The dynamic effects of visual complexity and scene cuts on viewer attention
研究利用计算机视觉和眼动追踪技术，分析视频广告中视觉复杂度和场景切换对观众注意力同步性的动态影响，发现简短简单的场景能提升注意力，为广告优化提供指导。
Abstract This research examines the dynamic effects of video advertising features on viewers’ attentional focus. We leverage state-of-the-art computer vision techniques to extract frame-level measures of visual complexity and scene cuts, and we combine these with eye-tracking data that capture attention via moment-to-moment attentional synchrony across viewers. Our findings indicate that short and visually simple scenes elicit higher levels of attentional synchrony. A time-series analysis of 2,520 individual viewing experiences across 42 public service announcements, reveals that visual complexity exerts a delayed, negative effect on attentional synchrony, whereas scene cuts (i.e., shorter scene durations) boost attention. Together, these results demonstrate that short and simple scenes help sustain attentional focus, which subsequently enhances narrative immersion, ad liking, and recognition. Our new analytical framework illustrates how computer vision, time-series modeling, and neuromarketing measures can be effectively integrated to advance video advertising research, while also offering actionable guidance for marketers to optimize viewer attention.