May 31, 2021.
How can neurotechnology help in marketing?
Using the portable EEG headband Muse and Naxon Explorer, Regina, a graduate researcher in ORT University in Uruguay, worked with Emotions classification of participants who watched a standard ad aimed at women versus a novel ad under the framework of “femvertising” (advertising that is carried out in favor of women, with messages and images that empower women and girls).
Using statistical data analysis, Regina was able to distinguish different brain responses in the participants according to which type of ad they were looking.
This provided Naxon Labs with valuable insights into developing our product Naxon Emotions.
Find out more at: https://naxonlabs.com/products/emotions
EEG data analysis
In the first instance, the signal-to-noise ratio of the data obtained was reviewed. Clipping of the EEG tracing was not considered necessary due to the little or no presence of artifacts that could affect the analysis. Subject No. 2's tracing was discarded, due to continuous mispositioning artifacts, so it was not possible to include it.
The analysis was based on the absolute Spectral Frequency Power (PSD) of the Alpha (7.5Hz - 13Hz) and Beta (13Hz - 30Hz) brain frequency bands. Being the evaluated electrodes AF7 and AF8 (located at the anterior-frontal level of the head) with reference electrodes in FPZ, taking into account the international EEG 10-10 placement standard and the modified combinatorial nomenclature (MCN) for the electrodes.
Frontal Lobe Frontal Activity Spectogram
As an exploratory analysis, the presence of statistically significant differences in the alpha and beta brain wave frequencies (associated with cognitive activity) was examined for the frontal channels of the portable EEG using the presented stimulus as a categorical variable, which makes up two groups, one from advertising. normal and another of the feminist.
For this, an analysis of variance (simple ANOVA) was defined for the case of alpha frequencies, since the assumptions of the procedure were verified, including the normal distribution of the variables.
Analysis of variance represents a collection of statistical models and their associated estimation procedures that are used to analyze differences between means. In its simplest form, ANOVA provides a statistical test of whether two or more population means are equal.
In the case of the beta frequency, a normal distribution of the data was not found, so the non-parametric analysis of Kruskal Willis was carried out.
Therefore, in our analysis, the absolute wave frequency power, both alpha and beta, was taken as a dependent variable for the frontal channels AF7 and AF8, which is an indicator of how present a certain wave frequency is in an EEG channel. The video presented to the participants was used as an independent variable.
Results:
A significant difference (p <.05) was found between both videos for the alpha frequency, both for the front left channel AF7 [F (1, 121126) = 64.25, p <.0001], and for the front right channel AF8 F (1, 121126) = 1855, p <.0001]. Given the value of F in the second case, we can see that the difference for alpha in this last channel is significantly greater.
In the case of the Kruskal Willis test for the beta frequencies of both frontal channels, significant differences were found both for the left channel AF7 (X2 = 3638, p <0.001, df = 1) and for the channel AF8 (X2 = 2265, p <0.001, df = 1), being in this case the most intense differences for the left frontal canal.
Considering the means of the powers, a greater alpha power is observed in the left channel for the feminist stimulus than for the common one. But the differences are statistically greater for the case of alpha of the right channel, where on the contrary we observe a greater power in the channel for the common video in relation to the feminist one.
In the case of beta, we find a stronger statistical difference for the left channel, where for the feminist stimulus the potency of beta is significantly higher than for the common stimulus. In the case of the right frontal canal, it is also observed, although to a lesser degree, a greater potency for feminist stimulation compared to the common one.
Interpretation of the obtained results:
The relationship of the frontal power asymmetries of frequency bands such as Alpha and Beta with differences in the valence of emotional states has been studied in recent years and allows us to orient the results obtained towards a conclusion that supports the subjective table collected in the experimentation.
The increase in the left frontal Alpha PSD frequency observed before stimulus B has been related in various studies with emotional states of positive valence and / or excitement (Vecchiato et al., 2011; Mikutta, Altorfer, Strik, & Koenig, 2012 ).
On the contrary, the increase in the right frontal Alpha PSD frequency has been related to negative emotional states (Diaz & Bell, 2012).
Regarding the increase in the frontal Beta PSD frequency band before stimulus B, it could be related to an increase in the concentration of the participants (Lim, Yeo, & Yoon, G. (2019), but the evidence is insufficient and little specific to conclude it accurately.
We can conclude then that the predominance of positive emotions during stimulus B over stimulus A has a neurobiological correlation.
References:
Diaz, A., & Bell, M. A. (2012). Frontal EEG asymmetry and fear reactivity in different contexts at 10 months. Developmental Psychobiology, 54(5), 536-545. doi: 10.1002/dev.20612
Lim, S., Yeo, M., & Yoon, G. (2019). Comparison between concentration and immersion based on EEG analysis. Sensors, 19(7), 1669. doi: 10.3390/s19071669
Mikutta, C., Altorfer, A., Strik, W., & Koenig, T. (2012). Emotions, arousal, and frontal alpha rhythm asymmetry during Beethoven’s 5th symphony. Brain topography, 25(4), 423-430. doi: 10.1007/s10548-012-0227-0
Vecchiato, G., Toppi, J., Astolfi, L., Fallani, F. D. V., Cincotti, F., Mattia, D., ... & Babiloni, F. (2011). Spectral EEG frontal asymmetries correlate with the experienced pleasantness of TV commercial advertisements. Medical & biological engineering & computing, 49(5), 579-583. doi: 10.1007/s11517-011-0747-x
Image Link ref: https://en.wikipedia.org/wiki/File:EEG_10-10_system_with_additional_information.svg