Current conversational agents (CA) have seen improvement in conversational quality in recent years due to the influence of large language models (LLMs) like GPT3. However, two key categories of...

ALT: a woman in a judge 's robes is pointing at someone

Current conversational agents (CA) have seen improvement ...

Background: Empathy is a critical component of effective mental health care communication. Positive perceptions of empathy in conversational agents (CAs) operating in the healthcare domain are needed to enhance the quality of care provided by these emerging technologies. Objective: This study investigates how demographic characteristics of individual users affect their perception of empathy in a mental health helpline service context. The impact of vocal cues engineered within nine CA prototypes is the focus of this study. Methods: A group of psychology students (n=306) were presented with several CA prototypes, and their perceptions of empathy were evaluated via an online survey. Perceptions of empathy were rated using the Perceived Emotional Intelligence (PEI) Scale and the Rater Scale (RS10). Results: A linear mixed model for PEI showed significantly higher ratings for high than low engineered empathy levels (t(8)=-2.34, p=0.048), and for participants of Asian and the Other category rather than Oceanic ethnicities (t(285)=2.54, p=0.01 and t(286)=2.25, p=0.03 respectively). RS10 ratings were also significantly higher for high rather than low engineered empathy levels (t(8)=-2.45, p=0.04) and for participants of the Other category rather than Oceanic ethnicities (t(284)=2.24, p=0.03). Women also showed significantly higher RS10 ratings than men (t(283) =1.94, p=0.05). Conclusions: Recognising empathy levels in CA prototypes proved challenging, highlighting possible complexities involved with voice-based empathy detection. The perception of empathy may also be affected by different ethnic and gender-based factors. The study findings emphasise the importance of personalised empathic communication by CAs with users based on their ethnicity and gender in the context of mental health helpline service provision