Estimating Trust in Conversational Agent with Lexical and Acoustic Features

Abstract

As NASA moves to long-duration space exploration operations, there is an increasing need for human-agent cooperation that requires real-time trust estimation by virtual agents. Our objective was to estimate trust using conversational data, including lexical and acoustic features, with machine learning. A 2 (reliability) × 2 (cycles) × 3 (events) within-subject study was designed to provoke various levels of trust. Participants had trust-related conversations with a conversational agent at the end of each event. To estimate trust, subjective trust ratings were predicted using machine learning models trained on three types of conversational features (i.e., lexical, acoustic, and combined). Results showed that a random forest model, trained on the combined lexical and acoustic features, best predicts trust in the conversational agent (R2adj = 0.67). Comparing models, we showed that trust is not only reflected in lexical cues but also acoustic cues. These results show the possibility of using conversational data to measure trust unobtrusively and dynamically.

Introduction to the Emerging Cognitive Science of Distributed Human‐Autonomy Teams

“…Lyons et al (2024) present factors including decision authority, social intent, and relational processes that predict changes in trust in a human-machine interaction. There has been recent growth in the area of trust dynamics (Li, Erickson, Cross, & Lee, 2022;Yang, Schemanske, & Searle, 2023), and we agree with Lyons et al that objective real-time metrics of trust in actual interactions are needed. We speculate that the spatiotemporal spread of trust and distrust as a form of social information is amenable to interactive and real-time team cognition analysis with, for example, influence (communicative or behavioral) being a generalizable mechanism of spread Huang et al, 2021;Zhou et al, 2023).…”

“…(2024) present factors including decision authority, social intent, and relational processes that predict changes in trust in a human–machine interaction. There has been recent growth in the area of trust dynamics (Li, Erickson, Cross, & Lee, 2022; Yang, Schemanske, & Searle, 2023), and we agree with Lyons et al. that objective real‐time metrics of trust in actual interactions are needed.…”

Bimodal Trust: High and Low Trust in Vehicle Automation Influence Response to Automation Errors

“…We also demonstrate how trust dynamics vary across drivers. Specifically, while all participants experienced the same automation, their responses to the automation differed (Li et al, 2023). Mixture models helped reveal that one group of drivers became complacent fail-ing to respond to automation errors promptly.…”

Explaining Trust Divergence: Bifurcations in a Dynamic System

“…Each level of reliability had 2 repeated cycles of the CDRS tasks, each including 3 events (i.e., startup, venting, shutdown). Details of the study were documented in (Li et al, 2022).…”