LL-DS-009: Comparative EEG Dataset of Software Engineering Students Performing Programming Activities With and Without Artificial Intelligence Support

Neurophysiological activity was continuously recorded using the Muse 2 EEG headband, which captures electroencephalographic signals from four channels (TP9, AF7, AF8, and TP10) at a sampling frequency of 256 Hz. The recorded signals support the analysis of cognitive states related to attention, concentration, mental workload, engagement, cognitive effort, and neural activation during software engineering tasks. The dataset includes demographic information together with synchronized EEG recordings obtained throughout the instructional activities, enabling temporal analyses of students' cognitive responses across both learning conditions.

The organization of the dataset facilitates comparisons between AI-assisted and non-AI-assisted learning environments while preserving the temporal structure of each experimental session. The primary purpose of this dataset is to investigate how Artificial Intelligence support may influence neurocognitive processes during software development activities in higher education. The dataset provides a valuable resource for research in educational neuroscience, learning analytics, human-AI collaboration, cognitive computing, and software engineering education, contributing empirical evidence regarding the cognitive implications of integrating generative AI tools into engineering education.