Harmonizing the mind: EEG-driven neuroadaptive visual stimuli control towards emotional arousal regulation

• Main Author: Lee, Karmen Clariss
• Other Authors: Guan Cuntai
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Machine learning; EEG; Neuroscience; Psychology; Emotional regulation
• Online Access: https://hdl.handle.net/10356/174936
• Institution: Nanyang Technological University

In recent years, the mental health crisis has become an increasingly pressing issue. This has created an important need for user-friendly, innovative, and personalised interventions. Notably, there is a bi-directional relationship between emotion and cognition which is imperative to emotional regulation. However, research into technologies that leverage this emotion-cognition interplay is lacking. Therefore, this study proposes an EEG-driven Neuroadaptive Emotion Arousal Regulation (NEAR) system, aimed at leveraging this relationship with EEG and data analytics to enhance emotional regulation. The aim is to accomplish this through individualised visual modulations provided based on reinforcement learning feedback from the multimodal sensors to elicit desirable emotions and control through adaptive psychotherapeutic interventions. Furthermore, this study adopts a personalised strategy, whereby besides personalised neuroadaptive visual modulations, the visual stimuli selected were also personalised based on their individual cognitive and emotional states. Multimodal sensor data was collected from a total of 20 healthy participants in a controlled experiment setting to evaluate the efficacy of a simulated version of the proposed neuroadaptive system. The findings indicate that the participants improved their emotion regulation capabilities after the Emotional Arousal Regulation Task. Moreover, the personalised stimuli were also effective in eliciting the desired emotional arousal levels. Overall, while there were several limiting factors, the proposed solution of a personalised EEG-driven neuroadaptive system exhibits potential as a method to enhance emotional regulation.