Designing for Spatial Computing: Exploring Natural and Traditional Interfaces in a VR Moodboarding Tool

This study investigates the effectiveness of multimodal interaction techniques within a spatial computing context, applied to a moodboarding system based on AI-generated images. Three interaction modes are tested: a traditional interface relying on virtual buttons; a gesture-driven configuration in which hand tracking is employed for specific key operations, including triggering image generation, initiating speech recognition, and navigating the user interface, and a hybrid interface combining both modalities. In addition to evaluating task accuracy, speed, and user engagement, the study explores user behavior within the hybrid setup to identify which interaction method users prefer for each type of task, providing insights into the most congenial input strategies in creative spatial workflows.