Rider Trust and the Role of the Operator in Automated Shuttles

Carriero, Amanda E.; Crabtree, Kaedyn W.; Cooper, Joel M.; Leonard, Blaine D. · 2020 · Proceedings of the Human Factors and Ergonomics Society Annual Meeting

DOI: 10.1177/1071181320641277

archive: archived pipeline: cataloged

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

**Research Question and Motivation** This study investigates the development of rider trust in low-speed, partially autonomous shuttles, specifically addressing how user preconceptions, repeated ridership, unexpected emergency stops (e-stops), and the presence of an onboard operator influence trust. As autonomous shuttles are deployed to solve first-mile/last-mile transportation gaps, understanding trust evolution is critical for maximizing acceptance. The research was motivated by the gap in literature regarding the specific role of the backup operator, who is currently required by NHTSA guidelines, in fostering rider confidence. The authors hypothesized that trust is founded on reliable, positive, and contextually dependent experience over time. **Methods** Data were collected over four weeks at the University of Utah, where an EasyMile autonomous shuttle operated a 0.5-mile route. The study involved 306 participants who completed pre- and post-ride surveys via iPad. Surveys used 7-point Likert scales with mirrored positive and negative framing to detect unthoughtful responses. Additionally, video footage was captured using a GoPro camera inside the shuttle. Researchers used BORIS software to code behavioral data from 19 riders who completed at least two trips. Coding focused on rider behaviors (e.g., casual conversation, cell phone use, anticipatory statements) and operator behaviors (e.g., manual interventions, gestures, preemptive stops). E-stop events were identified through telemetric data and cross-referenced with survey responses. **Findings** Results indicated that increased shuttle ridership significantly predicted more positive experiences, higher confidence in the technology, and better alignment with expectations. Conversely, riders who experienced one or more unexpected e-stops reported significantly lower trust, perceiving the shuttle as operating inconsistently with their expectations and requiring more operator intervention. Pre- and post-ride comparisons showed that actual riding improved perceptions of the shuttle’s role in daily routines and overall experience positivity compared to pre-ride expectations. Video analysis revealed that repeat riders engaged in more casual conversations and made more anticipatory statements about the shuttle’s behavior, suggesting increased comfort and understanding. Furthermore, the onboard operator actively fostered trust through interactions such as answering questions, gesturing to other road users, and performing preemptive stops to avoid abrupt autonomous halts. **Significance** The findings demonstrate that trust in automated shuttles is not static but evolves through positive, repeated exposure. The study highlights that the onboard operator plays a crucial, multifaceted role in maintaining safety and comfort, effectively bridging the gap between automated systems and human expectations. As the industry moves toward fully driverless operations, these results imply that specific challenges—such as managing e-stop discomfort and replacing the operator’s social and communicative functions—must be addressed to sustain rider trust. This research provides actionable insights for transportation authorities and autonomous vehicle developers aiming to improve safety, mobility, and public acceptance of autonomous transit systems.

Key finding

Repeated ridership predicted more positive experiences (tau_b=.164, p=.008), better-met expectations (tau_b=.252, p<.001), and greater confidence in timely arrival; riders also felt safer when the shuttle's informational displays could answer their questions.

Methodology

mixed_methods

Sample size: 306

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• trust calibration
• automation surprise
• acceptance adoption
• automation
• driverless ads

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: self report data