When Transparency May Backfire: Exploratory Signals on Voice and Explanation as a Home Robot Decides Against You

Abstract

A home robot can explain why it overrules a household member yet leave that member feeling less influential. A scenario-based 2 × 2 video-vignette pilot (N = 31) manipulated procedural voice—a depicted opportunity to express a preference before the decision—and explanation while holding an unfavorable outcome constant. Tests of procedural justice (proposed mediator) and decision acceptance (focal outcome) were imprecise and nonsignificant. Both checks differed by condition: explanation was strongly salient (M = 4.93 vs. 2.29; d = 1.77), whereas represented voice was moderately salient (M = 3.81 vs. 2.47). Exploratorily, explanation lowered felt opportunity and influence (M = 3.54 vs. 4.88). These uncorrected pilot signals motivate confirmatory tests; they do not show that explanation generally backfires.

인용 정보

@inproceedings{kim2026transparency,
  author = {Kim, Soo Y. and Lee, Changjun},
  title = {When Transparency May Backfire: Exploratory Signals on Voice and Explanation as a Home Robot Decides Against You},
  booktitle = {Proceedings of the 14th International Conference on Human-Agent Interaction},
  series = {HAI '26},
  year = {2026},
  location = {Osaka, Japan},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  numpages = {3}
}

키워드