A Probabilistic Approach to Understanding User Preferences for Adaptive Placement of AR Interfaces in Different...

Area of Science:

• Human-Computer Interaction
• Computer Graphics
• Cognitive Science

Background:

• Adaptive interfaces in augmented reality (AR) require understanding user preferences for optimal placement.
• Previous research has explored factors influencing user interaction with virtual elements in physical spaces.

Purpose of the Study:

• To develop a probabilistic model for predicting user preferences in adaptive AR interface placement.
• To identify key factors influencing user choices in AR layout adaptation.
• To provide design guidelines for adaptive AR systems.

Main Methods:

• Conducted two online crowdsourcing studies using simulated desktop and virtual reality (VR) environments.
• Assessed user-created layouts and analyzed subjective feedback to identify preference factors.
• Developed and validated a probabilistic optimization system for generating adapted AR interfaces.

Main Results:

• User preference for AR interface placement is significantly influenced by overall layout structure (shape, distribution) rather than environmental context.
• Quantified posterior probabilities for weighting coefficients in an optimization utility function based on user ratings.
• Established that layout consistency and structure are more critical than semantic associations with physical objects.

Conclusions:

• A statistical model with probabilistic distributions of factors can represent user preferences for adaptive AR interface placement.
• Concrete guidelines for adaptive AR systems emphasize layout consistency, structure, and the relationship between virtual widgets and physical objects.
• Findings inform the design of more user-centered and effective adaptive AR experiences.