Television and Internet Commercials Avoidance.

[摘要] This dissertation deals with the problem of video advertising avoidance by TV and Internet viewers. It focuses on the counterintuitive but systematic detrimental impact that brands (logos, trademarks pack shots) as well as emotional evoking scenes (joyous and surprising) may have on commercials avoidance (zapping). Theoretical and empirical support is provided for each phenomenon as well as rigorously tested methods to counteract the negative aspects of branding and emotional scenes without loosing their clear benefits. In Chapter 1, I develop a conceptual framework for understanding the impact that branding activity (the audio-visual representation of brands) and consumers’ dispersion of attention have on their moment-to-moment avoidance decisions during television advertising. I formalize this in a Dynamic Probit Model and estimate it with MCMC methods. New, simple metrics of attention dispersion are shown to strongly predict avoidance. I optimize the branding activity under marketing control for ads in the sample to reduce commercial avoidance. This reveals that pulsing the brand presence--while keeping total brand exposure constant--decreases commercial avoidance significantly. In Chapter 2, I study the concomitant effect of two positive emotions, joy and surprise, moment-to-moment on both the viewer’s visual attention and on their avoidance decisions (zapping) of Internet commercials. To do so, I propose a novel non-obtrusive means to automatically capture and classify emotions via images from their facial expressions while, at the same time, tracking their eye-movements. Using a simultaneous version ofthe model in Chapter 1, I find that Joy reduces zapping momentarily with little persistent effect and that this is largely a direct effect, with minimal influence on attention distraction. I also find that Surprise works mainly indirectly by momentarily reducing zapping but at the expense of causing major attentional distraction. The implication for advertisers is that Joyous ads should improve over time, ending strong. As for using Surprise, the risk in showing the unexpected is to loose the viewer’s engagement with the story line.

[发布机构] University of Michigan

[效力级别] Eye-tracking

[关键词] Advertising;Eye-tracking;Facial Expressions;Internet Ads;Dynamic Bayesian Models;Economics;Business;Science;Social Sciences;Business Administration