A Design Space of Virtual Bodies: Their Types, Effects, and Theoretical Foundations

Paper Title

A Design Space of Virtual Bodies: Their Types, Effects, and Theoretical Foundations

Publication Info

• Topic area: Virtual body manipulations in immersive virtual reality (VR)
• Keywords: Virtual bodies, VR, embodiment, design space, Proteus effect, multisensory integration, user experience, interaction, theory-building, HCI

Background and Problem

• Problem / challenge: Despite extensive research on virtual bodies, there is no comprehensive framework that maps what aspects of virtual bodies are manipulated, how these manipulations influence outcomes, and why these effects occur. Existing reviews are fragmented, focusing on specific manipulations, outcomes, or theories without crosscutting synthesis.
• Significance: A structured understanding of virtual body manipulations is critical for informed design, systematic experimentation, and theoretical advancement in VR applications, ranging from HCI to psychology and beyond.
• Motivation and related work: Prior work has cataloged specific manipulations (e.g., locomotion techniques, redirected hands) and synthesized outcomes (e.g., VR sickness, embodiment) or theories (e.g., presence, Proteus effect). However, these efforts lack integration, leaving gaps in understanding the relationships between manipulations, outcomes, and explanatory mechanisms.

Solution

• Proposed approach: The authors propose a design space of virtual bodies, categorizing manipulations into seven visual types and analyzing their effects and theoretical underpinnings.
• Novelty:
  1. A comprehensive design space with seven categories of visual manipulations: appearance, size, morphology, viewpoint, transfer, remapping, and control.
  2. A synthesis of effects across eight domains, including VR experiences, performance, and social outcomes.
  3. An analysis of theories explaining effects, such as embodiment, Proteus effect, and multisensory integration.
  4. Identification of gaps and underexplored areas in the design space and theoretical frameworks.
• Procedure and key techniques:
  • Systematic review of 208 studies from 2019–2023 using targeted searches in 159 venues.
  • Data extraction using Elicit, an LLM-based literature review tool, and iterative coding to identify key variables, effects, and theories.
  • Development of a multidimensional design space and mapping of studies onto it.
  • Analysis of connections between manipulations (what), effects (how), and theories (why).

Results

• Concrete findings:
  • Seven categories of visual manipulations: appearance (103 studies), size (25), morphology (32), viewpoint (41), transfer (30), remapping (10), and control (46).
  • Eight types of effects: VR experiences (71 studies), perception of self/environment (57), performance (38), behavioral outcomes (7), physiological responses (14), social outcomes (20), emotional responses (12), and technical evaluations (4).
  • Theories used include embodiment (107 studies), Proteus effect (25), presence (8), sensory conflict (3), peripersonal space (7), Bayesian models (2), and others.
• Advantage over baselines:
  • Provides a structured, multidimensional framework connecting manipulations, effects, and theories.
  • Highlights sparsely explored areas, such as many-to-one remapping and non-visual manipulations, and underutilized theories like Bayesian models.
• Experiments / evaluation:
  • Analysis based on a dataset of 208 studies, systematically selected and coded.
  • Effects and theories synthesized across diverse domains, including HCI, psychology, and neuroscience.
• Limitations and future work:
  • Limited engagement with non-visual manipulations and boundary conditions like multisensory interactions.
  • Lack of studies addressing societal behaviors (e.g., health, sustainability) through virtual body manipulations.
  • Future work should explore disembodiment, integrate mechanistic models, and expand manipulations to others’ virtual bodies.

Summary

This paper presents a comprehensive design space for virtual bodies in immersive VR, categorizing manipulations into seven visual types and synthesizing their effects across eight domains. It connects these manipulations to theoretical frameworks like embodiment and the Proteus effect, revealing gaps in both empirical and theoretical research. The findings provide actionable insights for designing and studying virtual bodies, while highlighting underexplored areas such as many-to-one mappings, non-visual modalities, and societal behavior change. The work offers a foundation for advancing VR research and applications in HCI, psychology, and beyond.