Supervisor: Laura Koesten
This project focuses on improving the Self-Assessment Manikin (SAM) by integrating nuanced methods to capture emotional dimensions such as dominance, arousal, and valence. We aim to design a new representation of an emotion scale using:
- Glyph-based representations: Exploring visual ways to represent complex emotional combinations.
- Haptic feedback: Incorporating force-feedback to enable tactile interaction and support emotional expression.
- Potentially: Physiological signals / Biometrics like heart rate or skin conductance
You will explore decision trees as a method to guide users through emotional assessment, emphasizing intuitive, explainable AI-supported systems. The goal is to measure subjective reactions and use these as inputs to adapt system interactions.
Key challenges include:
- Designing representations for complex emotional blends; addressing difficulties in defining and articulating emotions;
- Developing multimodal metrics to enhance user interaction and system adaptability.
Background literature: Margaret M Bradley and Peter J Lang. Measuring emotion: the self-assessment manikin and the semantic differential. Journal of behavior therapy and experimental psychiatry, 25(1):49–59, 1994.
Contact: Laura Koesten