External Self-Collected Dataset.zip

<b>Self-Collected EEG Dataset for Realistic Mental Arithmetic Workload Analysis</b>This dataset contains EEG recordings from <b>40 participants</b> engaged in mental arithmetic tasks, structured to simulate real-world workload conditions. In contrast to conventional lab-controlled EEG datasets, this collection deliberately includes a broad range of <b>physiological and environmental artifacts</b>, making it suitable for testing model robustness under noisy, real-life scenarios.Key Characteristics:<b>Noise Sources Captured</b>: ECG (cardiac activity), EMG (muscle movements), EOG (eye blinks/movements), respiratory artifacts, and power line interference.<b>Hardware Used</b>: EEG signals were acquired using the <b>KT88-3200 system</b> with <b>19 channels</b> at a <b>sampling rate of 200 Hz</b>.<b>Participant Background</b>: All participants had prior experience with dual-stage mental arithmetic protocols.Task and Recording Design:Each session lasted approximately <b>5 minutes</b>, with around <b>1-minute segments</b> extracted from early and late portions of the session for analysis.A <b>10-minute inter-task break</b> was included to introduce workload transitions.Each participant contributed around <b>35 minutes</b> of EEG data in total.<b>Session orders were randomized</b> to prevent bias from sequencing effects.File Naming Convention:Each EEG recording is provided in <b>EDF format</b>, named using the following structure:<code>"SubjectXX_Y.edf"</code><code>XX</code> – Participant number (e.g., <code>Subject00</code>, <code>Subject01</code>, ... <code>Subject39</code>)<code>Y</code> – Task condition:<code>0</code>: <b>No Workload (Resting State)</b><code>1</code>: <b>Workload – Bad Counter</b><code>2</code>: <b>Workload – Good Counter</b>This naming scheme allows easy identification of the subject and the workload level during each recording session.Applications:This dataset supports research in:EEG signal denoising and artifact rejectionWorkload classification and regression modelsEvaluation of brain–computer interface (BCI) algorithms in field conditionsReal-world cognitive workload monitoringBy combining structured task protocols with real-world noise variability, this dataset provides a practical benchmark for developing and validating EEG-based technologies outside controlled laboratory settings.