Related Experiment Video
Updated: Jul 16, 2026
07:14
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
Published on: December 23, 2025
Nonverbal AI-Based Communication Robot for Staff in Disaster-Affected Care Facilities: Exploratory ABAB Intervention
Daijiro Haba1, Riko Miyashita2, Takao Kondo3
1Department of Well-being Nursing, Graduate School of Nursing, Ishikawa Prefectural Nursing University, 1-1 Gakuendai, Kahoku City, Ishikawa, 929-1210, Japan, +81-76-281-8335.
JMIR Formative Research
|July 14, 2026
Summary
A pilot study found that nonverbal AI robots did not improve the quality of life (QoL) or well-being of care staff in disaster zones. While safe and feasible, their efficacy as a mental health support intervention requires further study.
Area of Science:
- Disaster medicine
- Human-robot interaction
- Mental health support
Background:
- Noto Peninsula earthquake impacted medical facilities and staff in Japan.
- Care facility staff in disaster areas are survivors with limited psychological support.
- Nonverbal social robots' impact on staff QoL in disasters is unknown.
Purpose of the Study:
- To explore AI robot's association with changes in staff QoL and well-being under disaster conditions.
- To evaluate the safety, acceptability, and continued use of AI robots.
- To assess the potential of AI robots as a mental health intervention for care staff.
Main Methods:
- Pragmatic, exploratory pilot study with an ABAB design (February-June 2025).
- Involved 49 care facility staff across dementia, general, and short-stay units.
- Measured health-related QoL (EQ-5D-5L), well-being (WHO-5), and positive mental health (MHC-SF).
Main Results:
- No significant phase effect observed for QoL, well-being, or mental health measures.
- No robot-related adverse events were reported; deployment was feasible.
- Some positive unadjusted feedback noted in dementia care, but not statistically significant after correction.
Conclusions:
- Short-term AI robot use did not measurably improve staff QoL or well-being in disaster settings.
- Robot deployment was feasible and safe, but efficacy as a mental health intervention is unproven.
- Further research with larger sample sizes is needed to confirm findings.
