Artificial Empathy in Therapy and Healthcare: Advancements in Interpersonal Interaction Technologies

Tianyu Jia, Feiyu Pan, Xingchen Yang, Linhong Ji, Dario Farina, Chong LiCyborg 2025 · DOI

TL;DR

This review explores how artificial empathy, through technologies like multiplayer games, social robots, and virtual agents, can address critical workforce shortages in healthcare and therapy. By simulating human-like interactions and emotional resonance, these advancements aim to improve patient well-being and treatment outcomes, revolutionizing care delivery.

What Problem Does This Paper Address?

The healthcare sector is facing a severe and growing workforce shortage, projected to leave millions of positions unfilled. While technology has been instrumental in automating tasks and improving efficiency, a crucial element often overlooked is the emotional support and empathy that human caregivers provide. This is particularly vital in therapy and healthcare, where a patient's emotional state significantly impacts their well-being and recovery. The paper addresses the challenge of replicating this essential human capacity for empathy in technological systems to augment human roles and improve the quality of care, especially in resource-constrained environments.

Methods

This review categorizes and examines three primary technical platforms for achieving artificial empathy:

  • Multiplayer Games: These platforms facilitate real-time human-to-human social interactions, extending genuine empathy into digital environments. They are used in rehabilitation to strengthen social connections and enhance patient engagement.
  • Social Robots: These are physically embodied robots designed to interact with humans through various social cues like gaze, gestures, facial expressions, and speech. They aim to provide engaging and effective complements to traditional human interaction, offering support in daily living, exercise guidance, and emotional expression.
  • Virtual Agents: These agents exist in digital environments and interact with users through computer interfaces, simulating human characteristics. They can serve roles similar to social robots, offering companionship, advice, and therapeutic support without the physical embodiment, thus reducing costs and logistical constraints.

The paper also highlights a general pipeline for eliciting artificial empathy, involving multimodal signal collection from humans and agents, processing through sensors and processors, and generating responses to create a closed-loop interaction.

Datasets

Not discussed in the paper. The paper is a review and does not present new experimental data or datasets.

Key Insights and Ideas

  • Empathy as a Critical Missing Piece: The paper strongly emphasizes that while technology excels at functional tasks, the emotional and empathic component of human interaction is crucial for effective therapy and healthcare, and its absence in current systems is a significant gap.
  • Multi-faceted Nature of Empathy: Empathy is presented as a complex construct with cognitive, emotional contagion, affective, and compassionate dimensions, highlighting the challenge in replicating it computationally.
  • Hybrid Nature of Multiplayer Games: Multiplayer games are seen as a unique approach, not directly simulating empathy algorithmically, but acting as a mediating technology that amplifies existing human-to-human empathy within digital therapeutic contexts.
  • The "Uncanny Valley" Challenge: For social robots, the paper points out the difficulty in designing anthropomorphic appearances that are appealing without falling into the "uncanny valley," where near-human likeness can evoke discomfort.
  • AI's Role in Enhancing Artificial Empathy: The integration of AI, particularly large language models (LLMs) like ChatGPT, is highlighted as a key enabler for creating more responsive, personalized, and adaptive virtual agents and robots, moving beyond pre-programmed interactions.
  • Ethical Considerations are Paramount: The review stresses the importance of artificial empathy serving as an adjunct to, rather than a replacement for, human interaction, and warns against misuse, false attachment, and the potential for AI hallucinations in critical healthcare applications.

Key Results

The review synthesizes findings from numerous studies across the three platforms:

  • Multiplayer Games: Studies show that patients often exhibit superior performance and engagement in multiplayer modes compared to single-player modes. However, benefits are not universal, with some reporting no significant gains or even stress responses.
  • Social Robots: Robots have shown considerable potential in supporting individuals with Autism Spectrum Disorder (ASD), improving their social behaviors and interaction skills. They have also been well-received in rehabilitation and elderly care, leading to increased motivation, adherence, and positive emotional outcomes.
  • Virtual Agents: Virtual agents are capable of simulating various interpersonal interactions and have been used effectively in roles like exercise advisors, tutors, and smoking cessation counselors. They have shown promise in improving user engagement and adherence to therapeutic programs.

Limitations and Future Work

  • Limitations:
    • Many studies reviewed use small sample sizes and short-term interventions, limiting generalizability.
  • A reliance on self-report questionnaires for outcome measures can limit comparability and objectivity.
  • The methodological heterogeneity across studies makes direct comparison difficult.
  • A lack of rigorous control conditions in many studies makes it hard to attribute benefits solely to artificial empathy.
  • The ethical implications of pseudo-empathy and potential over-reliance on AI systems need careful consideration.
  • AI models, especially GenAI, can still suffer from hallucinations, posing risks in healthcare.
  • Future Work:
    • Develop unified frameworks, integrate multimodal assessment indicators, and implement rigorous control designs for future research.
  • Focus on amplifying positive effects of interpersonal engagement through technological innovation, particularly emotional and cognitive resonance.
  • Develop artificial empathy that can perceive, understand, and respond to users' physiological and psychological needs in a personalized manner.
  • Enhance emotion recognition capabilities and integrate them with other psychological and cognitive metrics for a comprehensive understanding of human behavior.
  • Address the challenges of ecological validity, cultural differences in emotion expression, and ensuring reliable emotion recognition across diverse clinical environments.
  • Further explore the role of memory and personality in adapting interaction styles for more personalized and effective human-agent interactions.
  • Continue to advance AI technologies, especially LLMs and GenAI, to create more fluid, natural, and adaptive human-like behaviors in agents.
  • Ensure that the development and deployment of these technologies adhere to ethical and legal standards, prioritizing support over replacement of human care.

Summary updated

Abstract

The healthcare sector is challenged by critical workforce shortages, and this is causing an urgent need for innovative technologies to support or augment human roles. Although much of the research effort has focused on support and training of functional tasks, the emotional impacts that humans bring to the loop have often been overlooked. This gap is particularly pressing in healthcare and therapy, where empathy and emotional support are central to patient well-being. Unlike machines, humans possess a unique capacity for empathy, connecting emotionally with others and providing the essential support that fosters healing. Bridging this gap requires integrating affective elements, such as empathy, into therapeutic systems, which is the key to improving their effectiveness. This review explores groundbreaking techniques that integrate interpersonal interactions within therapy and healthcare, focusing on multiplayer games that strengthen real-time social connections, alongside social robots and virtual agents designed to simulate human-like affective interactions. Using artificial intelligence, these technologies aim to replicate complex human dynamics and foster artificial empathy, thus revolutionizing how we deliver care and support.
BibTeX
@inproceedings{jia2025artificial,
  author = {Tianyu Jia and Feiyu Pan and Xingchen Yang and Linhong Ji and Dario Farina and Chong Li},
  title = {{Artificial Empathy in Therapy and Healthcare: Advancements in Interpersonal Interaction Technologies}},
  booktitle = {Cyborg},
  year = {2025},
  doi = {10.34133/cbsystems.0473},
}
Figure