Tesi etd-02112015-112531 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
MILICIA, FEDERICA
URN
etd-02112015-112531
Titolo
BRAIN SCANS AS A TOOL TO PREDICT SUCCESS OF FASHION ITEMS
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA BIOMEDICA
Relatori
relatore Prof. Positano, Vincenzo
controrelatore Prof. Scilingo, Enzo Pasquale
tutor Dott. Baldo, Davide
controrelatore Prof. Scilingo, Enzo Pasquale
tutor Dott. Baldo, Davide
Parole chiave
- EEG
- neuromarketing
- sales forecasting
Data inizio appello
06/03/2015
Consultabilità
Completa
Riassunto
For thousands of years human beings have been trading goods and services. Trying to understand what happens in the mind of a customer has been akin to understanding a “black box” via trial and error. Very recent advances in neuroscience indicate a paradigm shift in this age-old challenge. The rising fields of neuroeconomics and neuromarketing shed light on decision processes and preferences of the consumer mind. Based on the analysis of the brain activity, this pilot study utilizes neuroscientific EEG recordings to understand and predict customers’ preference of shoes and relates this neuroscientific measure to sales in the market.
It was demonstrated that the left- and right-anterior brain regions are part of two separate neural systems underlying approach (e.g. joy, happy, interest) and withdrawal (e.g. fear, anger, disgust) emotions, respectively.
Decrease in synchronous activity in the alpha band in adults is associated with activation, thus when populations of cortical neurons become activated there is a decrease in amplitude. Relatively greater left frontal activity, either as a trait or a state, indicates a propensity to approach or engage a stimulus, while relatively greater right frontal activity indicates a propensity to withdraw or disengage from a stimulus.
From this, the analysis of this work was based on a parameter acquired from the EEG signals, the Frontal Alpha Asymmetry (FAA). It was computed to distinguish approach and withdrawal emotions as the difference between the right frontal EEG alpha power and the left frontal EEG alpha power. Being the EEG power inversely related to cortical activation, negative EEG asymmetry scores reflect greater relative right frontal EEG activity and, therefore, negative emotions.
These elicited emotions were directly linked with the success of a shoe.
In this work, it is proposed a novel approach based on brain data to forecast product performance that reaches an accuracy of 72%, higher than self-report data used as classical marketing methods.
It was demonstrated that the left- and right-anterior brain regions are part of two separate neural systems underlying approach (e.g. joy, happy, interest) and withdrawal (e.g. fear, anger, disgust) emotions, respectively.
Decrease in synchronous activity in the alpha band in adults is associated with activation, thus when populations of cortical neurons become activated there is a decrease in amplitude. Relatively greater left frontal activity, either as a trait or a state, indicates a propensity to approach or engage a stimulus, while relatively greater right frontal activity indicates a propensity to withdraw or disengage from a stimulus.
From this, the analysis of this work was based on a parameter acquired from the EEG signals, the Frontal Alpha Asymmetry (FAA). It was computed to distinguish approach and withdrawal emotions as the difference between the right frontal EEG alpha power and the left frontal EEG alpha power. Being the EEG power inversely related to cortical activation, negative EEG asymmetry scores reflect greater relative right frontal EEG activity and, therefore, negative emotions.
These elicited emotions were directly linked with the success of a shoe.
In this work, it is proposed a novel approach based on brain data to forecast product performance that reaches an accuracy of 72%, higher than self-report data used as classical marketing methods.
File
Nome file | Dimensione |
---|---|
BrainSca...licia.pdf | 2.88 Mb |
Contatta l’autore |