimuli) — tDCS‑EEG Dataset

This repository provides EEG recordings and behavioral data from the Emotion Processing and Regulation task conducted with transcranial direct current stimulation (tDCS).

• Preregistration: https://osf.io/qdp3w
• Preprint: https://osf.io/qtm8r

---

Overview

Each participant took part in two experimental sessions:

• ses-1 — Sham stimulation
• ses-2 — Active stimulation

The order of sham/active conditions was counterbalanced across participants.

---

Participants

• N = 120 right‑handed, neurologically healthy adults with normal or corrected‑to‑normal vision.
• Missing data: Participant sub-005 completed only ses-1 due to a recording error during ses-2.

---

Experimental Task

Participants completed 120 trials per session, evenly allocated to a 2 (content: social, non-social) × 3 (regulation requirement: watch-neutral, watch-negative, reappraise-negative) factorial design. On each trial, they viewed a static image for 5 s and either watched or reappraised it as instructed. After each image, participants rated its arousal and then valence on separate 9-point scales.

---

tDCS Stimulation

System: Starstim 8 (Neuroelectrics, Spain) with NIC2 software.

Electrode Montage

Stimulation was targeted to the dorsolateral prefrontal cortex (dlPFC) in two alternative montages:

• Right dlPFC stimulation

  • Anode: F4
  • Returns: FP2, FZ, FC2, FC6

• Left dlPFC stimulation

  • Anode: F3
  • Returns: FP1, FZ, FC1, FC5

Electrode areas

• Anodal: 8 cm²
• Return: π cm²
• Ground: left earlobe

Stimulation Protocol

• Active: 2 mA for 20 min (with 30 s ramp‑up)
• Sham: only ramp‑up periods at start and end; no sustained current
• Questionnaires: After each session, participants completed the tDCS Sensation Questionnaire (Polish version: https://osf.io/ufszr) to evaluate potential side effects. Additionally, after the final session, they indicated whether they believed each session involved real, sham, or I don’t know stimulation to assess blinding effectiveness.

---

EEG Acquisition

• Cap: 64‑channel QuickCap (32 EEG electrodes used)
• Amplifier: Neuroscan SynampsRT
• Sampling rate: 1000 Hz
• Impedance: kept < 10 kΩ

Active EEG electrodes (32): FP1, FP2, F7, F3, FZ, F4, F8, FT7, FC3, FCZ, FC4, FT8, T7, C3, CZ, C4, T8, M1, TP7, CP3, CPZ, CP4, TP8, M2, P7, P3, PZ, P4, P8, O1, OZ, O2

Additional sensors

• EOG: Horizontal (HEO) and Vertical (VEO) channels were available on the cap but were not connected during recording.
• Physio: ECG and GSR/EDA were recorded via auxiliary channels.

---

EEG Preprocessing

All preprocessing was performed in MATLAB R2020b using EEGLAB 2023.0 and ERPLAB 9.10. The full, commented pipeline is provided in code/Preprocessing_EEG.m.

Steps

1. Band‑pass filter: 0.1–30 Hz (zero‑phase Hamming‑windowed FIR)
2. Downsample to 250 Hz
3. Re‑reference to average mastoids (M1, M2)
4. Bad‑channel detection using clean_rawdata (autocorrelation criterion = 0.8)
5. ICA with runica
6. Automatic IC rejection using ADJUST and SASICA
7. Spherical interpolation of removed channels
8. Epoching: −200 to 5000 ms relative to stimulus onset
9. Baseline correction: −200 ms pre‑stimulus
10. Artifact rejection: Step 1 – absolute amplitude on channels 1–30, epochs rejected if amplitude exceeded ±200 µV within −200 to 5000 ms. Step 2 – FASTER epoch_properties on channels 1–30, epochs rejected if any metric exceeded |z| > 2.
11. Condition‑wise averaging using ERPLAB

---

Derivatives & Ancillary Data

derivatives/processed_erps/

Averaged ERP files (.erp) for each participant and session after preprocessing.

derivatives/side_effects_blinding_effectiveness/

• side_effects_blinding_effectiveness_english.csv — _blinding effectiveness and side effects questionnaire
• side_effects_blinding_effectiveness_data_dictionary.csv — data dictionary with variable names and value coding

code/

• MATLAB preprocessing script and documentation: Preprocessing_EEG.m

---