Bayesian Methods Yield Improved Knowledge on the Generation of Brain Functional Networks

HIIT-funded research illuminates the role of conduction delays between brain regions in generating brain functional networks.

Brain functional networks are key to executing all cognitive functions such as speaking, listening, remembering, planning and decision-making. Conversely, disrupted brain functional networks are observed in neuro-degenerative disorders such as Alzheimer’s Disease. While the consolidated results of hundreds of experiments have yielded knowledge on the distinct sets of brain regions, i.e., the functional networks, operating during specific cognitive tasks, we do not yet know precisely how these networks are generated. Conduction delays mediate the transmission of electrical activity between brain regions. Yet, the role of inter-regional conduction delays in generating brain functional networks has remained a mystery. A key hurdle in investigating the role of inter-regional delays has been the absence of computational methods to compare biologically plausible models with different assumptions on the role of delays in generating brain networks.

State-of-the-art Approximate Bayesian Computation (ABC) methods offer tools to fit and compare models with experimental data. In this project, we applied recently proposed ABC methods to compare biologically plausible models with different delay assumptions using human Magnetoencephalography (MEG) data. Docent Nitin Williams, a HIIT-funded postdoctoral researcher, led the study, Prof. Samuel Kaski guided the application of ABC methods and Prof. Matias Palva provided neurophysiological input and a large MEG dataset to compare the models. Comparing the models revealed for the first time, that inter-regional conduction delays proportional to the distance between brain regions are likely to underlie the generation of the experimentally observed brain functional networks. These findings have been published in NeuroImage (

Our findings yield new knowledge on the role of inter-regional conduction delays in generating large-scale brain networks. In the future, the knowledge and biologically realistic models resulting from this project could illuminate the pathophysiology of neuro-psychiatric and neurological disorders. In particular, the models could be used to better understand impaired mechanisms in Multiple Sclerosis (MS), in which inter-regional conduction delays are raised due to dramatic structural changes in the physical connections between regions.

Contact person

Docent Nitin Williams
Department of Neuroscience & Biomedical Engineering
Aalto University, Finland


Williams N, Ojanperä A, Siebenhühner F, Toselli B, Palva S, Arnulfo G, Kaski S, Palva M (2023) “The influence of inter-regional delays in generating large-scale brain networks of phase synchronization” NeuroImage 279:120318 doi: