Uncovering the differences and similarities between physical and virtual mobility.

Area of Science:

• Computational Social Science
• Network Science
• Human Mobility Studies

Background:

• Digital traces from information and communications technologies offer high-resolution insights into human movement.
• Existing mobility models explain empirical spatio-temporal trends but often focus solely on physical movement.
• The rise of the World Wide Web introduced virtual mobility, paralleling and sometimes replacing physical activities.

Purpose of the Study:

• To systematically analyze and compare the statistical patterns of physical and virtual human movement.
• To identify similarities and differences in the temporal and spatial regimes of these two mobility types.
• To propose a unified model explaining the common underlying mechanisms of human navigation.

Main Methods:

• Comparative analysis of empirical spatio-temporal data from physical and virtual movement.
• Statistical pattern identification in both temporal and time-independent (event sequence) spaces.
• Development and validation of a Metropolis-Hastings type optimization model for movement behavior.

Main Results:

• Physical and virtual mobility exhibit distinct temporal patterns due to spatial/economic constraints in physical movement.
• When analyzed as event sequences (time-independent), statistical patterns of physical and virtual mobility become identical.
• The proposed optimization model successfully reproduces observed statistical commonalities with minimal input.

Conclusions:

• A fundamental similarity exists in the statistical patterns of physical and virtual human mobility, particularly in event sequences.
• Human movement, whether physical or virtual, appears governed by a common decision-making process akin to cost-benefit analysis.
• The developed optimization model offers a parsimonious explanation for observed mobility regularities across different domains.