: Emotions processing is a complex mechanism that involves different physiological systems. In particular, the Central Nervous System (CNS) is considered to play a key role in this mechanism and one of the main modalities to study the CNS activity is the Electroencephalographic signal (EEG). To elicit emotions, different kinds of stimuli can be used e.g.: audio, visual or a combination of the two. Literature studies focus more on the correct classification of the different types of emotions or which kind of stimulation gives the best performance in terms of classification accuracy. However, it is still unclear how the different stimuli elicit the emotions and which are the results in terms of brain activity. In this paper, we analysed and compared EEG signals given by eliciting emotions using audio and visual stimuli or a combination of the latter two. Data were collected during experiments conducted in our laboratories using IAPS and IADS dataset. Our study confirmed literature physiological studies about emotions highlighting higher brain activity in the frontal and central regions and in the δ and θ bands for each kind of stimulus. However, audio stimulation was found to have higher responses when compared to the other two modalities of stimulation in almost all the comparisons performed. Higher values of the δ/β ratios, an index related to negative emotions, have been achieved when using only sounds as stimuli. Moreover, the same type of stimuli, resulted in higher δ-β coupling, suggesting a better attention control. We concluded that stimulating subjects without letting them know (seeing) what is actually happening may give a higher perception of emotions, even if this mechanism remains highly subjective. Clinical Relevance- This paper suggests that audio stimuli may give higher perception of the elicited emotion resulting in higher brain activity in the physiological areas and more focused subjects. Thus using only audio in emotion related studies may give more reliable and consistent results.
Comparison of different emotion stimulation modalities: an EEG signal analysis
Farabbi, Andrea;Polo, Edoardo M;Barbieri, Riccardo;Mainardi, Luca
2022-01-01
Abstract
: Emotions processing is a complex mechanism that involves different physiological systems. In particular, the Central Nervous System (CNS) is considered to play a key role in this mechanism and one of the main modalities to study the CNS activity is the Electroencephalographic signal (EEG). To elicit emotions, different kinds of stimuli can be used e.g.: audio, visual or a combination of the two. Literature studies focus more on the correct classification of the different types of emotions or which kind of stimulation gives the best performance in terms of classification accuracy. However, it is still unclear how the different stimuli elicit the emotions and which are the results in terms of brain activity. In this paper, we analysed and compared EEG signals given by eliciting emotions using audio and visual stimuli or a combination of the latter two. Data were collected during experiments conducted in our laboratories using IAPS and IADS dataset. Our study confirmed literature physiological studies about emotions highlighting higher brain activity in the frontal and central regions and in the δ and θ bands for each kind of stimulus. However, audio stimulation was found to have higher responses when compared to the other two modalities of stimulation in almost all the comparisons performed. Higher values of the δ/β ratios, an index related to negative emotions, have been achieved when using only sounds as stimuli. Moreover, the same type of stimuli, resulted in higher δ-β coupling, suggesting a better attention control. We concluded that stimulating subjects without letting them know (seeing) what is actually happening may give a higher perception of emotions, even if this mechanism remains highly subjective. Clinical Relevance- This paper suggests that audio stimuli may give higher perception of the elicited emotion resulting in higher brain activity in the physiological areas and more focused subjects. Thus using only audio in emotion related studies may give more reliable and consistent results.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
