Latent-Space Causal Discovery from Indirect Neuroimaging Observations
On the map: State-space sequence models (S4, Mamba) → Physiological signal modeling
Abstract (arXiv)
Neuroimaging does not observe causal variables directly: hemodynamics and volume conduction distort signals so that statistical dependence need not reflect latent neural influence. Before estimating graphs, one must specify under what assumptions delayed directed structure can be studied from such indirect observations. We formalize a conditional setting - recoverable inversion under modality physics together with nonstationary latent dynamics - and derive an inversion-error propagation bound under explicit assumptions. Building on this framing, we propose INCAMA (INdirect CAusal MAmba): physics-aware inversion coupled with a delay-aware Mamba encoder that uses mechanism shifts as informative variation for directed graph scoring. We use controlled simulations for quantitative validation and HCP motor-task fMRI as a zero-shot external transfer check based on anatomical and task-network consistency. Across TVB simulations, INCAMA improves directed-structure recovery by 2-3x in F1 over observation-space and two-stage baselines, and on HCP motor-task fMRI it produces sparse directed estimates concentrated in canonical visuo-motor pathways.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Physiological signal modeling, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 18 of its 55 references that are in the map. First posted on arXiv 30 Jan 2026.
Builds on
18 of 55 references.
References · 18
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
- 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
- 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
- 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,283
- 2022Simplified State Space Layers for Sequence Modelingcited by 776
- 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 745
- 2020HiPPO: Recurrent Memory with Optimal Polynomial Projectionscited by 600
- 2022On the Parameterization and Initialization of Diagonal State Space Modelscited by 418
- 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 329
- 2022It's Raw! Audio Generation with State-Space Modelscited by 171
- 2019Causal Discovery from Heterogeneous/Nonstationary Data with Independent Changescited by 166
6 more
- 2017Invariant Causal Prediction for Sequential Datacited by 63
- 2019Causal Discovery and Forecasting in Nonstationary Environments with State-Space Modelscited by 45
- 2020Reconstructing regime-dependent causal relationships from observational time seriescited by 28
- 2022Deep Recurrent Modelling of Granger Causality with Latent Confoundingcited by 8
- 2024Causal Discovery in Semi-Stationary Time Seriescited by 7
- 2025CausalMamba: Scalable Conditional State Space Models for Neural Causal Inferencecited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2022Neural State-Space Modeling with Latent Causal-Effect Disentanglement
- 2024Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics
- 2023Causal Reasoning in the Presence of Latent Confounders via Neural ADMG Learning
- 2025Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problem
- 2024Meta-Dynamical State Space Models for Integrative Neural Data Analysis
- 2026Coarse-to-fine Hierarchical Architecture with Sequential Mamba for Brain Reconstruction
- 2026CausalVAE as a Plug-in for World Models: Towards Reliable Counterfactual Dynamics
- 2025The Causal Round Trip: Generating Authentic Counterfactuals by Eliminating Information Loss