The study, led by Yutong Cheng and Qiuai Shu of The Second Affiliated Hospital of Xi’an Jiaotong University and published in Experimental Biology and Medicine, addresses a persistent problem in microbiome research. Observational studies have repeatedly reported microbial alterations in patients with functional gastrointestinal disorders, but the findings have been frustratingly inconsistent. Some studies found lower abundances of certain bacteria in patients compared with healthy controls, while others reported the opposite pattern or no difference at all. Without a way to establish the direction of these associations, it remained unclear whether the reported microbial changes were causes, consequences, or simply correlates of the disorders.

Mendelian randomization offers a way around this impasse by exploiting the random allocation of genetic variants at conception. Because single nucleotide polymorphisms are inherited independently of environmental factors and remain stable throughout life, they can serve as natural instruments for testing whether an observational association between an exposure and an outcome is consistent with a causal effect. The researchers drew their genetic data from genome-wide association studies of gut microbiota involving 14,306 individuals and from the FinnGen biobank, which provided summary statistics for irritable bowel syndrome based on 187,028 individuals and for functional dyspepsia based on 194,071 individuals, all of European ancestry.

The team selected 196 bacterial traits classified from phylum to genus level and applied rigorous quality control criteria to identify suitable instrumental variables. They used a relaxed significance threshold of P less than 5.0 times 10 to the negative 6 to obtain enough variants, performed linkage disequilibrium clumping with a threshold of r squared less than 0.001 and a window size of 10,000 kilobases, and calculated F statistics to ensure that all instruments exceeded the conventional threshold of 10, minimizing weak instrument bias. The inverse-variance-weighted method served as the primary analysis, with weighted median estimation as an auxiliary approach and a battery of sensitivity analyses to assess robustness.

The results revealed a striking asymmetry between the two disorders. For irritable bowel syndrome, higher genetically predicted abundances of the order Bifidobacteriales were associated with a lower risk of the condition, with an odds ratio of 0.741 and a 95 percent confidence interval of 0.570 to 0.963. The genus Eubacterium ventriosum group showed a similar protective pattern, with an odds ratio of 0.684 and a confidence interval of 0.524 to 0.893. Both associations were supported by both the inverse-variance-weighted and weighted median estimators, strengthening the case for their reliability.

For functional dyspepsia, the picture was more complex. A higher genetically predicted abundance of the genus Lachnospiraceae NK4A136 group was associated with an increased risk of the disorder, with an odds ratio of 1.368 and a confidence interval of 1.086 to 1.722. In contrast, the family Desulfovibrionaceae and the order Desulfovibrionales were protective, with an odds ratio of 0.649 and a confidence interval of 0.471 to 0.893. The researchers noted that the Desulfovibrionaceae and Desulfovibrionales analyses used the same seven instrumental variants and yielded nearly identical estimates, suggesting these represent highly overlapping rather than independent signals.

Sensitivity analyses provided strong support for the validity of these findings. Cochran’s Q test showed little evidence of heterogeneity among the variant-specific causal estimates. The MR-Egger intercept tests, which detect directional horizontal pleiotropy, returned P values ranging from 0.698 to 1.000 across the five significant associations, providing no evidence that the results were driven by pleiotropic pathways. MR-PRESSO global tests detected no overall pleiotropy and identified no outlier variants. Leave-one-out analyses confirmed that no single genetic variant disproportionately influenced any of the associations.

The reverse direction of the analysis yielded equally intriguing results. When the researchers asked whether genetically predicted disease risk influenced microbial abundance, they found that irritable bowel syndrome risk was associated with lower abundances of the genera Turicibacter and Slackia. The Turicibacter association was supported by both estimators, with a beta of negative 0.167 and a confidence interval of negative 0.283 to negative 0.052 in the primary analysis. No robust reverse associations were observed for functional dyspepsia. This bidirectional pattern suggests that the relationship between gut microbes and irritable bowel syndrome may be a two-way street, with the disorder itself capable of reshaping the microbial community.

The biological plausibility of these findings is supported by existing experimental evidence. Certain Bifidobacterium species can generate gamma-aminobutyric acid, an inhibitory neurotransmitter that may influence visceral sensitivity through the microbiota-gut-brain axis. A randomized controlled trial previously reported that a specific Bifidobacterium strain improved abdominal pain in women with irritable bowel syndrome. The Eubacterium ventriosum group includes butyrate-producing bacteria that support epithelial barrier integrity and mucosal immune regulation. Experimental studies have shown that Turicibacter interacts bidirectionally with host serotonergic signaling, providing a possible mechanistic link to the gut-brain communication pathway.

The authors are careful to frame these findings as hypothesis-generating rather than definitive. They emphasize that the identified microbial taxa are candidates for further mechanistic and clinical validation rather than established risk factors or therapeutic targets. The study has limitations, including the restriction to European ancestry populations, the limited number of instrumental variants for some microbial traits, and the reliance on 16S rRNA gene sequencing, which constrains analysis to the genus level and above. The fecal samples used in the microbiome studies may not represent the microbial composition at the gastroduodenal site most relevant to functional dyspepsia. Future studies should replicate these associations in independent and ancestrally diverse cohorts, employ strain-resolved shotgun metagenomics, and evaluate candidate strains and metabolites in mechanistic models before advancing to adequately powered intervention trials. Such stepwise validation could ultimately inform the development of more targeted and individualized microbiota-based strategies for these common and debilitating disorders.

Subject of Research: Genetics

Article Title: Causal effects of gut microbiota on functional gastrointestinal disorders: a bidirectional Mendelian randomization study

Article References: Cheng, Y., Shu, Q., Wang, Z., Wu, J., Peng, Y., Xie, J., Liang, X., Tao, Z., Shi, H., Wei, Z., Wang, J., & Xie, N. (2026). Causal effects of gut microbiota on functional gastrointestinal disorders: a bidirectional Mendelian randomization study. Experimental Biology and Medicine, 251, Article 11023. https://doi.org/10.3389/ebm.2026.11023

Image Credits: AI Generated

DOI: 10.3389/ebm.2026.11023

Keywords: gut microbiota, Mendelian randomization, irritable bowel syndrome, functional dyspepsia, functional gastrointestinal disorders, GWAS, Bifidobacteriales, Eubacterium ventriosum group, Lachnospiraceae NK4A136 group, Desulfovibrionaceae, gut-brain axis, causality

A team led by Hyojin Park, Nam-Hyun Yoo and Jinhong Yang has now released CD-CureNO, an open-source software package published in the journal SoftwareX that wraps neural-operator surrogates for composite curing in an unusually rigorous verification and provenance framework. The software, versioned at v0.0.3 and archived in Software Heritage under an Apache 2.0 license, is written in Python 3.10 through 3.12 using PyTorch, NumPy, SciPy and pandas, and it runs its audits and test suites on ordinary CPUs, with CUDA reserved for optional model training. What distinguishes the project is not a claim of new mathematics—the authors explicitly disclaim novelty for Fourier operators, joint temperature-and-cure prediction, or physics-based losses—but a domain-specific integration and verification layer built around inspectable scientific controls.

The motivating problem came from a legacy residual Fourier neural operator that stacked up to 51 independently trained positional models for curing analysis. Rather than silently replacing that code, CD-CureNO keeps two reproduction paths separately named: one that reproduces the legacy behavior exactly, and one that corrects known defects. The team regression-tested nine legacy defects, including normalization leakage and incomplete field outputs, so that the historical behavior is documented rather than erased. From that audited foundation, the software extends to joint temperature-and-degree-of-cure fields, causal response, conservative label generation, and transfer of learned knowledge across spatial dimensions.

The architecture is organized around explicit development and validation phases, labeled P0 through P6. P0 audits legacy behavior, P1 reproduces and corrects it, P2 develops joint one-plus-one-dimensional operators over the thickness-time field, P3 supplies a one-dimensional solver stage, P4 generates a true two-dimensional benchmark with two spatial coordinates evolving in time, P5 tests transfer, restriction and causality, and P6 is reserved for a future confirmatory held-out campaign. Each stage emits inspectable artifacts—configuration snapshots, Git and environment records, checkpoints, predictions and machine-readable receipts—that the next stage consumes, with cross-cutting guardrails verified again during postflight aggregation and reporting.

The authors define a result as auditable when its input, code, configuration, split, seed and checkpoint are all identifiable; when its reported metric can be recalculated from saved case outputs; when integrity violations or missing prerequisites fail explicitly; and when every claim can be traced back to those artifacts. Hashes, manifests, tensor mappings and receipts supply the observable evidence. Failed and superseded runs remain on record instead of being replaced by result-selected retries, and complete cure cycles or simulation cases—not individual grid points—define the samples. Crucially, the team distinguishes its independent postflight calculations, which are separate from the training and execution path, from validation by an unaffiliated research group or an independent physical model. A new NumPy-only example verifier imports neither training nor model code and reconstructs metrics from saved fields; a deliberately altered metric is rejected even after its file checksum is updated, and a failed recheck replaces any earlier success receipt.

Two technical contracts give the software its distinctive character. The first is restriction preservation: for laterally uniform coefficients, initial conditions and boundary data with zero lateral flux, an x-independent solution should reduce to the same through-thickness problem at every lateral position. The two-dimensional target therefore implements the lifted one-dimensional field plus a residual correction, with copied source tensors and zero-residual lateral branches that recover the source mapping exactly at initialization up to floating-point ordering. A verifier reconstructs the target state from the checkpoint, since output equality alone would not prove provenance. The second contract is causality: the causal model replaces temporal Fourier transforms with dilated convolutions, and perturbing all future input channels must leave every preceding output field unchanged. A nonnegative latent cure rate guarantees bounded, monotone cure evolution, and finite-volume-compatible fluxes handle material interfaces.

The verification evidence is unusually detailed. A GitHub Actions run on Ubuntu 24.04 passed 224 tests with 30 explicitly documented skips in about 42 seconds, and a fresh Windows run on a separate commit passed the same counts in about 48 seconds; continuous integration even exposed an input-stride-dependent CPU rounding difference, fixed by making shared target channels contiguous, with a new regression test added. The historical two-dimensional benchmark holds 512 temperature-and-cure cases of roughly 918 megabytes, anchored by 27 prespecified solver checks, zero failed cases, a maximum relative global energy residual of 1.345 times ten to the minus twelve, and exact float32 slice-hash replay for selected cases. Fresh regeneration of all 512 cases took 214.5 seconds on the documented Windows machine, and independent hashing confirmed that all eight output arrays match the frozen originals byte for byte. The team is candid that strict Linux plan validation rejected 33 heat-transfer values differing from the frozen Windows plan by at most 2.842 times ten to the minus fourteen, so canonical replay requires the documented compatible environment.

New in the revision is an independent mathematical verification of the production solver against elementary continuous reference solutions. Four synthetic case families—an anisotropic insulated transient, steady manufactured fields exercising all four Robin boundary faces with uniform and then discontinuous conductivity, and a solvable cure specialization coupled to reaction heating—were specified in an archived protocol with acceptance criteria fixed before execution. All 17 scheduled configurations and 72 recorded acceptance items passed. Measured spatial convergence orders clustered near two, temporal orders near one, consistent with backward Euler, and the synthetic cure equation gave orders near four, consistent with the production fourth-order Runge-Kutta update. Finest temperature relative errors ranged from 5.82 times ten to the minus five to 5.10 times ten to the minus four, and the finest cure maximum error was 1.80 times ten to the minus eight. The authors stress the boundaries: these studies verify anisotropic diffusion, boundary treatment and a special reaction-heat coupling, but not general kinetic validity, proprietary solver parity or experimental validation.

A fully public end-to-end example demonstrates the entire workflow without any private data or checkpoints. It generates eight true two-dimensional cases with laterally varying Robin coefficients, trains production causal operators with a fixed seed for 32 source and 48 target epochs, transfers knowledge through tensor copying and zero-residual initialization, and then recomputes held-out metrics in a separate NumPy-only postflight process. The two held-out cases yielded temperature relative errors of 0.006408 and 0.009741, temperature root-mean-square errors of 2.199 and 3.385 kelvin, and cure errors near 0.028, with future-prefix change exactly zero and independently recomputed metrics agreeing to within 4.44 times ten to the minus sixteen. The complete run took under ten seconds on a desktop CPU. The authors deliberately label this a functionality example, not evidence of superiority, and the figures show visible prediction errors rather than flattering comparisons.

Looking forward, the team has published a registered plan for a confirmatory P6 campaign: 40 held-out out-of-distribution cases across five target seeds, paired with a generic causal transfer baseline and analyzed with 10,000 crossed paired bootstrap replicates. A confirmatory claim would require a registered positive 95 percent improvement-interval bound, at least a 10 percent case-median improvement, physical-error guardrails and complete verified execution without test-based selection—and the authors note plainly that 40 cases across five seeds are not 200 independent cases and do not guarantee adequate statistical power. That discipline is the real story of CD-CureNO. In a field where surrogate models often arrive with cherry-picked demos and unverifiable numbers, the package demonstrates that neural operators for manufacturing can ship with the same provenance discipline as the experiments they hope to replace: every number traceable to a run, every failed attempt preserved, and every claim bounded by the evidence that produced it.

Subject of Research: Auditable restriction-preserving causal neural operators for thermochemical composite curing simulation

Article Title: CD-CureNO: Auditable software for restriction-preserving causal neural operators in composite curing

Article References: Park, H., Yoo, N.-H., & Yang, J. (2026). CD-CureNO: Auditable software for restriction-preserving causal neural operators in composite curing. SoftwareX, 36, Article 103086. https://doi.org/10.1016/j.softx.2026.103086

Image Credits: AI Generated

DOI: Not provided

Keywords: neural operators, composite curing, Fourier neural operator, software verification, reproducibility, provenance, causality, thermochemical modeling, surrogate models, manufacturing simulation, open-source software, convergence testing

The distinction between correlation and causation sits at the heart of the research. Prior work has established that language models can report calibrated confidence estimates: when a model assigns a high probability to its own answer, that answer is more likely to be correct, and when the model expresses doubt, errors become more frequent. But such correlations leave open an ambiguity. A model might compute an internal confidence signal and then act on it, or the confidence estimate might simply be a by-product of the same computational processes that generate the answer, exerting no independent influence on behaviour. Disentangling the two requires more than observational measurement; it requires intervention.

That is precisely the approach taken in the new work. Rather than passively recording what models say about their confidence, the researchers manipulated the internal states associated with confidence and observed whether behaviour changed as a result. By intervening directly on the representations that carry confidence information within a model’s activations, the team could test whether those representations play a genuine causal role in driving the model’s subsequent outputs. The logic mirrors classic interventionist methods in neuroscience and psychology, where researchers establish that a neural signal causes behaviour by perturbing it and watching what happens, rather than merely noting that the two tend to co-occur.

The results, according to the study, show that when confidence-related internal states are amplified, models behave in ways characteristic of high-confidence systems: they commit more firmly to their answers, become less likely to hedge or express uncertainty, and shift their response patterns in measurable ways. When those same states are suppressed, the opposite occurs. Models become more tentative, more prone to qualify their statements, and more likely to distribute probability across multiple candidate answers. Crucially, these behavioural shifts occur without altering the model’s underlying knowledge or the input prompt, indicating that the confidence signal itself is doing causal work in the decision-making pipeline.

The implications extend well beyond academic curiosity. If a language model’s confidence genuinely drives its behaviour, then the way a model expresses certainty is not a cosmetic layer added on top of reasoning—it is an integral part of how the system decides what to say. This reframes long-standing concerns about overconfident AI. Models that state falsehoods with unwavering assurance may do so not because they lack a mechanism for detecting their own errors, but because the confidence signal that should temper their claims has been miscalibrated or overridden. Conversely, models that hedge excessively may be suffering from systematically deflated internal confidence, even when their answers are largely correct.

The findings also speak to a growing body of research on introspection in artificial systems. One of the most debated questions in AI safety is whether language models have meaningful access to their own internal computational states—whether, in some functional sense, they know what they know. Self-report alone is a weak guide, because models can be trained to produce plausible-sounding confidence statements without those statements reflecting anything real. The new causal evidence strengthens the case that something substantive underlies these reports: the confidence a model expresses is connected, through identifiable internal mechanisms, to the behaviour it subsequently produces. That connection is exactly what one would expect if the model is, in a limited but genuine sense, monitoring and acting on its own uncertainty.

Methodologically, the study represents a maturing of interpretability research. Early work in the field focused on locating where particular concepts are encoded in a network’s layers, producing maps of representation that were suggestive but often causally inert. The field has since shifted toward intervention-based techniques—activating, suppressing, or editing internal features and measuring the downstream consequences. The new research applies this interventionist toolkit to metacognition, the capacity of a system to represent its own cognitive states. Demonstrating that confidence representations pass causal intervention tests places them in the same evidentiary category as other well-established internal features, such as those implicated in factual recall and instruction following.

For practitioners deploying language models in high-stakes settings, the research carries practical weight. Calibration techniques, uncertainty quantification, and abstention mechanisms—methods that allow models to say when they do not know—are increasingly treated as essential safety infrastructure for medical, legal, and scientific applications. The new findings suggest these mechanisms are not merely bolted-on output filters but are entangled with the model’s core decision processes. Interventions that improve calibration at the level of internal confidence representations could therefore propagate through the entire behavioural stack, making the model more honest about its limits in a way that end-to-end fine-tuning alone may not reliably achieve. At the same time, the causal link cuts both ways: anything that distorts a model’s internal confidence—adversarial inputs, distribution shift, or aggressive post-training—could distort not just its stated confidence but its actual behaviour in hard-to-predict ways.

The study also raises questions that the authors and the broader field will now need to address. Confidence in language models is not a single scalar; it is distributed across layers, tokens, and computational pathways, and different components of a network may encode different, sometimes conflicting, estimates of reliability. Understanding which of these representations are causally potent—and which are epiphenomenal—will be essential for building systems whose self-assessments can be trusted. There are also open questions about generality: whether the causal role of confidence holds across model scales, architectures, and training regimes, and whether the same mechanisms govern other forms of self-monitoring, such as a model’s assessment of the safety or harmfulness of its own outputs.

What the research establishes, for now, is a conceptual milestone. Language models are not black boxes that merely emit text; they are systems whose internal states, including states that encode confidence, play demonstrable causal roles in shaping behaviour. That realization moves the field closer to a science of machine metacognition—one in which a model’s relationship to its own uncertainty can be measured, verified, and engineered rather than assumed. As language models are entrusted with ever more consequential decisions, knowing that they use their confidence to drive what they do is a crucial step toward ensuring that what they do remains worthy of the confidence we place in them.

Subject of Research: Causal role of internal confidence representations in driving the behaviour of large language models

Article Title: Causal evidence that language models use confidence to drive behaviour

Article References: Kumaran, D., Daw, N., Osindero, S., Veličković, P., & Patraucean, V. (2026). Causal evidence that language models use confidence to drive behaviour. Nature Machine Intelligence. https://doi.org/10.1038/s42256-026-01293-x

Image Credits: AI Generated

DOI: 10.1038/s42256-026-01293-x

Keywords: large language models, confidence, causality, interpretability, AI safety, metacognition, calibration, uncertainty, neural networks, machine learning, Nature Machine Intelligence, self-monitoring