Neuromarketing Techniques to Enhance Consumer Preference Prediction

Abstract

This study evaluates the time-tested method of consumer self-reported measures against advanced neuromarketing algorithms to evaluate experience products. To do so, the authors utilize data from the public DEAP database, which contains both self-reports and EEG measurements of the same subjects. With self-reported measures of valence, arousal, and dominance, the authors then evaluate consumer liking, comparing effectiveness of three different methods: (1) the FFT-analysis of EEG, to (2) self-reported ratings, and (3) a combined method of EEG analysis with self-reported ratings. Results suggest that neuromarketing methods when combined with self-reported measures, will substantially increase accuracy, precision, recall, and F1 scores. Moreover, with the exception of utilizing self-reported valence, dominance and arousal combined, the FFT-analysis of EEG was a more powerful predictor of liking than self-reported measurements. Implications for digital marketing, management and business ethics are discussed.