Effects of Individuality, Education, and Image on Visual Attention: Analyzing Eye-tracking Data using Machine...

Area of Science:

• Cognitive Science
• Computer Science
• Architectural Psychology

Background:

• Machine learning classification algorithms build predictive models from labeled data.
• Eye-tracking studies analyze visual attention patterns during scene observation.
• Understanding how individual factors influence perception of architectural scenes is crucial.

Purpose of the Study:

• To investigate the impact of individuality, education, and image stimuli on the observation of architectural scenes using machine learning.
• To identify distinguishing eye-tracking patterns and parameters related to these factors.
• To uncover novel insights not previously reported in the literature.

Main Methods:

• Utilized machine learning classification algorithms to analyze multi-dimensional eye-tracking data.
• Investigated patterns related to individuality, educational background (architecture vs. other disciplines), and image stimuli.
• Employed velocity histograms and endogenous parameters for analysis.

Main Results:

• A unique velocity histogram was identified for each individual, highlighting personal viewing patterns.
• Students of architecture and other disciplines were distinguishable based on endogenous eye-tracking parameters.
• Distinct differences emerged in the focus on structural versus symbolic elements between student groups.

Conclusions:

• Machine learning effectively identified novel eye-tracking parameters and patterns in architectural scene observation.
• Individual differences and educational backgrounds significantly influence visual attention strategies.
• The study provides new insights into the cognitive processes underlying the perception of architectural environments.