Neuro-cognitive multilevel causal modeling: A framework that bridges the explanatory gap between neuronal activity and cognition

Moritz Grosse-Wentrup,Akshey Kumar,Anja Meunier,Manuel Zimmer
DOI: https://doi.org/10.1371/journal.pcbi.1012674
2024-12-17
PLoS Computational Biology
Abstract:Explaining how neuronal activity gives rise to cognition arguably remains the most significant challenge in cognitive neuroscience. We introduce neuro-cognitive multilevel causal modeling (NC-MCM), a framework that bridges the explanatory gap between neuronal activity and cognition by construing cognitive states as (behaviorally and dynamically) causally consistent abstractions of neuronal states. Multilevel causal modeling allows us to interchangeably reason about the neuronal- and cognitive causes of behavior while maintaining a physicalist (in contrast to a strong dualist) position. We introduce an algorithm for learning cognitive-level causal models from neuronal activation patterns and demonstrate its ability to learn cognitive states of the nematode C. elegans from calcium imaging data. We show that the cognitive-level model of the NC-MCM framework provides a concise representation of the neuronal manifold of C. elegans and its relation to behavior as a graph, which, in contrast to other neuronal manifold learning algorithms, supports causal reasoning. We conclude the article by arguing that the ability of the NC-MCM framework to learn causally interpretable abstractions of neuronal dynamics and their relation to behavior in a purely data-driven fashion is essential for understanding biological systems whose complexity prohibits the development of hand-crafted computational models. Despite several decades of research, the way in which neuronal activity generates cognition remains elusive. A significant obstacle to understanding this phenomenon is the lack of a rigorous mathematical framework for exploring such questions, including a precise definition of what constitutes a cognitive state. In this work, we introduce such a framework along with machine-learning algorithms capable of learning cognitive states from neuronal data. A distinctive feature of our approach is causal consistency; that is, we develop a cognitive model that aligns with the neuronal-level model in such a way that causal assertions at one level can always be translated to the other level. We demonstrate the application of these models using the simple model organism C. elegans, in which we find highly reproducible cognitive states across multiple animals.
biochemical research methods,mathematical & computational biology
What problem does this paper attempt to address?