Influence of agent's self-disclosure on human empathy

Area of Science:

• Human-Computer Interaction
• Artificial Intelligence Ethics
• Social Psychology

Background:

• Growing AI integration necessitates social acceptance of AI agents, including virtual agents and robots.
• Human empathy towards AI agents is crucial for positive interactions and acceptance.
• Agent self-disclosure is explored as a method to enhance human empathy.

Purpose of the Study:

• To experimentally investigate if agent self-disclosure facilitates human empathy towards anthropomorphic AI agents.
• To analyze conditions influencing human empathy with AI agents.

Main Methods:

• A three-way mixed experimental design involving 918 participants.
• Factors included agent appearance (human vs. robot), self-disclosure relevance (high, low, none), and empathy levels (pre/post video stimulus).
• Analysis of variance (ANOVA) was employed for data analysis.

Main Results:

• Agent appearance did not significantly affect empathy levels.
• High-relevance self-disclosure significantly increased human empathy towards AI agents.
• Absence of self-disclosure was found to suppress empathy.

Conclusions:

• Self-disclosure is a key characteristic that enhances human empathy and acceptance of anthropomorphic AI agents.
• The relevance of self-disclosure to the context is critical for its effectiveness.
• Findings support the hypotheses and highlight the importance of strategic self-disclosure in AI design.