The human brain is arguably the most complex network that we know of, consisting of billions of nerve cells interconnected by trillions of axonal fibres. Interactions unfolding within this intricate web of connectivity form the basis of all our thoughts, emotions and behaviour, and their derailment can lead to mental illnesses such as schizophrenia, depression, and obsessive-compulsive disorder.

To unravel this extraordinary complexity, our research combines human neuroimaging with techniques from neuroscience, genetics, psychology, psychiatry, physics, and mathematics. We use these approaches to map and model the connectivity of the brain in order to understand the biological basis of behaviour in health and disease.

Our research program is built on the four key pillars detailed below. For student and employment opportunities, contact us!

Cognitive and clinical neuroscience

We are interested in understanding how network interactions in the brain give rise to individual differences in behaviour. Questions that we tackle include:

  • How does the network organization of the brain support complex, adaptive behaviour?
  • How do individual differences in brain network organization relate to variations in personality, cognition, and risk for mental illness?
  • How do brain networks develop?
  • How is brain network connectivity disrupted by psychiatric disease?
  • Can we develop an empirically grounded framework for diagnosing mental illness?
  • Can an understanding of brain network dysfunction in psychiatric disease be used to develop more effective treatments?

Mapping the effects of adolescent development on brain connectivity. This figure shows brain-wide maps of connections that either decrease (left) or increase (right) in connectivity strength between 16 and 18 years of age. Source: Baker et al. (2015) Journal of Neuroscience


We aim to understand how genes shape the organization and function of large-scale brain networks. We conduct investigations into:

  • Genetic contributions to individual differences in brain connectivity
  • The genetic variants that shape brain network organization
  • How patterns of gene expression relate to brain network properties
  • How brain connectivity mediates the effects of genes on individual differences in behaviour and risk for mental illness

This figure shows different approaches to integrating brain-wide gene expression atlas data with different brain network properties. These analyses offer a powerful method for identifying molecular correlates of different imaging phenotypes. Source: Fornito et al. (2019). Trends in Cognitive Sciences.

Method development

We develop and validate diverse methods for brain mapping. Specific emphases of this work include:

  • Developing new Magnetic Resonance Imaging (MRI) methods for mapping brain connectivity in living humans
  • Developing effective processing pipelines for denoising MRI data
  • Developing new statistical and graph theoretic approaches for quantifying and visualizing the network organization of the brain

Schematic workflow for implementing the Diffuse Cluster Estimation and Regression (DiCER) method for denoising functional MRI data. This method more effectively removes anatomically widespread signal deflections than competing approaches, and improves statistical power for mapping brain connectivity networks. Source: Aquino et al. (2020). NeuroImage.

Mathematical modelling

Mathematical models provide a means for testing precisely specified hypotheses about the brain. Our research focuses on models of (1) large-scale neural dynamics that enable simulations of whole-brain activity patterns; and (2) network wiring, allowing us to understand how brain networks develop. Research questions that guide this work include:

  • How does brain structure constrain brain function?
  • Can mathematical models be used to evaluate the potential cellular and molecular mechanisms explaining macroscopic brain changes observed with neuroimaging?
  • What are the wiring principles that govern how different parts of the brain connect to each other?

Figure fro Mathmatical Modelling

Modelling the dynamical consequences of lesions to different brain regions. Blue and red edges represent connections in which functional coupling between brain regions is predicted to either decrease or increase, respectively, following a lesion to a specific brain region (shown on left). The predictions were made using a large-scale model of whole-brain dynamics. Lesioning a putative hub area that connects to many other regions (top) has a much greater effect on brain function than lesioning a less-connected area (bottom). Source: Fornito et al. (2015) Nature Reviews Neuroscience