# Testing Full Mediation of Treatment Effects and the Identifiability of Causal Mechanisms

arXiv: https://arxiv.org/abs/2603.04109 · HTML: https://mlsanity.com/paper/2603.04109/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 62 references that are in the map.
- First posted on arXiv 4 Mar 2026.

## Builds on (most cited first, 9 of 9)

- [Direct and Indirect Effects](https://mlsanity.com/paper/1301.2300/) (2013)
- [Estimating Treatment Effects using Multiple Surrogates: The Role of the Surrogate Score and the Surrogate Index](https://mlsanity.com/paper/1603.09326/) (2016)
- [Insights into the "cross-world" independence assumption of causal mediation analysis](https://mlsanity.com/paper/2003.10341/) (2020)
- [Doubly Robust Estimation of Direct and Indirect Quantile Treatment Effects with Machine Learning](https://mlsanity.com/paper/2307.01049/) (2023)
- [Contrasting Identifying Assumptions of Average Causal Effects: Robustness and Semiparametric Efficiency](https://mlsanity.com/paper/2111.15233/) (2021)
- [Ratio of Mediator Probability Weighting for Estimating Natural Direct and Indirect Effects](https://mlsanity.com/paper/2506.03284/) (2025)
- [Testing the identification of causal effects in observational data](https://mlsanity.com/paper/2203.15890/) (2022)
- [Testing Mechanisms](https://mlsanity.com/paper/2404.11739/) (2024)
- [Testing identification in mediation and dynamic treatment models](https://mlsanity.com/paper/2406.13826/) (2024)

## Abstract (arXiv)

In causal analysis, understanding the causal mechanisms through which an intervention or treatment affects an outcome is often of central interest. We propose a test to evaluate (i) whether the causal effect of a treatment that is randomly assigned conditional on covariates is fully mediated by, or operates exclusively through, observed intermediate outcomes (referred to as mediators or surrogate outcomes), and (ii) whether the various causal mechanisms operating through different mediators are identifiable conditional on covariates. We demonstrate that if both full mediation and identification of causal mechanisms hold, then the conditionally random treatment is conditionally independent of the outcome given the mediators and covariates. Furthermore, we extend our framework to settings with non-randomly assigned treatments. We show that, in this case, full mediation remains testable, while identification of causal mechanisms is no longer guaranteed. We propose a double machine learning framework for implementing the test that can incorporate high-dimensional covariates and is root-n consistent and asymptotically normal under specific regularity conditions. We also present a simulation study demonstrating good finite-sample performance of our method, along with two empirical applications revisiting randomized experiments on maternal mental health and social norms.