Exploring the Interplay Between Voice, Personality, and Gender in Human-Agent Interactions

To foster effective human-agent interactions, designers need to identify characteristics that could affect how agents are perceived and accepted, and to what extent they could impact rapport-building. Aiming to explore the role of user-agent synchrony, we assessed 388 participants to determine whether they could perceive personality traits from four artificial voices we selected and adapted from human samples, considering gender (male or female) and personality (introvert or extrovert) as grouping factors. Our findings suggest that participants were able to significantly differentiate female agents by personality, while male agents were not consistently distinguished. We also observed evidence of personality synchrony, where participants tended to perceive the first agent as more similar to their own personality, with this effect driven mainly by male participants, especially toward male agents. This paper contributes findings and insights to consider the interplay of user-agent personality and gender synchrony in the design of human-agent interactions.

💡 Research Summary

This paper investigates how voice, gender, and personality interact to shape user perceptions of voice‑only virtual agents. The authors recruited 388 participants and created four synthetic agent voices derived from real human speakers: male‑introvert, male‑extrovert, female‑introvert, and female‑extrovert. Participants were divided into two groups (male‑only and female‑only agents) and each interacted with two agents of the same gender but opposite personality, presented sequentially. Personality was measured using the Ten‑Item Personality Measure (TIPI) focusing on the extroversion dimension; participants also completed the TIPI for themselves prior to the experiment.

The study addressed two research questions: (1) how voice, gender, and personality cues jointly influence user perceptions of agents, and (2) the extent to which users project their own personality onto agents (personality synchrony). Statistical analyses (chi‑square tests and mixed‑effects models) examined (a) participants’ ability to correctly identify the agent’s personality and (b) the relationship between participants’ self‑reported extroversion and their perception of the first agent.

Key findings are threefold. First, participants reliably distinguished the personality of female agents (significant differentiation between introvert and extrovert voices), whereas the personality of male agents was not consistently identified. This suggests that vocal cues in female speech (pitch, timbre, prosodic variation) convey personality information more effectively than those in male speech. Second, a personality synchrony effect emerged: participants tended to rate the first agent as more similar to their own personality, with the effect driven primarily by male participants. Third, the synchrony effect was strongest when male participants interacted with male agents, indicating that gender homophily may amplify personality alignment.

Methodologically, the paper’s strengths include the use of real human voice samples to generate synthetic agents, a concise yet validated personality measure (TIPI), and a within‑subject design that allows assessment of order effects. Limitations are noted: (a) personality was reduced to a single extroversion dimension, ignoring other Big Five traits; (b) the study isolated voice from visual or gestural cues, limiting ecological validity for multimodal agents; (c) the participant pool appears to be largely U.S. college‑age individuals, restricting generalizability across cultures and age groups.

The authors discuss implications for design: voice‑only agents should tailor vocal parameters differently for male and female personas, and designers might consider adaptive voice synthesis that aligns with users’ self‑reported personality, especially for male users. Future work is suggested in three areas: expanding to a full multidimensional personality model, incorporating multimodal cues (visual avatars, gestures) to examine whether synchrony persists, and conducting cross‑cultural studies with broader demographic samples. Additionally, real‑time user modeling could enable agents to dynamically adjust pitch, tempo, and prosody to match user personality and gender, fostering greater rapport and trust.

In conclusion, the study provides empirical evidence that gender and personality interact with vocal characteristics to affect perception of voice‑only agents. Female voices convey personality cues more clearly, while male users exhibit stronger personality synchrony with male agents. These insights highlight the importance of considering gender‑specific vocal design and personality alignment when developing conversational agents for diverse user populations.