Analysis of Consumer Decision on E-Commerce Platforms with Video Advertising and Brain Data

Abstract

In recent years, video advertising has become one of the most powerful tools for influencing consumer decisions, often using visuals, music, and celebrities to attract buyers. However, this creates a problem: even low-quality products are sometimes promoted through high-quality visual media, leading consumers to purchase them, while good-quality products may be ignored when presented in poor-quality videos. The amount of research in this area is quite limited,, specifically addressing this mismatch between video presentation and product quality. This study investigates consumer purchasing interest under two conditions: good-quality video and poor-quality video. Particular focus is given to the mismatched cases, where attractive advertisements may lead to the purchase of poor-quality products, and poor advertisements may reduce interest in good-quality products. EEG signals were recorded from participants while they viewed advertisements, and neurophysiological features such as Mean Absolute Value (MAV) and Shannon Entropy were extracted a total of 35 participants took part in the experiment,viewing two advertising videos (High- and low-quality). Extracted EEG features (Mean Absolute Value and Shannon Entropy) from beta and gamma bands achieved near-perfect classification performance (AUC ≈ 1.0) in discriminating buy vs not-buy decisions from beta (β) and gamma (γ) frequency bands. The results reveal how attractive advertisements can sometimes mislead consumers, while EEG analysis helps identify when purchasing decisions align with or deviate from actual product quality.