Soma-noti: Delivering Notifications Through Under Clothing Wearables

Title of the Paper

Soma-noti: Delivering Notifications Through Under-clothing Wearables

Paper Information

• Domain: Wearable Technology, Haptic Interaction
• Keywords: Notification Mechanism, Wearable Devices, Haptic Output, Body Perception, Feedback Mechanism, Error Rate, Comfort, Response Time, Wearable Interaction, Location Awareness
• Conference: CHI 2021
• Link: DOI

Research Background and Problem Statement

• Identified Problems or Challenges:

  • Current wearable devices are predominantly designed as wristbands or rings, while the area on the inner side of clothing, which is close to the skin, remains underexplored as an output channel.
  • Vibration is the dominant notification mechanism, while other haptic signals (e.g., pinching, blowing) have not been sufficiently studied.
  • There is a lack of research comparing the performance of various haptic signals across different body locations.

• Significance of the Research:

  • Placing devices on the inner side of clothing eliminates the need for direct skin attachment, potentially enabling the design of non-invasive, privacy-friendly notification mechanisms.
  • Diversifying haptic notification forms can mitigate the limitations of single vibration signals (e.g., being easily ignored or distracting users).

• Motivation and Related Work:

  • Commercial applications of similar devices (e.g., LumoLift) inspired the authors, but these devices lack in-depth studies on perception effects and notification types.
  • Previous research has mostly focused on single haptic signals (e.g., heat or vibration) or specific locations (e.g., wrist), neglecting comparisons across different locations and signal types.

Proposed Solution

• Proposed Approach:

  • Design a prototype device (referred to as "badge") supporting 10 different haptic signals, including pressing, poking, pinching, heat, cold, blowing, suction, vibration, moisture, and brushing sensations.
  • Conduct experimental tests on six common upper-body locations covered by clothing: collarbone, shoulder, stomach, side of the body, upper arm, and lower back.

• Innovations:

  • The first study to compare 10 types of haptic signals and their performance differences across multiple body locations.
  • Introduced the "badge" as a novel device form and explored its design potential for placement on the inner side of clothing near the skin.

• Implementation Steps and Key Technologies:

  • Hardware Design:
    • Each signal type was implemented using actuators with different functionalities, such as modules driven by pressure, motors, or temperature differences to achieve various haptic effects.
    • Devices were secured to the inner side of clothing using magnets, ensuring skin contact and ease of wear and removal.
  • Experimental Design:
    • WiFi-controlled ESP32 microcontrollers were used to power and operate the signal devices.
    • Measured user response time, error rate, comfort, and urgency across different signals and locations.
  • Software:
    • Developed two Android applications: one for experimenters to control signals and another for users to record response times by pressing a button after perceiving a signal.

Research Results

• Specific Findings:

  • Response Time:
    • "Poking," "pinching," and "vibration" signals had the shortest response times, averaging around 2 seconds.
    • Temperature-related signals (e.g., heat/cold) had the longest response times, at 3.1 seconds and 3.57 seconds, respectively.
  • Error Rate:
    • The collarbone, shoulder, and upper arm locations had the lowest error rates (< 8%), indicating better signal perception due to tighter clothing fit.
    • The lower back had the highest error rate (15%), likely due to loose clothing affecting signal transmission.
  • Comfort:
    • Cold signals (4 points) and vibration signals (3.85 points) were rated as the most comfortable by users.
    • The upper arm location received the highest comfort score, indicating a better user experience for signals at this location.
  • Urgency:
    • The poking signal (4.0 points) was perceived as the most urgent, while the blowing signal (2.08 points) was considered the least urgent.

• Advantages Compared to Existing Solutions:

  • Provided a systematic comparison of multiple signals, offering valuable references for future notification design.
  • Highlighted the potential and applicability of "inner clothing" as a novel location for wearable devices.

• Limitations and Future Directions:

  • Limitations:
    • Individual factors such as clothing tightness were not fully standardized, which might have influenced the results.
    • The prototype device could be further miniaturized, as the current design slightly affects comfort.
  • Future Directions:
    • Conduct in-depth studies on individual signals, such as optimizing texture or stimulation patterns.
    • Compare the output effects of different device forms (e.g., rings, pendants, armbands).
    • Explore the combination and temporal optimization of haptic signals.

Conclusion

This study systematically explored the design and performance of under-clothing wearable notification devices by designing and testing a "badge" capable of delivering 10 different haptic signals. The results demonstrated significant impacts of body location and device type on user experience, providing valuable guidance and frameworks for future research on haptic notifications and wearable technologies.