Does Care Lead to Bonds? Exploring the Relationship Between Human Caregiving for Robots and Human-Robot Bonding

Research Background and Issues

• Identified Problems or Challenges:
  Existing research primarily focuses on robots providing care to humans, emphasizing that caring behaviors help establish human-robot relationships. However, compared to scenarios where robots care for humans, the reverse situation—humans caring for robots—has not been sufficiently studied. This leaves a significant knowledge gap: whether human caring behaviors toward robots can effectively promote human-robot relationships and the mechanisms behind this process remain unknown.

• Significance of the Issue:
  In scenarios requiring long-term interaction, such as daily life, healthcare, and education, emotional connections between humans and robots can enhance user acceptance, increase the frequency of service robot usage, and potentially alleviate loneliness and stress. Moreover, building strong human-robot relationships can extend the emotional and practical lifespan of robotic technologies, supporting sustainable social development.

• Research Motivation and Related Work:
  Although caring behaviors among humans, animals, and objects often lead to emotional connections (e.g., the "care effect"), there is a lack of consistent empirical research on the specific contexts and effects of such behaviors in human-robot interactions. Additionally, the distinction between types of caring behaviors (emotional vs. instrumental) and their impact on human-robot relationships has not been explicitly tested.

Proposed Solution

• Proposed Method or Solution:
  The authors designed experiments to investigate the effects of two types of caring behaviors (emotional and instrumental) in human-robot interactions. Specifically, participants provided emotional care (e.g., comforting the robot) or instrumental care (e.g., helping the robot recharge), and the strength of the human-robot relationship was assessed using both explicit and implicit measures.

• Innovative Aspects of the Solution:

  1. Systematically distinguishes the effects of emotional and instrumental care on human-robot relationships, addressing a gap in existing research.
  2. Introduces implicit measures (e.g., reluctance to replace the robot and compliance with robot requests) to complement traditional explicit measures based on subjective questionnaires.
  3. Extends the classic "caring behavior-interpersonal relationship" theoretical model to the field of human-robot interaction and develops an effective experimental paradigm.

• Implementation Steps and Key Techniques:

  1. Experimental Design:
     • 88 participants completed three rounds of a game in which the robot failed and requested care from participants: emotional care (e.g., comforting) or instrumental care (e.g., recharging). In the control group, the robot did not request any care.
  2. Measurement Metrics:
     • Explicit metrics: such as perceived closeness, social attractiveness, and intention for future interaction, collected via subjective questionnaires.
     • Implicit metrics: such as reluctance to replace the robot and compliance with voluntary tasks requested by the robot, measured through behavioral observations.
  3. Statistical Analysis:
     • One-way ANCOVA and survival analysis were used to evaluate the effects of the three conditions on both explicit and implicit metrics of human-robot relationships.
     • Control variables: participants' prior familiarity with robots.

Research Findings

• Specific Findings:

  1. Emotional care effectively enhances human-robot relationships:
     • In implicit measures, participants providing emotional care exhibited stronger reluctance to replace the robot and were more willing to spend time completing tasks requested by the robot.
  2. Instrumental care has weaker but potential effects:
     • Instrumental care showed slight advantages in implicit measures (e.g., reluctance to replace the robot) but did not reach statistical significance.
  3. Implicit metrics outperform explicit metrics:
     • In explicit measures (e.g., social attractiveness, intention for future interaction), no significant differences were observed among the three groups. However, implicit metrics (e.g., behavioral compliance) successfully revealed the potential impact of caring behaviors on human-robot relationships.

• Comparison with Existing Solutions:
  Unlike traditional research focusing on how robots care for humans, this study adopts a reverse perspective by examining human caring behaviors toward robots. It addresses the limitations of existing research that relies solely on explicit subjective data by incorporating implicit measures.

• Experimental and Evaluation Results:

  • In the emotional care condition, participants demonstrated significantly higher:
    • Reluctance to replace the robot (compared to the control group).
    • Compliance with voluntary tasks (with significantly lower task interruption risks than the control condition).
  • Instrumental care had a near-significant effect on reluctance to replace the robot but was less effective than emotional care.

• Limitations and Future Directions:

  1. Difficulty isolating the independent effects of caring behaviors:
     • The current experiment intertwined the robot's display of vulnerability with participants' caring behaviors, making it challenging to precisely separate their independent effects. Future studies could design scenarios with stricter control over these components.
  2. Lack of research on the long-term dynamics of human-robot relationships:
     • This study only examined initial interaction contexts. Future research should explore the sustained impact of caring behaviors on long-term human-robot relationships.
  3. Sample characteristics and external validity:
     • Participants primarily had technical backgrounds, which may reflect the preferences and perceptions of a limited user group. Future studies should validate the findings with a broader and more diverse population.

In summary, this study is the first to systematically evaluate the differential impacts of emotional and instrumental care on human-robot relationships. It highlights the potential of implicit measures and provides empirical support for designing care-oriented interactions and advancing theoretical development in the field of human-robot interaction (HRI).