Effects of Emotion and Agency on Presence in Virtual Reality

Title of the Paper

Effects of Emotion and Agency on Presence in Virtual Reality

Paper Information

• Research Domain: Virtual Reality (VR), Emotion, Agency, and their impact on Presence
• Keywords: Virtual Reality, Presence, Emotion, Agency, Emotional Intensity, Emotional Valence, Structural Equation Modeling, User Experience

Research Background and Problem Statement

• Identified Issues or Challenges:

  • Virtual reality technology, known for its significantly higher presence compared to traditional media, has been widely applied in psychological research, behavioral therapy, training, and gaming. However, findings on how human factors (such as emotion and agency) influence presence remain inconsistent.
  • The impact of emotion on presence is inconclusive: some studies suggest that different emotions trigger varying levels of presence, while others argue that the type of emotion (positive/negative) does not significantly affect presence.
  • The relationship between agency and presence is similarly unclear, and the potential interaction between agency and emotion has not been sufficiently explored.

• Importance of the Research:

  • Presence is a critical factor for the success of virtual reality applications. Understanding the impact of emotion and agency on presence can help optimize virtual environment design and enhance user experience.

• Research Motivation and Related Work:

  • Investigating two dimensions of emotion: valence (positive or negative) and intensity (degree of emotional experience), to understand their effects on presence.
  • Exploring how agency influences presence independently or interacts with emotion.
  • Proposing a unified model to explain the combined effects of emotion and agency on presence, providing guidance for designers to optimize virtual environments.

Solution

• Methods or Solutions:

  • Designing virtual environments (VEs) that elicit "happiness" and "fear" emotions while controlling for the presence or absence of user agency, forming a 2×2 experimental design.
  • Collecting subjective data on emotion, agency, and presence from 121 participants and constructing a Structural Equation Model (SEM) to analyze the direct and interactive effects of emotion and agency on presence.

• Innovative Contributions:

  • Employing a unified experimental design to simultaneously study the effects of emotion (valence, intensity) and agency on presence for the first time.
  • Developing the Presence, Emotion and Agency (PEA) model, which separates virtual environment design factors from user experience, providing a unified explanation of the mechanisms by which emotion and agency affect presence.
  • Proposing a layered modeling approach that considers design factors (EmotionVE, AgencyVE) and subjective user experiences (Intensity, Agency).

• Implementation Steps and Key Techniques:

  1. Virtual Environment Design:
     • Construct dynamic virtual scenarios evoking "happiness" and "fear," with conditions for user interaction (interactive/non-interactive).
     • Enhance fear through adjustments in lighting, sound effects, and nighttime visuals; increase happiness through interactions with friendly dogs.
  2. Measurement Tools:
     • Use the "User Experience Questionnaire" and a 10-point Likert scale to assess participants' emotional intensity, agency, and presence.
  3. Experimental Procedure:
     • Randomly assign participants to one of four scenarios: happiness-interactive, happiness-non-interactive, fear-interactive, fear-non-interactive.
     • Participants complete questionnaires immediately after the experiment for subjective evaluation.
  4. Data Analysis:
     • Employ ANOVA and Pearson correlation analysis to evaluate the effects of emotion and agency on presence.
     • Construct a Structural Equation Model (SEM) based on experimental data to validate model fit and quantify direct/indirect effects.

Research Outcomes

• Specific Findings:

  1. Dominant emotions elicited in virtual environments positively correlate with presence, with emotional intensity playing a significant role.
  2. Agency significantly and positively influences presence and moderates the impact of emotion on presence.
  3. When virtual environments induce fear, providing agency significantly enhances presence; however, agency does not significantly increase presence in happiness-inducing scenarios.
  4. The PEA model quantifies the direct and indirect effects of emotion and agency, revealing the complex relationship between virtual environment design factors and user experience.

• Advantages Compared to Existing Solutions:

  • Introduced a systematic and structured analytical framework (PEA model) capable of separating and explaining the effects of design factors and subjective perception on presence.
  • Clarified the specific interaction patterns between emotional intensity and agency under different emotional valence conditions.

• Experimental or Evaluation Results:

  • SEM model validation demonstrated good model fit (RMSEA < 0.01, Comparative Fit Index > 0.97).
  • The standardized direct effect of Agency on Presence (0.69) was significantly higher than that of Intensity (0.20), indicating the dominant role of agency.

• Limitations and Future Directions:

  • The use of non-human characters may limit the impact of social emotions on presence; further research on human avatars is needed.
  • The experiment did not address the potential effects of non-human factors such as virtual graphics resolution and sound quality on presence; future studies should comprehensively control external variables.
  • This study primarily focused on emotional intensity rather than physiological arousal (e.g., skin conductance or heart rate); incorporating real-time physiological indicators may deepen the understanding of the emotion-presence relationship.