Neuroaesthetics, as defined by Zeki in 1999, is the scientific approach to the study of aesthetic perceptions of art, music, or any other experience that can give rise to aesthetic judgments. One way to understand the processes of neuroaesthetics is studying the electroencephalogram (EEG) signals that are recorded from subjects while they are exposed to some expression of art, and study how the differences among such signals correlate to the differences in their subjective judgments; typically, such studies are conducted on limited data with a purely statistical signal analysis. In this paper we consider a larger data set which was previously used in an experiment on beauty perception; we apply a novel machine learning-based data analysis methodology that allows us to extract symbolic like/dislike rules on the voltage at the most relevant frequencies from the most relevant electrodes. Our approach is not only novel in this particular area, but it is also reproducible and allows us to treat large quantities of data.
Statistical and Symbolic Neuroaesthetics Rules Extraction From EEG Signals
Maddalena Coccagna;Federico Manzella;Sante Mazzacane;Giovanni Pagliarini;Guido Sciavicco
2022
Abstract
Neuroaesthetics, as defined by Zeki in 1999, is the scientific approach to the study of aesthetic perceptions of art, music, or any other experience that can give rise to aesthetic judgments. One way to understand the processes of neuroaesthetics is studying the electroencephalogram (EEG) signals that are recorded from subjects while they are exposed to some expression of art, and study how the differences among such signals correlate to the differences in their subjective judgments; typically, such studies are conducted on limited data with a purely statistical signal analysis. In this paper we consider a larger data set which was previously used in an experiment on beauty perception; we apply a novel machine learning-based data analysis methodology that allows us to extract symbolic like/dislike rules on the voltage at the most relevant frequencies from the most relevant electrodes. Our approach is not only novel in this particular area, but it is also reproducible and allows us to treat large quantities of data.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
