Creepy Technology: What Is It and How Do You Measure It?

Document Title

"Creepy Technology: What Is It and How Do You Measure It?"

Document Information

• Subject Area: Human-Computer Interaction (HCI), focusing on first-time user perceptions and quantification of "creepiness."
• Keywords: creepiness, user experience, HCI, assessment scale, first impressions, questionnaire, technology acceptance model, ethics

Research Background and Issues

• Issues and Challenges:

  • As interactive technologies increasingly integrate into users' daily lives, these technologies (e.g., wearable devices, home voice assistants) not only bring convenience but also provoke psychological discomfort—often described as a sense of "creepiness."
  • HCI research needs to address users' intuitive emotional responses during initial exposure to technology to avoid negative first impressions.
  • How to quantify this ambiguous, intuitive "creepiness" and use it as a basis for design remains underexplored.

• Significance:

  • Eliminating the "creepiness" experienced during initial technology use can enhance user acceptance and experience.
  • Systematically addressing this phenomenon can help designers develop technologies that are non-threatening or emotionally neutral.

• Research Motivation and Related Work:

  • In previous studies, "creepiness" in privacy and robotics research was often associated with "privacy invasion" or the "uncanny valley" effect in robots, which limited its scope.
  • Literature reviews and preliminary studies reveal that "creepiness" spans broader interaction scenarios (e.g., user interaction with autonomous vehicles, augmented reality devices) and is related to uncertainty, design intent, and social adaptability.
  • Currently, there is no unified theoretical model or standardized quantitative tool to measure "creepiness"; existing research methods are fragmented and lack systematicity.

Solution

• Methods and Framework:

  • Proposed the "Perceived Creepiness of Technology Scale" (PCTS).
  • The development process includes the following steps:
    1. Literature Review: Reviewing existing materials on the concept of "creepiness" in HCI, privacy research, and human-machine interaction.
    2. Focus Group Studies: Discussing sources of "creepiness" with users and collecting feedback on their experiences with different technologies.
    3. Conceptual Modeling: Developing a theoretical model of "creepiness" based on qualitative data from literature and focus groups, including three core dimensions.
    4. Questionnaire Development and Optimization: Drafting initial questionnaire items and refining them through expert feedback and statistical analysis.
    5. Scale Validation: Conducting exploratory factor analysis, confirmatory factor analysis, and reliability and validity testing on the scale.

• Innovations:

  • Systematically defined and validated the core dimensions of "creepiness" for the first time in HCI research.
  • Developed a standardized scale adaptable to diverse technological scenarios, enabling quick comparisons of "creepiness" levels across technologies.

• Model Dimensions:

  • Implied Malice: Users perceive the technology design as potentially violating ethical principles or personal values.
  • Undesirability: The technology appears intrusive or inappropriate in specific contexts.
  • Unpredictability: Users cannot fully understand the technology's purpose, functionality, or expected behavior.

Research Outcomes

• Specific Results:

  • Successfully designed a questionnaire based on a 7-point Likert scale, including three dimensions (8 items, total score range 9-63).
  • Validated the scale's internal consistency (Cronbach’s α=0.74) and validity through qualitative data and statistical analysis.

• Advantages Over Existing Solutions:

  • Unlike social acceptance or technology acceptance models, PCTS focuses on users' emotional first impressions of technology rather than functionality or social adaptability.
  • The scale enables early identification of design issues causing "creepiness," avoiding high costs of later adjustments.

• Experiments and Evaluation Results:

  • When comparing different interactive technologies (e.g., traditional keyboards vs. EEG login methods), PCTS effectively distinguished the "creepiness" differences between technologies.
  • Validation experiments (repeated measurements of known "creepy" items) demonstrated the scale's temporal stability (test-retest reliability κ=0.82).

• Limitations and Future Directions:

  • Limitations:
    • The scale currently applies only to measuring "creepiness" during initial technology experiences; changes in "creepiness" over prolonged use remain unclear.
    • Participants predominantly come from European or Western backgrounds, and the scale's applicability in multicultural contexts requires further testing.
  • Future Directions:
    • Extend PCTS to measure changes in "creepiness" during long-term technology use.
    • Conduct cross-cultural validation in different languages and cultural contexts.

This study lays a foundation for exploring new unified quantitative standards in the field of user experience, particularly guiding the design of technologies involving automation, personalization, or direct human interaction.