Latent Dynamics · 2026-11-03
Identifiable Signed Causal Learning via Latent Stochastic Differential Equations for Inferring Brain Dynamic Effective Connectivity
Yiding Wang, Chen Qiao
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
ISCL is a continuous-time latent stochastic differential equation framework for dynamic effective connectivity. It derives a signed, observation-space causal influence operator from learned latent dynamics, with a rationality theorem (directed interventional response per brain state) and an identifiability result under invariance assumptions. On Lorenz96 and NetSim it recovers causal structure and reports lowest mean SHD; on HCP Social Cognition it yields a signed, seed-stable architecture and task reconfiguration.
Why it matters
The authors target dynamic effective connectivity that is both causal-rational and identifiable, and that outputs signed excitatory vs inhibitory influences in observation space (not only unsigned associations). Their signed, identifiable operator supports cross-seed consistency, enabling mechanistic and cross-subject comparison, and it supports task-dependent reconfiguration analysis in HCP.
Method
- Model neural dynamics as a latent SDE with drift μψ and isotropic diffusion, and map latent states to observations via a smooth decoder fθ (encoder mφ infers latent points).
- Define an observation-space causal influence operator Cx~ from decoder Jacobian/Hessian and latent drift, with a rationality theorem relating it to local interventional response in observation velocity.
- Under latent SDE invariance assumptions, prove identifiability: different parameterizations that match adjacent observation distributions induce the same observation-space causal operator up to isometric latent transformations; report signed dEC without temporal smoothing/windowing.
Abstract (from arXiv)
Understanding how brain regions interact to support cognitive tasks requires characterizing directed causal influences among them, known as effective connectivity (EC). As inter-regional coupling changes over time, a dynamic EC (dEC) estimate is required to track this temporal variation. Existing dEC learning methods typically lack a causal rationality guarantee or an identifiability analysis, and output unsigned connectivity that cannot distinguish excitatory from inhibitory influences, which limits further study of brain mechanisms. We propose ISCL, a latent stochastic differential equation framework that derives dEC from latent dynamics through a causal influence operator defined in observation space. As established by a rationality theorem, the operator captures a directed interventional response at each brain state, and under invariance assumptions the operator is identifiable from data, with its signed output separating excitatory from inhibitory influences. On synthetic benchmarks, \model recovers the causal structure on Lorenz96 and achieves the lowest mean structural Hamming distance among compared methods on NetSim, while preserving signed influence structures that distinguish excitatory from inhibitory interactions. Applied to HCP Social Cognition task-fMRI, ISCL reveals a signed architecture whose stability across seeds supports the identifiability guarantee, identifies dominant source and target networks consistent with social-cognition systems, and shows that condition-specific training strengthens causal outflow from the default-mode and frontoparietal networks during theory-of-mind processing. In this task, the recovered signed structure reveals network-level excitatory and inhibitory organization, with the dorsal-attention network as the only consistently positive out-strength source.
Related papers
- Identifying Neural Dynamics Using Nonlinear Interventional State Space Models
- Learning Transferable Policies from Action-free Time Series Through Dynamical Embeddings
- Similar Predictive Fit but Different Latent Dynamics: Characterizing Learned Dynamical Structure in Personalized Models of Brain Disorders
- A Bayesian Mirror Architecture for Emergent Consciousness: Circular Hierarchies, Self-Manifolds, and Hybrid Event-Self Binding
- Learning Computational Structure for Sample-Efficient Prediction
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First announced by arXiv: 2026-11-03 · Indexed by Pub-AI: 2026-11-03 UTC
Identifiable Signed Causal Learning via Latent Stochastic Differential Equations for Inferring Brain Dynamic Effective Connectivity
First announced by arXiv: 2026-11-03 · Original record ↗
- arxiv
- 2611.02876
Explicit source versions (1)
- v1 · Date not supplied
- Preprint available
Source date: 2026-11-03 · First detected: 2026-11-03 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · RSS guid=oai:arXiv.org:2611.02876v1; arxiv:announce_type=new; item pubDate (official metadata)
Abstract · arxiv
Understanding how brain regions interact to support cognitive tasks requires characterizing directed causal influences among them, known as effective connectivity (EC). As inter-regional coupling changes over time, a dynamic EC (dEC) estimate is required to track this temporal variation. Existing dEC learning methods typically lack a causal rationality guarantee or an identifiability analysis, and output unsigned connectivity that cannot distinguish excitatory from inhibitory influences, which limits further study of brain mechanisms. We propose ISCL, a latent stochastic differential equation framework that derives dEC from latent dynamics through a causal influence operator defined in observation space. As established by a rationality theorem, the operator captures a directed interventional response at each brain state, and under invariance assumptions the operator is identifiable from data, with its signed output separating excitatory from inhibitory influences. On synthetic benchmarks, \model recovers the causal structure on Lorenz96 and achieves the lowest mean structural Hamming distance among compared methods on NetSim, while preserving signed influence structures that distinguish excitatory from inhibitory interactions. Applied to HCP Social Cognition task-fMRI, ISCL reveals a signed architecture whose stability across seeds supports the identifiability guarantee, identifies dominant source and target networks consistent with social-cognition systems, and shows that condition-specific training strengthens causal outflow from the default-mode and frontoparietal networks during theory-of-mind processing. In this task, the recovered signed structure reveals network-level excitatory and inhibitory organization, with the dorsal-attention network as the only consistently positive out-strength source.
