Abstract
In the interactions between human and AI agents, the language strategies of the agent play a crucial role in shaping user perception and overall user experience. While language strategies have been shown to influence users’ emotions in various application domains, limited research has examined their effects in driving contexts, particularly in interactions based on external human–machine interfaces (eHMIs). In practical traffic environments, eHMIs serve as a key communication channel through which autonomous vehicle agents convey information to other road users. This study introduces an eHMI language strategy framework based on text and emoji that enables autonomous agents to express distinct language styles and tonal variations. A user study (N = 20) was conducted to investigate how these language strategies influence drivers’ emotional responses, visual attention, and overall interaction experience. Participants operated a vehicle in a simulated driving cockpit, during which an autonomous vehicle using eHMIs with varying language strategies executed a lane-cutting maneuver. The results indicate that anthropomorphic language styles significantly reduce the negative emotions typically induced by cut-in events and improve both attentional focus and user acceptance of the interaction. However, in certain scenarios, an affinitive tone was less effective than a neutral tone, suggesting the tone should be contextualized. This study reveals the cognitive-affective dynamics triggered by language strategies in autonomous driving eHMIs and provides empirical evidence and design recommendations for the development of emotionally adaptive and semantically rich eHMI systems.
Keywords
Intelligent traffic agent; External human-machine interface; Language style; Emotion regulation
DOI
https://doi.org/10.21606/iasdr.2025.561
Citation
Hu, Y., Cao, Q., Liu, Y., Fang, S.,and Zhang, H.(2025) Cognitive-affective dynamics in AI eHMI: adaptive language strategy for emotion and attention modulation, in Chang, C.-Y., and Hsu, Y. (eds.), IASDR 2025: Design Next, 02-05 December, Taiwan. https://doi.org/10.21606/iasdr.2025.561
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Conference Track
Track 4 - Human-Centered AI
Cognitive-affective dynamics in AI eHMI: adaptive language strategy for emotion and attention modulation
In the interactions between human and AI agents, the language strategies of the agent play a crucial role in shaping user perception and overall user experience. While language strategies have been shown to influence users’ emotions in various application domains, limited research has examined their effects in driving contexts, particularly in interactions based on external human–machine interfaces (eHMIs). In practical traffic environments, eHMIs serve as a key communication channel through which autonomous vehicle agents convey information to other road users. This study introduces an eHMI language strategy framework based on text and emoji that enables autonomous agents to express distinct language styles and tonal variations. A user study (N = 20) was conducted to investigate how these language strategies influence drivers’ emotional responses, visual attention, and overall interaction experience. Participants operated a vehicle in a simulated driving cockpit, during which an autonomous vehicle using eHMIs with varying language strategies executed a lane-cutting maneuver. The results indicate that anthropomorphic language styles significantly reduce the negative emotions typically induced by cut-in events and improve both attentional focus and user acceptance of the interaction. However, in certain scenarios, an affinitive tone was less effective than a neutral tone, suggesting the tone should be contextualized. This study reveals the cognitive-affective dynamics triggered by language strategies in autonomous driving eHMIs and provides empirical evidence and design recommendations for the development of emotionally adaptive and semantically rich eHMI systems.