Ten challenges for EEG-based affective computing

Abstract

The emerging field of affective computing focuses on enhancing computers’ ability to understand and appropriately respond to people’s affective states in human-computer interactions, and has revealed significant potential for a wide spectrum of applications. Recently, the electroencephalography (EEG) based affective computing has gained increasing interest for its good balance between mechanistic exploration and real-world practical application. The present work reviewed ten theoretical and operational challenges for the existing affective computing researches from an interdisciplinary perspective of information technology, psychology, and neuroscience. On the theoretical side, we suggest that researchers should be well aware of the limitations of the commonly used emotion models, and be cautious about the widely accepted assumptions on EEG-emotion relationships as well as the transferability of findings based on different research paradigms. On the practical side, we propose several operational recommendations for the challenges about data collection, feature extraction, model implementation, online system design, as well as the potential ethical issues. The present review is expected to contribute to an improved understanding of EEG-based affective computing and promote further applications.

A Literature Review of EEG-Based Affective Computing in Marketing

“…The phenomena of coexistence should be considered in affective tagging. We recommend that future studies should focus on two issues to develop a more accurate affective definition and conduct better forecasting: (1) they should aim for a deeper understanding of consumer ambivalence, characterized by the co-occurrence of positive and negative emotions (Kreibig and Gross, 2017;Hu et al, 2019); and (2) consider emotion dyads, namely, a mix of primary emotions, raised by Plutchik (1980). The multidimensional and multimodal feature fusion can obtain better recognition performance.…”

FusionSense: Emotion Classification Using Feature Fusion of Multimodal Data and Deep Learning in a Brain-Inspired Spiking Neural Network

“…As discussed previously, several studies have shown that including EEG features considerably improves the classification accuracy. However, there are several challenges in using EEG for emotion recognition [ 126 ], including the selection of robust features, continous decoding of affective states, reliable decoding of long-term reliability of EEG recordings for such studies, long preparation time, and, most importantly, adopting a proper model of emotion with regard to EEG and understanding the EEG representation of affective states. For an excellent overview of these challenges the reader is directed to [ 126 ].…”

Individual differences in trait creativity moderate the state-level mood-creativity relationship

“…Also, here we relied on the participants’ self-evaluation of their state-level mood and creativity. Given the rapid development of wearable bio-sensing technologies and machine learning methods [ 70 – 72 ], it is expected to have a momentary evaluation of one’s state creativity and mood states in an objective way that could further our understanding of the mood-creativity link. Last but not least, while the mood states were categorized into positive and negative moods, it might be necessary to have a more fine-grained categorization.…”