Continuous EEG Recordings During Coherent and Scene-Scrambled Film Viewing

Description

This dataset contains continuous EEG recordings collected during naturalistic film viewing for a study of transition-related potentials (TRPs) and narrative comprehension. Thirty-four participants watched two short narrative films, The Art of Love and City Lights. Each film had a coherent original version and an incoherent scene-scrambled version. The experiment used a counterbalanced crossover design: each participant watched one coherent film and one incoherent film. EEG was recorded with a BioSemi ActiveTwo system at 512 Hz using a 32-channel scalp montage, with additional electrodes recorded during acquisition. The dataset includes continuous EEG recordings, condition labels, participant/session identifiers, and frame-accurate annotations of cinematic cuts. These cut annotations allow extraction of cut-locked EEG responses and comparison of TRPs across coherent and incoherent narrative contexts. The dataset supports analyses of naturalistic EEG, event-locked responses during continuous stimulation, narrative-context effects, and EEG-based detection of cinematic transitions. It was used to evaluate whether cut-locked TRPs exhibit ERP-like temporal structure, whether they are modulated by narrative coherence, and whether cut-related EEG signatures can be recovered from group-averaged continuous EEG using deep neural networks. The data are intended for research on naturalistic cognition, EEG event detection, film comprehension, narrative processing, and transition-related potentials. Folder structure: EEG_Cut_dataset/ ├── recordings/ │ ├── art_movies/ │ │ ├── art_l/ │ │ │ └── BioSemi .bdf EEG recordings for The Art of Love, linear/coherent condition │ │ └── art_nl/ │ │ └── BioSemi .bdf EEG recordings for The Art of Love, non-linear/incoherent condition │ └── city_movies/ │ ├── city_l/ │ │ └── BioSemi .bdf EEG recordings for City Lights, linear/coherent condition │ └── city_nl/ │ └── BioSemi .bdf EEG recordings for City Lights, non-linear/incoherent condition └── cuts/ ├── all/ │ └── CSV files containing all cinematic cut annotations for each film-condition combination └── matched/ └── CSV files containing matched cut annotations for each film-condition combination EEG files are named using the participant identifier and condition label, for example "01_city_l.bdf" or "16_art_l.bdf".

Institutions

• Eötvös Loránd University

  Budapest, Budapest

Categories

Licence