Dimensionality reduction techniques in pupillometry research: A primer for behavioral scientists.

Area of Science:

• Psychophysiology
• Eye-tracking technology
• Data analysis

Background:

• Pupil size measurement is a classic psychophysiological tool.
• Recent advancements in eye-tracking have increased its popularity.
• Existing analytical methods emphasize the temporal aspect of pupillary responses.

Purpose of the Study:

• To propose a novel analytical approach for pupillary data.
• To suggest that significant pupil size changes are late indicators of underlying physiological processes.
• To introduce dimensionality reduction techniques for analyzing these earlier processes.

Main Methods:

• Utilizing temporal principal components analysis (PCA) and related rotations.
• Applying dimensionality reduction to uncover a latent 'pupillary manifold'.
• Demonstrating the utility of the R library 'Pupilla'.

Main Results:

• Dimensionality reduction reveals a low-dimensional 'pupillary manifold' representing core physiological processes.
• This manifold captures the underlying physiology more effectively than direct pupil size measurements.
• The approach provides scores that are statistically manageable and less arbitrary.

Conclusions:

• Analyzing the pupillary manifold offers an alternative and appealing solution for pupillary data analysis.
• This method accounts for physiological processes preceding observable pupil size changes.
• The R library 'Pupilla' facilitates this advanced analytical approach.