用新模型从脑影像数据中更准地推断神经因果关系。
CausalMamba: Scalable Conditional State Space Models for Neural Causal Inference
- 分两步走：先解卷积还原神经活动，再用条件Mamba推断因果图
- 在真实数据上识别出88%已知神经通路，传统方法失败率超99%
- 发现工作记忆中大脑会动态切换主导网络，传统方法看不到
我们提出CausalMamba，一种可扩展的框架，解决基于fMRI的神经因果推断中的根本问题：从血流动力学扭曲的BOLD信号中推断神经因果关系的病态性，以及现有方法（如动态因果建模，DCM）的计算不可行性。该方法将复杂逆问题分解为两个可处理阶段：先通过BOLD解卷积恢复潜在神经活动，再使用新型条件Mamba架构进行因果图推断。在模拟数据上，CausalMamba比DCM准确率高出37%。关键的是，在真实任务fMRI数据上，该方法以88%保真度恢复了公认神经通路，而传统方法在超过99%受试者中未能识别这些经典回路。此外，对工作记忆数据的网络分析揭示，大脑会根据刺激策略性地切换主要因果枢纽——调动执行或注意网络，这种精细重组是传统方法无法察觉的。本工作为神经科学家提供了大规模因果推断的实用工具，能捕捉认知功能背后的底层环路模式与灵活网络动态。
原文摘要 · Abstract (English)
We introduce CausalMamba, a scalable framework that addresses fundamental limitations in fMRI-based causal inference: the ill-posed nature of inferring neural causality from hemodynamically distorted BOLD signals and the computational intractability of existing methods like Dynamic Causal Modeling (DCM). Our approach decomposes this complex inverse problem into two tractable stages: BOLD deconvolution to recover latent neural activity, followed by causal graph inference using a novel Conditional Mamba architecture. On simulated data, CausalMamba achieves 37% higher accuracy than DCM. Critically, when applied to real task fMRI data, our method recovers well-established neural pathways with 88% fidelity, whereas conventional approaches fail to identify these canonical circuits in over 99% of subjects. Furthermore, our network analysis of working memory data reveals that the brain strategically shifts its primary causal hub-recruiting executive or salience networks depending on the stimulus-a sophisticated reconfiguration that remains undetected by traditional methods. This work provides neuroscientists with a practical tool for large-scale causal inference that captures both fundamental circuit motifs and flexible network dynamics underlying cognitive function.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。