EEG Neuromarketing: Measuring Emotion in Ads
EEG records electrical activity at the scalp, and the feeling is an inference added afterwards. Four studies show what the signal can predict about movie trailers and product choices, and where the emotion label needs more evidence.
Answer to the question
What does EEG neuromarketing measure?
EEG neuromarketing records electrical activity at the scalp while people watch advertisements and reads engagement or emotion into the pattern. Some signals have added predictive value beyond surveys, in studies of movie trailers and later product choices, but the step from signal to emotion is an inference, and resting frontal asymmetry has weak support as a personality marker.
Key takeaways
- Maarten Boksem and a co-author (2015) recorded EEG during movie trailers and found that beta and gamma oscillations added information about individual and population-wide preference beyond stated preferences.
- Ariel Telpaz and colleagues (2015) found that a midfrontal EEG signal recorded while participants viewed products tracked the choices they made later between pairs of those products.
- Niclas Kuper and colleagues (2019) pooled 79 samples and about 5,700 participants and concluded that the validity of resting frontal asymmetry as a marker of personality is not supported.
- Adam Hakim and a co-author (2019) reviewed EEG-based preference prediction and asked studies to show that neural measures add predictive value beyond traditional methods.
- A signal is not a feeling: the emotion label is an inference added to the recording, and it needs its own evidence.
A cap of sensors on the scalp picks up faint electrical activity from the brain beneath it, and none of that activity arrives labelled with a feeling. EEG neuromarketing is the practice of recording the signal while people watch advertisements and then reading engagement, liking or emotion into its pattern. Martin Lindstrom's Buyology argued the wider case that measuring the brain can reach reactions people cannot or do not report, and electroencephalography is one of the ways to attempt it.
Read EEG neuromarketing as a signal
Each recording can be cut into features, and a review by Adam Hakim and a co-author lists four that researchers lean on: event-related components, inter-subject correlations, hemispheric asymmetry and spectral band powers.
| Feature | What it records | Where it appears in the studies here |
|---|---|---|
| Event-related component | The brain's time-locked response to a stimulus | A midfrontal response in a study of later product choices |
| Inter-subject correlation | How closely different viewers' signals move together | Named in the review as a promising feature |
| Hemispheric asymmetry | The difference in activity between the left and right sides | Frontal asymmetry, tested as a marker of personality |
| Spectral band power | The strength of activity within a frequency range | Beta and gamma activity in a study of movie trailers |
Every row records a signal. The move from a signal back to a feeling is a separate step, often called reverse inference, in which a pattern seen when people report an emotion is taken as evidence that the emotion is present. That step is where the word emotion enters EEG neuromarketing. The wider vocabulary of the field is set out in Neuromarketing Definition: How Brands Read the Buyer's Brain.
Two studies that predicted preference
Trailers for commercially released films gave Maarten Boksem and a co-author an EEG neuromarketing test with a real-world outcome. They recorded EEG while participants watched the trailers, and they also collected stated preferences, meaning what participants said they preferred. The authors report that beta and gamma oscillations carried unique information about individual preferences and about preference across the general population, beyond what the stated measures held. They add that such measures could in principle serve as a neural marker of commercial success.
The authors describe the result as the first evidence that EEG measures relate to real-world outcomes. They frame the gain as added value: models that included the EEG measures predicted choice better than models with stated preferences alone. The stated measures stayed in the model, and EEG was added to them.
Another team asked whether EEG recorded while nobody was choosing could predict a choice made later. Ariel Telpaz and colleagues recorded EEG as participants viewed consumer products, then asked the same participants to choose between pairs of those products. A midfrontal signal, described as a larger early negative response with weaker theta power, tracked preference, and prediction was more accurate when the EEG difference between two products was larger. The authors present EEG as a relatively low-cost way to obtain such a measure, which bears on its appeal in EEG neuromarketing.
Viewing without choosing is close to how an advertisement is met, though the stimuli were products rather than advertisements and the outcome was a laboratory choice rather than a purchase.
Check the emotion label
Frontal asymmetry, often measured in the alpha band as frontal alpha asymmetry, is the difference in activity between the left and right front of the scalp. It has been widely used as a marker of emotion and motivation, and the resting-state version is regarded as an index of trait approach and withdrawal motivation. Niclas Kuper and colleagues checked that reading by pooling data from 79 independent samples with about 5,700 participants in all.
Less than 0.4 per cent of the variation in extraversion and neuroticism could be explained by resting frontal asymmetry. Effects for anger and defensiveness were small to medium-sized in few studies, and impulsivity showed no significant effect. Adjusting for publication bias reduced the effects further, and the authors concluded that the validity of resting frontal asymmetry as a marker of personality is not supported.
That meta-analysis concerns people at rest and personality traits, so it does not rule on asymmetry recorded while someone watches an advertisement. It does show how an emotion label for a signal gets tested: across many samples, with publication bias in view.
Hakim and his co-author reviewed EEG neuromarketing work on preference prediction and describe a field that has grown quickly but still lacks a unified methodology and conclusive results for industry use. Their guidance is concrete. Studies should show that neural measures add predictive value beyond traditional methods, a property researchers call incremental validity, and should adopt the data science and prediction-modelling practices used in fields such as computer science and engineering. The movie-trailer study was built around that test, comparing models with and without the EEG measures.
For owners deciding whether any neural measure is worth buying, What Is Neuromarketing? A Simple Guide for Business Owners sets out the cheaper alternatives. For a brand weighing an EEG study of its own advertising, the sensible request is a design that compares the EEG measure with survey ratings on the same ads, tests the prediction on viewers and ads the model has not seen, and names in advance which outcome counts as success, so that the signal is judged by what it adds and the state it is said to show is earned by evidence rather than assumed from a waveform.
Common questions
What does EEG neuromarketing measure?
EEG neuromarketing measures electrical activity at the scalp while people watch advertisements or products. Researchers summarise it as event-related components, inter-subject correlations, hemispheric asymmetry and spectral band powers. None of these is a feeling by itself, and an emotion label is an inference added to the signal.
Can EEG predict how an advertisement will perform?
In one study of movie trailers, beta and gamma oscillations added information about preference, including preference across the general population, beyond stated preferences. The authors present this as in-principle evidence for a neural marker of commercial success. A review of the field asks for more studies that show added value over traditional methods.
Is frontal asymmetry a reliable emotion measure?
A meta-analysis of 79 samples found that resting frontal asymmetry explained less than 0.4 per cent of the variation in extraversion and neuroticism and concluded that its validity as a marker of personality is not supported. That result concerns the resting state and personality, so asymmetry recorded during viewing is a separate question.
Does EEG neuromarketing replace surveys?
The movie-trailer study kept stated preferences in its models and found that EEG measures added to them. A review of the field urges studies to show that neural measures add predictive value beyond traditional methods, which treats EEG as an addition to conventional research.
Sources
Lindstrom, M. Buyology. Seed book; vault hub Books/Buyology.
Boksem, M. A. S., & Smidts, A. (2015). “Brain Responses to Movie Trailers Predict Individual Preferences for Movies and Their Population-Wide Commercial Success.” Journal of Marketing Research, 52(4). doi:10.1509/jmr.13.0572 https://doi.org/10.1509/jmr.13.0572
Telpaz, A., Webb, R., & Levy, D. J. (2015). “Using EEG to Predict Consumers' Future Choices.” Journal of Marketing Research, 52(4). doi:10.1509/jmr.13.0564 https://doi.org/10.1509/jmr.13.0564
Kuper, N., Käckenmester, W., & Wacker, J. (2019). “Resting Frontal EEG Asymmetry and Personality Traits: A Meta-Analysis.” European Journal of Personality, 33(2). doi:10.1002/per.2197 https://doi.org/10.1002/per.2197
Hakim, A., & Levy, D. J. (2019). “A gateway to consumers' minds: Achievements, caveats, and prospects of electroencephalography-based prediction in neuromarketing.” WIREs Cognitive Science, 10(2). doi:10.1002/wcs.1485 https://doi.org/10.1002/wcs.1485