Quantitative Computation of Emotion Perception and Visual Cognition: Advances and Applications

Affective computing is a field at the intersection of psychology and computer science that focuses on the analysis of human emotions through multiple modalities, such as facial expression recognition (dynamic expressions and micro-expressions), speech emotion recognition, multimodal emotion recognition, and physiological signals (e.g., EEG/ECG/MEG). Although many current computational techniques for emotion recognition rely on artificial intelligence (AI), researchers increasingly aim to integrate psychological theory and neuroscience evidence to improve the accuracy and interpretability of emotion detection and classification. Their goal is to increase precision in emotion categorization and broaden the range of emotions that can be identified, while also advancing cognitive theories of how emotions interact with perception, attention, memory, decision-making, and cognitive control. We particularly welcome work that links computational measures of emotion to cognitive processes and behavior in well-controlled experimental paradigms, and that provides interpretable models aligned with neural and behavioral evidence.1. Emerging theories, methods, and applications for measuring emotional intelligence using AI, including (but not limited to) self-awareness, self-management, self-motivation, empathy, and social skills—core capacities that support how people understand and manage emotions and relationships.2. Emerging theories, methods, and applications in emotion research integrating psychology and AI.3. Emerging theories, methods, and applications in dynamic facial expression recognition, micro-expression recognition, action unit detection, speech emotion recognition, multimodal emotion recognition, and biosignal-based emotion recognition (e.g., EEG/ECG/MEG).4. We also encourage submissions that employ multimodal approaches to test explicit hypotheses about cortical communication and integration.5. Emotion–cognition interactions: computational and experimental work on how emotion shapes (and is shaped by) perception, attention, learning, memory (working and long-term), decision-making, and cognitive control (e.g., inhibition, shifting, updating).6. Computational cognitive modeling under affective states: model-based analyses linking affective signals to latent cognitive variables (e.g., learning rates, uncertainty, value representations, drift diffusion parameters), including reinforcement learning, Bayesian models, and predictive processing frameworks.7. Interpretable and mechanistic approaches: studies that improve model interpretability, representation analysis, or alignment between AI representations and behavioral/neural measures, in order to explain cognitive mechanisms rather than only classify emotions.This Research Topic invites original research and review articles on affective computing, focusing on the integration of AI, psychology, and neuroscience for advanced emotion recognition. Topics include emerging methods in emotional intelligence, multimodal emotion recognition (facial, speech, physiological signals), and computational models of emotional and cognitive processes. Submissions should present novel insights, methodologies, or applications, be previously unpublished and not under consideration elsewhere, and adhere to ethical guidelines. Peer review will ensure scientific quality and relevance to the field. Topic editors: Feng Liu, Yijun Ge, Panagiotis Tzirakis, Fabien Ringeval, Changzeng Fu Source: https://www.frontiersin.org/research-topics/77305/quantitative-computation-of-emotion-perception-and-visual-cognition-advances-and-applications