Causality in Cognitive Neuroscience: Concepts, Challenges, and Distributional Robustness

2020/02/29 by Sebastian Weichwald, Jonas Peters
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · Medicine · Neuroscience · Psychology · #Artificial intelligence #Bayesian Modeling and Causal Inference #Causal inference #Causal model #Causal structure #Causality (physics) #Cognition #Cognitive neuroscience #Cognitive psychology #Cognitive science #Computer science #Dementia and Cognitive Impairment Research #Developmental psychology #Econometrics #Functional Brain Connectivity Studies #Generalizability theory #Inference #Mathematics #Neuroscience #Probabilistic logic #Psychology #Robustness (evolution) #q-bio.NC #stat.AP #stat.ME

paper · pdf · doi:10.1162/jocn_a_01623

published as Journal of Cognitive Neuroscience, 33(2):226-247, 2021

arxiv created 2020/07/03 · openalex publication_date 2020/08/19 · openalex created_date 2020/11/23 · arxiv updated 2021/04/12 · openalex updated_date 2026/08/05

Abstract

Whereas probabilistic models describe the dependence structure between observed variables, causal models go one step further: They predict, for example, how cognitive functions are affected by external interventions that perturb neuronal activity. In this review and perspective article, we introduce the concept of causality in the context of cognitive neuroscience and review existing methods for inferring causal relationships from data. Causal inference is an ambitious task that is particularly challenging in cognitive neuroscience. We discuss two difficulties in more detail: the scarcity of interventional data and the challenge of finding the right variables. We argue for distributional robustness as a guiding principle to tackle these problems. Robustness (or invariance) is a fundamental principle underlying causal methodology. A (correctly specified) causal model of a target variable generalizes across environments or subjects as long as these environments leave the causal mechanisms of the target intact. Consequently, if a candidate model does not generalize, then either it does not consist of the target variable's causes or the underlying variables do not represent the correct granularity of the problem. In this sense, assessing generalizability may be useful when defining relevant variables and can be used to partially compensate for the lack of interventional data.

Citations

• Visual Causal Feature Learning
• Synergy and redundancy in the Granger causal analysis of dynamical networks
• What we can do and what we cannot do with fMRI
• Global and local fMRI signals driven by neurons defined optogenetically by type and wiring
• Geometry of the faithfulness assumption in causal inference
• Identifiability of Gaussian structural equation models with equal error variances
• On the Identifiability of the Post-Nonlinear Causal Model
• Investigating Causal Relations by Econometric Models and Cross-spectral Methods
• Causal Discovery from Subsampled Time Series Data by Constraint Optimization
• Recovery of non-linear cause-effect relationships from linearly mixed neuroimaging data
• Anchor regression: heterogeneous data meets causality
• Structural causal models for macro-variables in time-series
• The Hardness of Conditional Independence Testing and the Generalised Covariance Measure
• On Causal and Anticausal Learning
• Invariant Risk Minimization
• Approximate Causal Abstraction
• Learning Sparse Nonparametric DAGs
• The lure of misleading causal statements in functional connectivity research
• The History, Development and Impact of Computed Imaging in Neurological Diagnosis and Neurosurgery: CT, MRI, and DTI
• The organization of the human cerebral cortex estimated by intrinsic functional connectivity
• Causal models for dynamical systems