Review on Stimuli Presentation for Affect Analysis Based on EEG

Abstract

This work presents a comprehensive review on stimuli presentation, which is an important stage of any emotion elicitation experiment in affect analysis. Due to lack of standard guidelines, the researchers employ their self-devised methods which are not always sufficiently informative-making this area very inconsistent and ambiguous. In addition, an ample study about this stage including how to select, design and present the stimuli has not been reported properly earlier. In this purpose, an inclusive study has been conducted aiming to summarize various aspects of stimuli presentation including type of stimuli, available database, presentation tools, subjective measures, ethical issues and so on. Certainly, among several methods of emotion recognition (e.g., facial expression, speech, gesture and physiological signal), the EEG based emotion recognition works have been considered here due to availability of sufficient number of works, reliability and well-established technology. In total, 137 peer reviewed articles have been studied and the results show that about 83% of emotion elicitations have been performed by employing visual stimuli (mostly pictures and video). Therefore, presentation of visual stimuli has been explored with great emphasis covering laboratory setup, presentation timing, subjective issues, and ethical issues. Finally, an extensive recommendations regarding stimuli presentation has been provided which could guide to conduct the emotion elicitation experiments effectively.

BED: A New Data Set for EEG-Based Biometrics

“…1) Affective Stimulus (AS): These are a set of images that aimed to elicit typical emotional responses during EEG acquisition. They were obtained from two publicly available image datasets that are commonly used in affect detection studies [52], i.e., the Geneva Affective Picture Dataset (GAPED) [53] and the Open Affective Standardised Image Set (OASIS) [54]. GAPED contains 730 different images with a resolution of 640 × 479 pixels encoded in JPEG format and rated in terms of arousal and valence within the range [0, 100].…”

A Review of EEG and fMRI Measuring Aesthetic Processing in Visual User Experience Research

“…Recently, review research has primarily focused on the relationship between emotions and EEG studies [ 7 , 17 ]. They discussed how valence and arousal influence the user experience quality and summarized the corresponding activated brain regions for anger, disgust, fear, happiness, and sadness.…”

“…It reveals the perception of beauty, especially of geometric shapes [69]. (6) Recognition rather than recall (7) Flexibility and efficiency of use (8) Aesthetic and minimalist design (9) Help for users to recognize and diagnose errors and to recover from them (10) Help and documentation Many interactive research experiments have used these ten principles to test users' evaluations on the development of HCI installations and prototypes. ese evaluations are quite useful on 2D program applications, but psychophysiological data could be more appropriate for UX or UI designs in the virtual environment [88].…”

“…Visual interaction is a critical part of UX, and therefore, neuroaesthetics can contribute to the development of current UX biofeedback. A past brain research review reported that 40% of stimuli concern visual content, 43% of stimuli combine visual and audio, 15% of stimuli contain audio content, and 2% contain others [ 7 ]. Neuroaesthetics has been described as a study investigating the neural substrates of aesthetic appreciation [ 8 ], and it comprises three aspects: “cognitive neuroscience of aesthetics,” “cognitive neuroscience of art,” and “cognitive neuroscience of beauty.” However, the sense of “beauty” or “aesthetic” is primarily subjective and is a part of neuroaesthetics.…”

Fear-based health messages for Malay diabetes patients in Singapore

“…These stimuli are most often retrieved from well-known stimuli databases such as the IAPS, GAPED, and NAPS (Balsamo, Carlucci, Padulo, Perfetti, & Fairfield, 2020). In recent years, stimuli in multiple formats such as visual, audio, and visual-audio have also emerged (Sarma & Barma, 2020). Despite the importance of socially relevant emotion stimuli databases (Wei, Roodenrys, Miller, & Barkus, 2020), most databases fail to delineate the actual emotion because there is a disconnect between the stimuli and the actual research participants.…”