An Odyssey with complexity and network science: From the brain to social organisation
Author(s)
Expert, Paul
Type
Thesis
Abstract
Complexity science is the study of systems that give rise to a priori unexpected
macroscopic patterns, at different scales, emerging from the simple
microscopic rules governing the evolution of the system. Method to explore
complex systems are based on tools from a wide range of sciences, including
statistical mechanics. Recently, the study of the emerging properties
of complex systems has been enriched by a new toolbox derived by an extension
of graph theory, namely, complex network science. We use both
approaches to investigate two complex systems, the human brain and the
communications patterns of a social network.
The brain could be considered as an archetypical complex system; its
fundamental constituting units, the neurones, communicate via simple inhibitory
and excitatory interactions. These give rise to an extraordinarily
rich hierarchical complex system, the human mind, that enables one to
apprehend and interact with the universe.
Understanding how the brain functions is essential, both for the general
knowledge of humanity, but also for medical purposes; a better understanding
of brain functions could lead to finding cures. We investigate the dynamics
of brain activity at the smallest scale accessible by functional Magnetic
Resonance Imaging, the voxel, while subjects are at the resting-state. We
apply real-space renormalisation from the statistical mechanics toolbox, and
our findings confirm that brain dynamics displays characteristic signatures
of a critical system.
At a coarser level, we study the structural differences in the functional
networks of a healthy cohort and one made of people at-risk of developing a
mental disorder during a verbal fluency task. We find that a key brain region
plays a different role in the network organisation of the two populations,
which is in agreement with previous findings on the disease schizophrenia.
Finally, we investigate community structure in complex systems. Social
interactions in humans are also a prime example of a system with emerging
structure, the nature of which is dependent on the types of interactions
between individuals. We use and develop new methods for community
detection to uncover structures due to spatial and linguistic interactions in
a mobile phone network.
macroscopic patterns, at different scales, emerging from the simple
microscopic rules governing the evolution of the system. Method to explore
complex systems are based on tools from a wide range of sciences, including
statistical mechanics. Recently, the study of the emerging properties
of complex systems has been enriched by a new toolbox derived by an extension
of graph theory, namely, complex network science. We use both
approaches to investigate two complex systems, the human brain and the
communications patterns of a social network.
The brain could be considered as an archetypical complex system; its
fundamental constituting units, the neurones, communicate via simple inhibitory
and excitatory interactions. These give rise to an extraordinarily
rich hierarchical complex system, the human mind, that enables one to
apprehend and interact with the universe.
Understanding how the brain functions is essential, both for the general
knowledge of humanity, but also for medical purposes; a better understanding
of brain functions could lead to finding cures. We investigate the dynamics
of brain activity at the smallest scale accessible by functional Magnetic
Resonance Imaging, the voxel, while subjects are at the resting-state. We
apply real-space renormalisation from the statistical mechanics toolbox, and
our findings confirm that brain dynamics displays characteristic signatures
of a critical system.
At a coarser level, we study the structural differences in the functional
networks of a healthy cohort and one made of people at-risk of developing a
mental disorder during a verbal fluency task. We find that a key brain region
plays a different role in the network organisation of the two populations,
which is in agreement with previous findings on the disease schizophrenia.
Finally, we investigate community structure in complex systems. Social
interactions in humans are also a prime example of a system with emerging
structure, the nature of which is dependent on the types of interactions
between individuals. We use and develop new methods for community
detection to uncover structures due to spatial and linguistic interactions in
a mobile phone network.
Date Issued
2011-12
Date Awarded
2012-06
Copyright Statement
Attribution NoDerivatives 4.0 International Licence (CC BY-ND)
Advisor
Christensen, Kim
Jensen, Henrik
Publisher Department
Physics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)