Lecture: 'Beyond Discrete Brain States: Rethinking Functional'

Understanding how large-scale brain networks dynamically organize over time is a central goal of contemporary neuroscience, yet many widely used analytic frameworks rely on simplifying assumptions about how functional brain states are defined and evolve. This talk will discuss how recent advances in functional connectivity and dynamic modeling motivate a shift from discrete “brain state” descriptions toward superposition-based accounts of brain dynamics.

This talk will begin with an overview of core concepts in functional neuroimaging and connectomics, with an emphasis on how functional connectivity is commonly estimated and interpreted from functional Magnetic Resonance Imaging (fMRI) data. I will then provide a high-level overview of some of my own methodological contributions to the analysis of functional connectivity, including the motivation, design principles, and dissemination of these methods through the CONN toolbox.

In the second half of the talk, I will focus on some of my most recent work on dynamic functional connectivity, and in particular on a new framework based on dynamic independent component analysis (dynamic ICA). This work challenges the notion of discrete, mutually exclusive “brain” or “mind” states, and instead argues for a superposition-based view, in which multiple concurrent functional processes jointly shape observed brain dynamics. I will discuss how this perspective better captures the richness and flexibility of brain organization across different tasks and contexts, and its implications for how we conceptualize mental states in neuroscience research.

About the speaker

Alfonso Nieto-Castañón is a computational neuroscientist and Director of the Computational Neuroscience Research Lab (CNRLab). He is an Invited Research Scientist in the Department of Speech, Language, & Hearing Sciences, Boston University (BU), and a Research Affiliate at the McGovern Institute for Brain Research, Massachusetts Institute of Technology (MIT). His research focuses on functional neuroimaging, brain connectivity, and statistical modeling of brain dynamics. He has authored over 80 peer-reviewed publications, and his work has had a sustained and broad influence on neuroimaging research (16,867 citations overall; 2,545 citations in 2025 alone).

He is also the lead developer of the CONN functional connectivity toolbox (CONN toolbox), a widely used software platform for the analysis of resting-state and task-based fMRI data. His methodological work spans multiple aspects of functional connectivity analysis, from data denoising and quality control to multivariate inference at the connectome level and dynamic connectivity methods. His current work focuses on time-varying functional connectivity methods, including dynamic ICA approaches for characterizing brain network dynamics across tasks and contexts.