Using EEG to Predict Consumers’ Future Choices

Abstract

It is well established that neural imaging technology can predict preferences for consumer products. However, the applicability of this method to consumer marketing research remains uncertain, partly because of the expense required. In this article, the authors demonstrate that neural measurements made with a relatively low-cost and widely available measurement method—electroencephalography (EEG)—can predict future choices of consumer products. In the experiment, participants viewed individual consumer products in isolation, without making any actual choices, while their neural activity was measured with EEG. At the end of the experiment, participants were offered choices between pairs of the same products. The authors find that neural activity measured from a midfrontal electrode displays an increase in the N200 component and a weaker theta band power that correlates with a more preferred product. Using recent techniques for relating neural measurements to choice prediction, they demonstrate that these measures predict subsequent choices. Moreover, the accuracy of prediction depends on both the ordinal and cardinal distance of the EEG data; the larger the difference in EEG activity between two products, the better the predictive accuracy.

Consumer neuroscience for marketing researchers

“…EEG's high temporal resolution and ability to measure surface cortical electrical activity of submilliseconds indicates that EEG, rather than fMRI, is more useful for measuring neural responses of dynamic messages in TV commercials (Ohme & Matukin, 2012;Ohme et al, 2009). Currently, EEG is frequently used in advertising research to measure consumer responses to emotional engagement (Ohme & Matukin, 2012), personal relevance (Pynta et al, 2014), cognitive attention (Gountas, Ciorciari, Gountas, & Huddle, 2014;Rothschild et al, 1986;Vecchiato et al, 2010), facial recognition (Gountas et al, 2014), prediction of success (Boksem & Smidts, 2015), and brand choice evaluation to predict future choices (Telpaz, Webb, & Levy, 2015). Table 2 presents a summary of advertising research using EEG research methodology.…”

“…Table 4 summarises some key studies using MEG. Boksem and Smidts (2015) Beta frequencies Individual preference and predictor of movies' future commercial success Telpaz et al (2015) Midfrontal electrode, increase N200, preferred product correlation with weaker theta band power Evaluation of pairs of brands to predict future choices. Accuracy of prediction dependent on larger difference.…”

A gateway to consumers' minds: Achievements, caveats, and prospects of electroencephalography‐based prediction in neuromarketing

“…There is another event‐related component that has been investigated in recent years in connection to valuation. Telpaz, Webb, & Levy () found that the N200 component, a negative wave peaking between 200 and 350 ms after stimulus onset, can differentiate between high‐ and low‐preferred goods. This finding was later replicated by an independent research team (Goto et al, ).…”

Pathways to Consumers’ Minds: Using Machine Learning and Multiple EEG Metrics to Increase Preference Prediction Above and Beyond Traditional Measurements

“…Despite the fact that EEG is commonly used in the neuromarketing industry (see NMSBA Website), and that there is accumulating data linking various EEG signals with value-based choice (Dmochowski, Sajda, Dias, & Parra, 2012;Fuentemilla et al, 2013;Khushaba et al, 2013;San Martin, Appelbaum, Pearson, Huettel, & Woldorff, 2013;Sutton & Davidson, 2000), only several academic studies attempted to predict subjects' stated preferences or actual choices (Kong, Zhao, Hu, Vecchiato, & Babiloni, 2013;Ravaja, Somervuori, & Salminen, 2013;Telpaz, Webb, & Levy, 2015;Vecchiato et al, 2011;Yadava, Kumar, Saini, Roy, & Prosad Dogra, 2017), or population marketing success (Barnett & Cerf, 2017;Boksem & Smidts, 2015;Dmochowski et al, 2014;Guixeres et al, 2017;Venkatraman et al, 2015). However, importantly, nearly all these previous studies did not examine if their prediction accuracy was above and beyond the prediction accuracy of traditional marketing measurements.…”