Adopting whole-brain computational modelling to investigate neurophysiological features associated with cognition

Understanding human cognition and intelligence requires integrating diverse psychological capacities and the complex biological interactions that underpin them. This work explores the potential of whole-brain computational modelling to illuminate these interactions and address the inherent challenges of studying the human brain. While traditional methods like lesion studies and non-invasive brain stimulation offer valuable insights, they are limited in scope and resolution. Computational modelling, by simulating brain activity, can provide a complementary approach, allowing for the prediction of system changes and the exploration of mechanistic relationships. We provide an introduction to whole-brain modelling, tracing its development from single-neuron models to complex simulations of neural populations and networks. The relevance of model selection based on research objectives is discussed, highlighting trade-offs between microscopic and mesoscopic scales. Furthermore, we detail the essential components of building a whole-brain model, including brain parcellation, structural connectivity, population dynamics models, parameter estimation, and software. Models of specific neurophysiological features relevant to cognition, such as neuromodulatory systems, the cerebrovascular system, and astrocytes, are then discussed. Examples of existing modelling studies are presented to illustrate how modelling techniques can be used to investigate how these features influence brain dynamics and cognitive processes. By providing an introductory foundation in whole-brain modelling, this paper aims to inspire more neuroscience researchers to integrate these powerful tools into their investigations of human cognition.