From Following to Understanding: Investigating the Role of Reflective Prompts in AR-Guided Tasks to Promote User Understanding

Research Background and Issues

• What problems or challenges did the authors identify?
  The authors found that although augmented reality (AR) technology provides powerful tools for task guidance, it may lead users to mechanically follow steps without understanding the deeper principles of the task. This issue reduces the transferability of knowledge and may result in errors or over-reliance on the system.

• Why is this issue important?
  Deep understanding is key to internalizing task knowledge and transferring skills. Solely relying on mechanical step execution may make it difficult for users to adapt to new scenarios or task changes. This is particularly risky in tasks where safety is critical or complexity is high.

• Research motivation and related work
  The authors were inspired by the importance of reflective thinking in deep learning, as emphasized in educational and learning theories, and planned to integrate this mechanism into AR guidance. However, previous studies have mostly focused on traditional educational settings rather than task guidance contexts, and have not specifically examined whether reflection can promote deeper understanding in AR task guidance.

Solution

• What methods or solutions did the authors propose?
  The authors proposed integrating reflective prompts into AR-guided tasks. These prompts include three main types:

  1. Challenging Assumptions: Prompting users to question the necessity or method of the current step.
  2. Connection to Outcomes: Helping users link specific actions to the final outcomes.
  3. Hypothetical Scenarios: Encouraging users to consider how the task could be completed in different contexts.

• What is innovative about this solution?
  It embeds reflective elements into AR task guidance systems, aiming to balance the need for task completion with deep understanding. This approach seeks to break the mechanical execution pattern by forcing users to pause and reflect on the meaning and consequences of task steps.

• What are the implementation steps and key technologies used?

  1. Formative Research: Deriving reflection types from educational theories and optimizing these reflective prompts through a co-design process with 9 participants.
  2. AR System Prototype Development: Developing a prototype system using Apple Vision Pro, featuring instructional text, clickable keyword information retrieval, and automatic reflective prompt display.
  3. Evaluation Study: Conducting a comparative experiment (N=16) to test the impact of reflective prompts on task performance and deep understanding.

Research Outcomes

• What specific outcomes were achieved?

  1. Tasks with reflective prompts resulted in a 0.66 standard deviation increase in quiz scores, significantly improving participants' task understanding (p < .05).
  2. Task-related information retrieval behaviors increased by 68.75% under the reflective prompt condition (from 0.32 to 0.54, p < .01), demonstrating its role in stimulating cognitive curiosity.
  3. Most participants (15/16) found the reflective prompts non-intrusive and easy to ignore during task completion.

• What advantages does this solution have compared to existing ones?
  Unlike standard AR step-by-step guidance, this method encourages users to actively think about the logic behind tasks by adding reflective prompts. It reduces mechanical reliance on the system without significantly increasing cognitive load.

• What were the experimental or evaluation results?

  1. Task Success Rate: The task success rate under the reflective prompt condition was 94.8%, significantly higher than the 68.8% without reflective prompts.
  2. Subjective Feedback:
     • 11/16 participants felt that reflective prompts increased their motivation to understand the principles behind the task.
     • 9/16 participants believed the prompts enhanced their interest in the task.
     • Reflective prompts did not significantly increase cognitive load or reduce system usability.

• Limitations and future directions

  1. Limitations:
     • The scope of tasks was limited to coffee-making and circuit assembly, which may not generalize to all task types.
     • Participants only used the AR system once, so long-term behavior patterns were not observed.
     • Reflective prompts might reduce users' subjective task self-efficacy.
  2. Future Directions:
     • Expanding the types and application scenarios of reflective prompts, such as tasks involving high risk or requiring team collaboration.
     • Introducing long-term tracking studies to evaluate the long-term effects on knowledge transfer and skill development.
     • Combining reflective prompts with automated error feedback mechanisms to provide more precise and context-sensitive guidance.

This study presents an innovative approach to addressing the issues of over-reliance on AR guidance and lack of deep understanding by integrating reflective prompts. It offers valuable directional insights for the design of future AR guidance systems.