Human-Robot Teaming Field Deployments: A Comparison Between Verbal and Non-verbal Communication

Abstract: Healthcare workers (HCWs) encounter challenges in hospitals, such as retrieving medical supplies quickly from crash carts, which could potentially result in medical errors and delays in patient care. Robotic crash carts (RCCs) have shown promise in assisting healthcare teams during medical tasks through guided object searches and task reminders. Limited exploration has been done to determine what communication modalities are most effective and least disruptive to patient care in real-world settings. To address this gap, we conducted a between-subjects experiment comparing the RCC's verbal and non-verbal communication of object search with a standard crash cart in resuscitation scenarios to understand the impact of robot communication on workload and attitudes toward using robots in the workplace. Our findings indicate that verbal communication significantly reduced mental demand and effort compared to visual cues and with a traditional crash cart. Although frustration levels were slightly higher during collaborations with the robot compared to a traditional cart, these research insights provide valuable implications for human-robot teamwork in high-stakes environments.

• The paper demonstrates that verbal communication significantly reduces cognitive load during emergency procedures compared to non-verbal cues and standard methods.
• The study employs a between-subject design with 115 participants using NASA-TLX and ACIR-Q to assess workload and attitudes toward robotic collaboration.
• The findings highlight the need for improved RCC designs that balance individual efficiency with wider organizational acceptance in high-pressure clinical settings.

An Evaluation of Communication Modalities in Human-Robot Teams: Insights from Emergency Healthcare Deployments

The presented paper investigates the efficacy of verbal and non-verbal communication in the deployment of robotic crash carts (RCCs) in emergency healthcare settings, specifically within the fast-paced, cognitive-demanding environment of hospital emergency rooms. This research addresses a critical gap in determining the optimal communication modalities that support healthcare workers (HCWs) during high-stakes procedures while minimizing potential disruption to patient care.

Study Context and Methodology

The examination was performed in a medical simulation setting mimicking real-world emergency scenarios involving resuscitation procedures. The study engaged 115 participants, including nurses, child life specialists, and fellows at various training levels, who were assigned to one of three conditions: verbal communication via robotic speech, non-verbal communication using LED cues, and standard crash carts (control group), thereby employing a between-subjects experimental design.

Data were collected using self-reported workload measures (NASA-TLX) and attitudes toward cooperative robots in the workplace (ACIR-Q), alongside video recordings of simulation activities. Additionally, Wizard-of-Oz (WoZ) methodology was deployed to operationalize RCC interactions, allowing highly controlled investigations into RCC behavior and communication efficacy.

Key Findings

Workload Impact

Analyses revealed that verbal communication via RCCs significantly reduces mental demand and effort compared to traditional and non-verbal communication methods. This reduction in perceived cognitive load highlights the importance of direct verbal prompts in facilitating rapid item retrieval and thereby optimizing task execution under time constraints.

Conversely, while verbal communication reduced the mental and physical effort, RCC usage introduced a slight increase in frustration compared to traditional crash carts. The paper posits that this could arise from participants' unfamiliarity with robotic technology or unmet expectations related to object availability within the RCC, emphasizing the need for transparent communication of robotics capabilities and limitations.

Attitudes Toward Robotic Integration

The ACIR-Q results signaled general acceptance toward RCCs in healthcare environments, although respondents reported concerns relating to job security and changes in colleague interactions. Interestingly, workplace outcomes were more positively perceived in the non-verbal condition, contrary to the lower workload ratings associated with this modality.

This dichotomy suggests that while verbal prompts are advantageous for individual task efficiency, non-verbal cues might support broader organizational dynamics, highlighting a complex interaction between individual and collective perspectives on robotic integration.

Implications and Future Directions

This work underscores the potential for RCCs to improve ER team performance by alleviating cognitive burdens through optimal communication strategies. Practically, deploying RCCs in high-stress environments necessitates design adjustments, such as improved LED visibility and refined verbal prompts, which could enhance human-robot interaction efficacy and user experience.

Theoretically, these findings contribute to the body of HRI knowledge by revealing interdisciplinary dynamics of robotics utility in healthcare, emphasizing the need for synchronized individual and organizational evaluations of robots.

Future endeavors should consider enhancing RCC autonomy through sensing capabilities and closed-loop communications to augment their real-time adaptability to dynamic clinical workflows. Further validation of RCCs' ability to minimize object retrieval times and potential expansions into diverse medical contexts could reinforce their utility in emergency medicine.

Overall, this research provides a methodological framework and actionable insights for advancing human-robot teamwork, leveraging multimodal communication to enhance healthcare delivery in critical care settings.