NeuroPlace: Categorizing urban places according to mental states

Area of Science:

• Neuroscience
• Urban Computing
• Human-Computer Interaction

Background:

• Urban environments significantly influence human emotion and behavior.
• Understanding brain responses to diverse urban spaces is crucial for well-being.
• Affordable Electroencephalogram (EEG) technology enables novel research into in-situ mental state monitoring.

Purpose of the Study:

• To develop a novel urban place recommendation system using in-situ EEG data.
• To investigate how urban environments affect brain activity and mental states.
• To create an association map for recommending therapeutic urban spaces.

Main Methods:

• Utilized mobile EEG headsets to record brain activity in various urban settings.
• Analyzed and modeled EEG data to identify mental state signatures associated with different places.
• Developed a classifier to categorize urban locations based on user brain activity patterns.

Main Results:

• Successfully classified three distinct urban place types based on user mental state signatures.
• Achieved a classification accuracy of 90.8% for the mental states classifier.
• Created an association map to guide users toward places that reduce brain fatigue and promote rejuvenation.

Conclusions:

• The NeuroPlace system offers a novel approach to ubiquitous brain monitoring in urban computing.
• The findings provide insights for urban planners on the impact of urban design on mental well-being.
• Further research is ongoing to enhance the prototype's practicality, robustness, and reliability.