An official website of the United States government
Abstract Many real-world decision-making tasks require learning causal relationships between a set of variables. Traditional causal discovery methods, however, require that all variables are observed, which is often not feasible in practical scenarios. Without additional assumptions about the unobserved variables, it is not possible to recover any causal relationships from observational data. Fortunately, in many applied settings, additional structure among the confounders can be expected. In particular, pervasive confounding is commonly encountered and has been utilised for consistent causal estimation in linear causal models. In this article, we present a provably consistent method to estimate causal relationships in the nonlinear, pervasive confounding setting. The core of our procedure relies on the ability to estimate the confounding variation through a simple spectral decomposition of the observed data matrix. We derive a DAG score function based on this insight, prove its consistency in recovering a correct ordering of the DAG, and empirically compare it to previous approaches. We demonstrate improved performance on both simulated and real datasets by explicitly accounting for both confounders and nonlinear effects.
more »
« less
Whole-brain causal discovery using fMRI
Abstract Despite significant research, discovering causal relationships from fMRI remains a challenge. Popular methods such as Granger causality and dynamic causal modeling fall short in handling contemporaneous effects and latent common causes. Methods from causal structure learning literature can address these limitations but often scale poorly with network size and need acyclicity. In this study, we first provide a taxonomy of existing methods and compare their accuracy and efficiency on simulated fMRI from simple topologies. This analysis demonstrates a pressing need for more accurate and scalable methods, motivating the design of Causal discovery for Large-scale Low-resolution Time-series with Feedback (CaLLTiF). CaLLTiF is a constraint-based method that uses conditional independence between contemporaneous and lagged variables to extract causal relationships. On simulated fMRI from the macaque connectome, CaLLTiF achieves significantly higher accuracy and scalability than all tested alternatives. From resting-state human fMRI, CaLLTiF learns causal connectomes that are highly consistent across individuals, show clear top-down flow of causal effect from attention and default mode to sensorimotor networks, exhibit Euclidean distance dependence in causal interactions, and are highly dominated by contemporaneous effects. Overall, this work takes a major step in enhancing causal discovery from whole-brain fMRI and defines a new standard for future investigations.
more »
« less
- Award ID(s):
- 2239654
- PAR ID:
- 10565333
- Publisher / Repository:
- DOI PREFIX: 10.1162
- Date Published:
- Journal Name:
- Network Neuroscience
- Volume:
- 9
- Issue:
- 1
- ISSN:
- 2472-1751
- Format(s):
- Medium: X Size: p. 392-420
- Size(s):
- p. 392-420
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
ABSTRACT Experiments have long been the gold standard for causal inference in Ecology. As Ecology tackles progressively larger problems, however, we are moving beyond the scales at which randomised controlled experiments are feasible. To answer causal questions at scale, we need to also use observational data —something Ecologists tend to view with great scepticism. The major challenge using observational data for causal inference is confounding variables: variables affecting both a causal variable and response of interest. Unmeasured confounders—known or unknown—lead to statistical bias, creating spurious correlations and masking true causal relationships. To combat this omitted variable bias, other disciplines have developed rigorous approaches for causal inference from observational data that flexibly control for broad suites of confounding variables. We show how ecologists can harness some of these methods—causal diagrams to identify confounders coupled with nested sampling and statistical designs—to reduce risks of omitted variable bias. Using an example of estimating warming effects on snails, we show how current methods in Ecology (e.g., mixed models) produce incorrect inferences due to omitted variable bias and how alternative methods can eliminate it, improving causal inferences with weaker assumptions. Our goal is to expand tools for causal inference using observational and imperfect experimental data in Ecology.more » « less
-
Abstract MotivationUnderstanding causal effects is a fundamental goal of science and underpins our ability to make accurate predictions in unseen settings and conditions. While direct experimentation is the gold standard for measuring and validating causal effects, the field of causal graph theory offers a tantalizing alternative: extracting causal insights from observational data. Theoretical analysis has shown that this is indeed possible, given a large dataset and if certain conditions are met. However, biological datasets, frequently, do not meet such requirements but evaluation of causal discovery algorithms is typically performed on synthetic datasets, which they meet all requirements. Thus, real-life datasets are needed, in which the causal truth is reasonably known. In this work we first construct such a large-scale real-life dataset and then we perform on it a comprehensive benchmarking of various causal discovery methods. ResultsWe find that the PC algorithm is particularly accurate at estimating causal structure, including the causal direction which is critical for biological applicability. However, PC does only produces cause-effect directionality, but not estimates of causal effects. We propose PC-NOTEARS (PCnt), a hybrid solution, which includes the PC output as an additional constraint inside the NOTEARS optimization. This approach combines PC algorithm’s strengths in graph structure prediction with the NOTEARS continuous optimization to estimate causal effects accurately. PCnt achieved best aggregate performance across all structural and effect size metrics. Availability and implementationhttps://github.com/zhu-yh1/PC-NOTEARS.more » « less
-
Abstract Recent philosophical work on causation has focused on distinctions across types of causal relationships. This paper argues for another distinction that has yet to receive attention in this work. This distinction has to do with whether causal relationships have “material continuity,” which refers to the reliable movement of material from cause to effect. This paper provides an analysis of material continuity and argues that causal relationships with this feature (1) are associated with a unique explanatory perspective, (2) are studied with distinct causal investigative methods, and (3) provide different types of causal control over their effects.more » « less
-
Abstract Relationships between plant biodiversity and productivity are highly variable across studies in managed grasslands, partly because of the challenge of accounting for confounding's and reciprocal relationships between biodiversity and productivity in observational data collected at a single point in time. Identifying causal effects in the presence of these challenges requires new analytical approaches and repeated observations to determine the temporal ordering of effects.Though rarely available, data collected at multiple time points within a growing season can help to disentangle the effects of biodiversity on productivity and vice versa. Here we advance this understanding using seasonal grassland surveys from 150 managed grassland sites repeated over 2 years, along with statistical methods that are relatively new in ecology, that aim to infer causal relationships from observational data. We compare our approach to common methods used in ecology, that is, mixed‐effect models, and to analyses that use observations from only one point in time within the growing seasons.We find that mixed models overestimated the effect of biodiversity on productivity by two standard errors as compared to our main models, which find no evidence for a strong positive effect. For the effect of productivity on biodiversity we found a negative effect using mixed models which was highly sensitive to the time at which the data was collected within the growing season. In contrast, our main models found no evidence for an effect. Conventional models overestimated the effects between biodiversity and productivity, likely due to confounding variables.Synthesis. Understanding the biodiversity‐productivity relationships is a focal topic in ecology, but unravelling their reciprocal nature remains challenging. We demonstrate that higher‐resolution longitudinal data along with methods to control for a broader suite of confounding variables can be used to resolve reciprocal relationships. We highlight future data needs and methods that can help us to resolve biodiversity‐productivity relationships, crucial for reconciling a long‐running debate in ecology and ultimately, to understand how biodiversity and ecosystem functioning respond to global change.more » « less
