Affective Reaction And Emotion Identification Induced By Different Sound Stimuli

International Affective Digitized Sounds (IADS-2) were used to investigate emotional responses to different acoustic stimuli in Chinese population. The Self-Assessment Manikin (SAM) was used to assess the two dimensions of pleasure and arousal. The results showed that the distribution of the ratings in the affective space was virtually identical to that found in the original normative study of the IADS-2. However, for a specific sound sample, there are some differences between these two countries. Cultural and educational factors may be responsible for these differences. Therefore, sounds should be selected based on the local assessment results when study the two dimensions of pleasure and arousal in future.The sounds used in the experiment above were sampled and analyzed using ArtemiS software to acquire their acoustic parameters. The importance of various acoustic parameters for pleasure and arousal evaluation was statistically analyzed. On this basis, artificial neural network models was been explored at two levels. The results showed that the dimension of arousal had a relatively high correlation with acoustic parameters; prediction results showed that the two models were able to predict arousal rating levels with great accuracy and Model Two present a more wide application prospect because of its simple structure. In contrast, the dimension of pleasure showed a relatively low correlation with acoustic parameters; prediction results showed the two models had low reconition rates, mainly because pleasure is highly related to personal preference, experience and other social and cultural factors.To explore the relationship between EEG and the two dimensions of pleasure and arousal,24sounds were selected according to the local assessment results. EEG of fifteen students was recorded when listening to the sounds, and theta, alpha, beta power was analyzed. The results showed that midline theta wave power increased significantly under the pleasant sound stimulation; a decrease of left frontal and frontal pole alpha power was also found during the pleasant sound stimulation. While a decrease of right frontal pole alpha power was found during the unpleasant sound stimulation and midline alpha power also decreased compared with pleasant sound stimulation. Alpha power in the temporal area reduced during low-arousal sound stimulation and beta power increased in the right temporal area.