EXPLORA: A teacher-apprentice methodology for eliciting natural child-computer interactions

Document Title

EXPLORA: A teacher-apprentice methodology for eliciting natural child-computer interactions

Document Information

• Subject Area: Child-Computer Interaction (CCI) and its design and evaluation methodologies
• Keywords: methodology, child-computer interaction, means-ends analysis, contextual studies, multi-stage approach, online research design

Research Background and Issues

• Research Questions and Challenges:

  • As children increasingly interact with technological devices and applications, there is a need to study their interactions in natural environments to improve child-friendly technology design.
  • Obtaining context-rich information is particularly complex in research involving children, requiring trust-building while overcoming challenges such as insufficient interaction data due to time and spatial constraints.
  • Current context-rich research methods, such as situational surveys and natural observation, provide valuable insights but often face obstacles in translating findings into concrete design recommendations.

• Significance:

  • Understanding children's interactions with computer technologies in real-life environments contributes to developing technologies that align with their developmental needs and respect their context. This is especially critical for supporting learning and development processes.

• Research Motivation and Related Work:

  • Contextual studies and ethnography methods have demonstrated the potential for collecting behavioral data in natural environments, but a clear systematic framework is needed to organize observations into actionable design recommendations.
  • Inspired by the teacher-apprentice relationship, this study integrates the existing means-ends analysis framework to explore the complex relationships between cognition and context.

Solution

• Method Overview:

  • EXPLORA is a multimodal, multi-stage online methodology designed to study natural child-computer interactions, structured around three core steps:
    1. Establishing a teacher-apprentice relationship.
    2. Learning from child teachers.
    3. Evaluating and reinforcing the knowledge acquired by researcher apprentices.
  • The methodology incorporates attitudinal data collection (interviews) and behavioral data (natural observation) within the research design. It employs the means-ends analysis framework to dissect context, interaction patterns, and cognitive processes.

• Innovations:

  • Developed a method that positions children as expert participants in the research, granting them greater agency and expression.
  • Extended the means-ends analysis framework to the CCI domain, providing a systematic approach to analyze the complex relationships between context, interactions, and individual choices.
  • Introduced a multi-stage design to guide and observe natural interactions progressively, minimizing interference with observed behaviors.

• Implementation Steps:

  1. Initial Session: Introduce the research purpose and methodology to children and caregivers (dyads), establish background knowledge, and build trust.
  2. Interview Phase: Use non-directive questions to explore participants' technology usage habits (e.g., guided questions like "Can you teach me this?").
  3. Observation Phase: Conduct remote video observations of natural interactions in home environments, including screen recordings and online interactions.
  4. Event-Based Conversations: Invite children to recreate or explain recently completed tasks and related interaction processes.
  5. Reflective Interviews: Validate research hypotheses with participants, confirming alignment between findings and their experiences.
  6. Data Analysis: Apply deductive and inductive thematic analysis combined with the means-ends method to organize and derive interaction patterns and design recommendations.

Research Outcomes

• Key Findings:

  • EXPLORA enhanced participants' ability to express themselves through the teacher-apprentice relationship, particularly aiding children in recalling and recreating processes, which enriched the data collected.
  • The methodological sequence (interview-observation-event conversation) effectively captured critical contexts and behaviors in child-technology interactions, facilitating the generation of design recommendations.
  • The study identified common challenges faced by children during homework-related information searches (e.g., ineffective search terms, difficulty understanding issues) and summarized problem-solving strategies.

• Design Recommendations:

  • Proposed the concept of a "Homework Tutor" based on common scenarios, offering functionalities such as support for information retrieval, problem framing, and knowledge supplementation.
  • Promoted the development of children's autonomous problem-solving abilities and key search skills.

• Advantages and Uniqueness:

  • Compared to traditional ethnography, this method provides a more systematic process, combining behavioral and cognitive analyses to enhance data interpretability and practical applicability.
  • The means-ends framework effectively identified specific contextual constraints and their impacts.

• Limitations and Future Directions:

  • Sample limitations: Focused only on children aged 9-11 and their caregivers, with a small sample size (9 dyads).
  • The online approach restricted the capture of background activities; future iterations could integrate more on-site observations and offline interactions.
  • The method was successfully validated in the Brazilian context but requires further cross-cultural validation to assess its transferability and universality.
  • Future plans include introducing a participatory design phase, inviting children to co-design solutions, and incorporating anonymization and privacy protection technologies.

Conclusion

The EXPLORA methodology provides an innovative tool for capturing natural child-computer interaction behaviors, integrating attitudinal and behavioral data, implementing a multi-stage learning process, and deriving design recommendations through the means-ends framework. This study offers valuable insights into understanding children's interaction needs and cognitive challenges within specific contexts. The structured approach of EXPLORA demonstrates its broad application potential within the CCI domain.