Paper Number

ICIS2025-1924

Paper Type

Complete

Abstract

This study investigates how voice tone and visual instructions in augmented reality (AR) influence AI-guided diagnosis. Drawing on the instructional design literature and dual-coding theory, two experiments examined the effects of dominant versus submissive-toned AI verbalizations, and schematic versus structural AR. The results revealed that dominance-toned AI instructions generally improved diagnostic completion, particularly when paired with structural AR. Experiment 2 identified cognitive effort and confused emotions as mediating mechanisms, with user attitudes toward AI moderating these effects. This study contributes to HCI and IS by highlighting the importance of processing consistency across verbal and visual modalities and by integrating physical and digital interaction design. It offers novel insights into AR design and multimodal communication, emphasizing the cognitive-emotional dynamics of AI-guided diagnosis.

Comments

08-ImmersiveTech

Share

COinS
 
Dec 14th, 12:00 AM

Command or Collaborate in Augmented Reality? Investigating Voice Tone and Visual Instruction in Augmented Reality Human-AI Diagnosis

This study investigates how voice tone and visual instructions in augmented reality (AR) influence AI-guided diagnosis. Drawing on the instructional design literature and dual-coding theory, two experiments examined the effects of dominant versus submissive-toned AI verbalizations, and schematic versus structural AR. The results revealed that dominance-toned AI instructions generally improved diagnostic completion, particularly when paired with structural AR. Experiment 2 identified cognitive effort and confused emotions as mediating mechanisms, with user attitudes toward AI moderating these effects. This study contributes to HCI and IS by highlighting the importance of processing consistency across verbal and visual modalities and by integrating physical and digital interaction design. It offers novel insights into AR design and multimodal communication, emphasizing the cognitive-emotional dynamics of AI-guided diagnosis.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.