From Sensation to Meaning: Comparing Sensory and Animate Visual Features in Anti-Vaping PSAs

Overview

Anti-vaping public service announcements (PSAs) play a vital role in conveying the health risks of vaping and motivating healthier behavior, particularly in light of the post-2020 rise in adult vaping rates. While prior research has emphasized the emotional power of sensory-based imagery, our study explores the potential influence of animate features, such as human or pet presence. Using computer vision analysis and large language model (LLM)–based annotation, we examine internet-sourced PSAs and their simulated emotional responses. We compare two regression models: one based on sensory-driven features, and the other on animate content. Results indicate that social cues, especially the presence of children or pets, were stronger predictors of moral emotions such as empathy. In contrast, the sensory-based model more effectively predicted fear and disgust but was less successful in explaining moral or prosocial emotions. These findings suggest that incorporating animacy in anti-vaping PSAs can enhance emotionally resonant responses known to promote persuasion, offering practical insights for designing more impactful health communication campaigns.

Competition history

• CSEF 2026 Behavioral & Social Sciences (Senior Division) · Entry S-03-24

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Behavioural and Social Sciences projects
• CSEF Behavioural and Social Sciences projects
• Every CSEF 2026 project