Impact of Robot Design Features on Worker-Robot Interaction: A Meta-Analysis and Systematic Review

作者：Siu Shing Man, Xin Zhang, Zhen Qin, Mu Tong, Alan H. S. Chan · 发表于：International Journal of Human-Computer Interaction · 年份：2026 · DOI：10.1080/10447318.2026.2693612 · 研究领域：Human-Automation Interaction and Safety、Social Robot Interaction and HRI、AI in Service Interactions

With the advent of Industry 5.0 and the widespread use of robots, worker-robot interaction (WRI) became crucial. However, there was a lack of systematic reviews on the impact of robot design features on WRI. This study aimed to conduct a meta-analysis and systematic review, searching the relevant literature published from 2015 to 2024 in Google Scholar, Taylor & Francis, Scopus, and ScienceDirect. The data collection from 22 studies that totally recruited 1,254 participants provided 113 effect sizes. The total of 1,254 participants reflects the sum of unique samples across all included studies, with no double-counting of overlapping participants. The 113 effect sizes stemmed from the standard meta-analysis practice of extracting multiple effect sizes per study when appropriate. Robot design features included interaction, appearance, and function. WRI was categorized into subjective evaluation, mental workload, task performance, and situation awareness. Results showed all three types of features had a positive impact on WRI after adjustment for publication bias (overall adjusted effect size g = 0.425), among which functional features had the largest effect (adjusted effect size g = 0.605), while interaction features (adjusted effect size g = 0.341) and appearance features (effect size g = 0.427, no publication bias) had smaller effects. Meta-regression showed that participant type and robot design features were significant moderators of the effect size, while age and automatio...