This paper discusses the importance of incorporating emotional intelligence into next-generation human–computer interaction (HCI) systems to make them more human-like, effective, and efficient. Emotional intelligence involves recognizing and managing affective states, which are modulated by nonverbal cues such as facial expressions, body movements, and vocal and physiological reactions. While humans naturally detect and interpret these cues, developing an automated system to do so is challenging. The paper surveys past research on this topic and provides recommendations for developing an automatic personalized analyzer of nonverbal affective feedback. Emotional intelligence is crucial for successful interpersonal communication and has been argued to be a better predictor of success than IQ. The paper highlights the need for affect-sensitive HCI systems that can sense and respond to users' affective states. It discusses the challenges of detecting and interpreting affective states, including the lack of consensus on the nature of emotions and the cultural dependency of emotional expressions. The paper also explores the importance of multimodal analysis, which combines multiple sensory inputs to improve the accuracy and robustness of affect recognition. The paper outlines the problem domain of affective computing, including the challenges of detecting and interpreting affective states. It discusses the psychological issues related to the nature of emotions and the importance of context in interpreting affective cues. The paper also addresses the technical challenges of integrating different modalities and the need for robust, accurate, and context-sensitive affect recognition systems. The paper reviews the current state of the art in machine analysis of human affective feedback, highlighting the challenges and opportunities in this field. It discusses the importance of multimodal analysis, the role of temporal information, and the need for context-sensitive systems. The paper also addresses the challenges of automatic facial expression analysis, including the need for robust and accurate systems that can handle variations in lighting, head movements, and distractions. The paper concludes that the development of affect-sensitive HCI systems is essential for creating more natural, effective, and efficient interactions between humans and computers. It emphasizes the need for further research and development in this area to overcome the challenges and improve the performance of affect recognition systems.This paper discusses the importance of incorporating emotional intelligence into next-generation human–computer interaction (HCI) systems to make them more human-like, effective, and efficient. Emotional intelligence involves recognizing and managing affective states, which are modulated by nonverbal cues such as facial expressions, body movements, and vocal and physiological reactions. While humans naturally detect and interpret these cues, developing an automated system to do so is challenging. The paper surveys past research on this topic and provides recommendations for developing an automatic personalized analyzer of nonverbal affective feedback. Emotional intelligence is crucial for successful interpersonal communication and has been argued to be a better predictor of success than IQ. The paper highlights the need for affect-sensitive HCI systems that can sense and respond to users' affective states. It discusses the challenges of detecting and interpreting affective states, including the lack of consensus on the nature of emotions and the cultural dependency of emotional expressions. The paper also explores the importance of multimodal analysis, which combines multiple sensory inputs to improve the accuracy and robustness of affect recognition. The paper outlines the problem domain of affective computing, including the challenges of detecting and interpreting affective states. It discusses the psychological issues related to the nature of emotions and the importance of context in interpreting affective cues. The paper also addresses the technical challenges of integrating different modalities and the need for robust, accurate, and context-sensitive affect recognition systems. The paper reviews the current state of the art in machine analysis of human affective feedback, highlighting the challenges and opportunities in this field. It discusses the importance of multimodal analysis, the role of temporal information, and the need for context-sensitive systems. The paper also addresses the challenges of automatic facial expression analysis, including the need for robust and accurate systems that can handle variations in lighting, head movements, and distractions. The paper concludes that the development of affect-sensitive HCI systems is essential for creating more natural, effective, and efficient interactions between humans and computers. It emphasizes the need for further research and development in this area to overcome the challenges and improve the performance of affect recognition systems.