Neuromarketing using EEG takes advantage of frontal asymmetry as a measure of emotional engagement but confounds abound and better methods are needed for real results.
For decades consumer product companies have relied on “traditional” market research techniques which typically ask people to rate and evaluate a product or piece of marketing material using survey tools: “How much do you like the advert?”; “Which brand design do you prefer?”; “How likely would you be to purchase the product?” These are all simple and straightforward questions used to ascertain how popular or influential the end product or promotional material is likely to be once released on the marketplace.
But with a significant proportion of new products never attaining the level of success that the respective marketing teams at first hoped for, some companies, both small and large, have started looking for alternative ways to measure metrics such as consumer liking, emotional engagement, preference and purchase intent. They have therefore started turning to consumer neuroscience, and, more specifically, the much talked about subfield of neuromarketing.
Neuromarketing around the world
The range of companies offering neuromarketing include long-standing consumer market research companies like Nielsen (who bought both Innerscope and NeuroFocus to set up it’s consumer neuroscience division) and IPSOS, as well as smaller start-up enterprises like Mindlab International, NEUROHM and SalesBrain. Nowadays, neuromarketing companies can be found in many countries around the globe (see here for the directory of neuromarketing companies).
Each company offers their own suite of proprietary methods which can be tailored to the needs of the specific client and product material. In addition, many companies now offer EEG as a core method. But how can you actually use EEG from a neuromarketing perspective? What insights are possible? And how do the waveforms on the computer screen translate to the questions that an Account Director wants answered?
See related post What Does the EEG Signal Measure?
One commonly cited metric which neuromarketing companies focus on is the degree of “emotional engagement”. To try and get an accurate measure of this, they often examine frontal asymmetry in the EEG output. This measure involves computing the difference in activity between left frontal and right frontal regions, generally in the alpha band.
See related post Alpha Oscillations and Attention
Greater left frontal activity is suggested to be a measure of approach motivation, whilst greater right frontal activity is suggested to be related to avoidance. However, this measure can be impacted by a number of factors. For instance it can flip directions if there is engagement and approach motivation but where the emotional valence is negative or more significantly it can be impacted by things contracting your left or right hand and also basal cortisol levels which have in turn shown to be modulated by seasons and a host of other factors (see last section in this review). The number of confounds suggest that frontal asymmetry, in the absence of an understanding of the individuals basal state, may be fraught with challenges of interpretation.
These challenges notwithstanding, neuromarketing folks have forged bravely ahead. For example, a study by Professor Rafal Ohme from the University of Warsaw and the neuromarketing company NEUROHM used the frontal asymmetry index to measure emotional engagement whilst people (n=45) viewed the 3 different Sonia Bravia TV adverts – Bouncing Balls, Paint, and Playdoh. They tracked the difference between left and right frontal activity within the alpha band (8-12Hz) over the time course of the adverts to identify the moments which elicited the greatest asymmetry.
The study revealed that overall the Bouncing Balls advert elicited the greatest left hemispheric dominance compared to the other adverts during “emotional” part – the part with music and moving scenery. part which placed music and accompanying video. The magnitude increase of left hemispheric dominance from the Balls advert was ~55% compared to a decrease of ~45% for the Playdoh advert and an increase of ~14% for the Paints advert.
Another study from a team researchers based in Italy, Poland as well as representative form the neuromarketing company Brainsigns used EEG and autonomic markers to explore the emotional impact of a Prada and a Cartier perfume advert (n=28). The computed various metrics including a memorization Index (theta activity), an approach-withdrawal index (alpha frontal asymmetry) and an emotional Index (from the physiological markers of galvanic skin response and heart rate). Like the above study, they also tracked these indices through the different scenes of the adverts. Using the approach-withdrawal index as a measure of interest they found that different sections of the advert appealed to men and women at different points. For example the magnitude of interest during the opening scenes of the Prada advert was ~ 63% greater in men compared to women.
See related post What does EEG have to do with Perfume?
Challenges of Interpretation
The studies may use EEG metrics that have a long history in the literature. However, there are several challenges in their interpretation in these experimental contexts, and a simplistic association of these measures to emotional engagement is fraught with minefields. The lack of imposed standards of methodological and scientific rigor by companies means that some of the results arising from neuromarketing studies are of poor quality and should be met with skepticism. Poor analysis, misinterpretation, overinterpretation and retroactive fitting of the data to the original question of interest are far too common.
Take the first study on Sony TV adverts. The interpretation that the Balls ad created the most emotional engagement is based on simply a difference in frontal asymmetry between two aggregate groups without insight into the nature of individual differences. Is the aggregate difference because a few outliers showed a large difference while the rest showed none? ‘Significant’ p-values often don’t call these out, particularly when tests meant for normal distributions are applied to skewed ones. The Z-scores of the second study on perfume is a better measure but still masks individual differences. Furthermore, there is no attention to the natural fluctuation of frontal asymmetry that arises from other factors outside of the advertisement.
The studies also do little to link back the EEG results to real results in advertising. How do these EEG differences result in an individual’s perceived emotional engagement? And most important from the advertisers perspective, how do they translate into a decision to buy?
The future of Neuromarketing
Despite the numerous challenges, there are undoubtedly research and commercial opportunities in neuromarketing and consumer neuroscience. With a larger repertoire of EEG measures, better designed studies to understand individual trajectories and control for confounds, it can indeed lead to better understanding of consumer decision making and become a valuable tool for understanding brands, aiding product development and overlapping with the co-emerging field of neuroeconomics.