Empathic computing shifts toward augmented reality for real-time human connection and emotion recognition

A systematic review published on February 25, 2025, in the inaugural issue of the journal Empathic Computing provides a comprehensive overview of how immersive technologies foster human connection, emphasizing the untapped potential of Augmented Reality (AR) and Mixed Reality (MR).

Authored by Umme Afifa Jinan, Mark Billinghurst and colleagues from Kennesaw State University, the University of Louisiana at Lafayette, and the University of Auckland, the article examines 24 years of research (2000–2024). The team systematically analyzed 77 high-impact studies to map out key areas of empathic computing, including emotion elicitation, emotion recognition, and cross-disciplinary applications like healthcare therapy, education, and remote collaboration.

"Virtual Reality has been the undisputed champion of empathic computing over the last two decades, allowing users to vividly experience the world from another's perspective," said lead author Umme Afifa Jinan, a researcher in Computer Science at Kennesaw State University. "However, AR and MR remain significantly underexplored. We need to expand into these mediums to unlock their massive potential for real-time, face-to-face empathic collaboration."

The review highlights several critical insights into how machines understand human feelings. Researchers found that dimensional emotional models—which measure continuous states like arousal and valence—have influenced the field far more than discrete models that simply categorize basic emotions like happiness or sadness.

Furthermore, while current systems heavily rely on physiological data such as heart rate, eye tracking, and facial expressions, the study identified a fascinating blind spot: the lower body.

"Most studies have explored hand tracking, head movements, and upper body gestures for recognizing emotions, but few have investigated lower body postures," noted the research team. "Analyzing gait patterns and leg movements could be a groundbreaking new avenue for understanding emotional states, especially for virtual therapy and rehabilitation."

The authors argue that addressing these gaps is essential for the future of human-computer interaction. By expanding into AR, integrating hybrid emotion frameworks, and exploring new behavioral cues like gait analysis, future technologies can evolve from simple immersive environments into deeply personalized, emotionally intelligent systems that genuinely foster human empathy.