Advancements in Sensors and Analyses for Emotion Sensing

Wataru Sato

DOI: https://doi.org/10.3390/s24134166

IF: 3.9

2024-06-28

Sensors

Abstract:Exploring the objective signals associated with subjective emotional states has practical significance [...]

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation

What problem does this paper attempt to address?

The paper aims to explore the latest advancements in sensors and analytical methods in affective sensing technology to overcome the limitations of existing methods. Specifically, the paper focuses on the following issues:

1. **Limitations of Subjective Emotion Assessment**:
- The scientific rigor of self-reports is controversial.
- Subjective assessments may be biased by demand characteristics and social expectations.
- Continuous subjective assessment is difficult during dynamic tasks.
- Certain populations (e.g., individuals with alexithymia, dementia patients, and young children) find it challenging to assess emotional experiences.
- Emotional responses in the body and behavior may be triggered without subjective experience.

2. **Limitations of Existing Affective Sensing Methods**:
- There is controversy over using facial expressions or physiological signals to estimate emotion categories.
- The correlation between facial electromyography (EMG) activity and subjective pleasure experience is weak in some studies.
- Only a small portion of wearable affective sensing devices have been empirically validated.
- Affective analysis faces advanced challenges in multimodal and real-world data analysis.

3. **New Sensors and Analytical Methods**:
- Exploring new sensors and measurement methods, such as automatic facial action analysis via video, fingertip temperature measurement, etc.
- Testing the effectiveness of wearable devices in implementing affective sensing in real-world contexts, such as recording physiological or physical signals during shopping, collaborative tasks, driving fatigue detection, and children's storytelling activities.
- Proposing new analytical methods, such as multimodal information analysis, heart rate variability signal feature extraction, machine learning model optimization, etc., to improve the accuracy of emotion classification.

In summary, the paper attempts to address the issues in existing affective sensing technology by exploring new sensors and analytical methods, thereby more accurately estimating subjective emotional states and applying these technologies in more realistic scenarios.