Visual Affective Stimulus Database: A Validated Set of Short Videos

Area of Science:

• Psychology and Cognitive Science
• Neuroscience
• Media Studies

Background:

• Standardized emotional stimuli are crucial for reproducible psychological research.
• Existing stimulus sets may lack diversity or dynamic elements essential for capturing complex emotional responses.
• The need for a validated database of dynamic, multi-scene emotional stimuli is evident.

Purpose of the Study:

• To create and validate a comprehensive database of short emotion-inducing videos.
• To assess the reliability and scoring distribution of emotional dimensions (valence, arousal, dominance) within the video stimuli.
• To establish a versatile tool for research in emotion and related fields.

Main Methods:

• Standardization of 299 videos (people, animals, objects, scenes) using Adobe Premiere Pro CC 2018 for consistent duration (3s), resolution (1080 pixels/inch), and size (1920 × 1080 mm²).
• Participant assessment (N=116) using a 9-point scale for valence, arousal, and dominance.
• Classification of video valence (positive, neutral, negative) based on >60% participant consensus.

Main Results:

• A database of 242 validated short videos was established: 112 positive, 47 neutral, and 83 negative.
• Emotional dimensions exhibited distributions consistent with fundamental emotional characteristics.
• High internal consistency reliability was achieved for valence (0.968), arousal (0.984), and dominance (0.970), with inter-dimensional reliability ranging from 0.799 to 0.968.

Conclusions:

• The developed emotion short-video system provides reliable, multi-scene dynamic stimuli.
• The database demonstrates good scoring distribution across emotional dimensions.
• This system is applicable and valuable for emotion research and related scientific investigations.