arxiv: 2605.11420 · v1 · submitted 2026-05-12 · ⚛️ physics.hist-ph · astro-ph.SR

Recognition: 2 theorem links

· Lean Theorem

Causality and Scientific Inquiry: Lessons from Space Physics and Medical Sciences

Amene Asgari-Targhi, Edward J. (Ned) Hall, Mahboubeh Asgari-Targhi, Marzieh Asgari-Targhi

Pith reviewed 2026-05-13 01:00 UTC · model grok-4.3

classification ⚛️ physics.hist-ph astro-ph.SR

keywords causalitymechanistic causalitydifference-making causalityapplied mathematicsstatistical modelingspace physicsmedical sciences

0 comments

The pith

Distinguishing mechanistic from difference-making causality allows better selection of mathematical and statistical methods in science.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines how the growth of statistical modeling from big data might reduce reliance on applied mathematics for understanding causes in science. Using examples from space physics and medical research, it illustrates that causality has two sides: one about the mechanisms at work and another about what factors make a difference in outcomes. Recognizing these differences allows scientists to pick the right tools—mathematical models or statistical analyses—for their specific causal questions. This choice can resolve conflicting findings and make overall scientific conclusions more dependable.

Core claim

The paper claims that over the past two decades, the rapid surge in data-intensive computational techniques for statistical modeling may have diminished the use of applied mathematics in causal scientific inquiry. By drawing on examples from space physics and medical sciences, it demonstrates how applied mathematical and statistical methods can differentiate between two fundamental facets of causality, i.e., mechanistic and difference-making. Understanding such foundational differences may help explain discrepant or erroneous research results and has the potential to strengthen the rigor and reliability of scientific inquiry through optimal selection of mathematical and/or statistical method

What carries the argument

The distinction between mechanistic causality and difference-making causality, as elucidated through applied mathematical models versus statistical modeling alone.

If this is right

Optimal selection of mathematical and/or statistical methods based on the type of causal question improves research rigor.
In space physics and medical sciences, this distinction can help resolve discrepant or erroneous findings.
Neglecting applied mathematics in favor of statistical modeling alone risks missing important causal mechanisms.
Understanding the relationship between causality and analytical approaches enhances the reliability of scientific results.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This perspective could be applied to data science practices in other disciplines to avoid similar pitfalls in causal claims.
It implies a need for interdisciplinary collaboration between mathematicians, statisticians, and domain scientists when designing studies.
A testable extension would be to examine historical cases where shifting from statistical to mathematical causal modeling revised prior conclusions.

Load-bearing premise

The rapid surge in data-intensive computational techniques has diminished the use of applied mathematics in causal scientific inquiry, and distinguishing mechanistic from difference-making causality can explain discrepant or erroneous research results.

What would settle it

A comparative study in space physics or medicine that applies both applied mathematical modeling for mechanisms and statistical methods for difference-making to the same causal question and measures whether one approach yields fewer discrepancies or errors than the other.

read the original abstract

Over the past two decades, the rapid surge in data-intensive computational techniques for statistical modeling may have had the effect of diminishing the use of applied mathematics in causal scientific inquiry. In this paper, co-authored by an astrophysicist, a mathematician, and philosophers, we assess the hazards of neglecting the branch of mathematics that constructs models to address causal questions in favor of statistical modeling alone. Causality is relevant in all branches of science and is often elucidated through applied mathematics. Here, we illuminate the idea with examples drawn from space physics and medical sciences. We examine causal questions to demonstrate how applied mathematical and statistical methods may differentiate between two fundamental facets of causality, i.e., mechanistic and difference-making. Understanding such foundational differences in causality may, in some cases, help explain discrepant or erroneous research results. Most importantly, understanding the relationship between causality and analytical approaches used in science has the potential to strengthen the rigor and reliability of scientific inquiry through optimal selection of mathematical and/or statistical methods.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a short discussion piece that applies the standard mechanistic versus difference-making causality split to space physics and medicine without new theorems, data, or derivations.

read the letter

The paper's main point is that distinguishing mechanistic causality from difference-making causality can help scientists pick between applied mathematical models and statistical techniques, and that ignoring the former risks weaker causal claims. The authors, who include an astrophysicist and a mathematician alongside philosophers, use examples from their fields to illustrate the idea. That cross-background setup is the clearest strength here; it lets them speak directly to practitioners who might otherwise treat the distinction as purely abstract philosophy. The abstract also states the practical payoff plainly: better method selection and fewer discrepant results in some cases. If the full text actually walks through one or two concrete cases where the choice between the two kinds of causality changed the analysis or explained a mismatch, that would be the part worth taking away. The rest stays at the level of reminder rather than discovery. The opening claim that data-intensive methods have diminished applied mathematics in causal work is presented without trend data, citations, or even a rough count of papers, so it functions more as background opinion than documented premise. The examples are described as illuminative but the abstract gives no specifics on what the models looked like or how the distinction resolved anything, which leaves the central suggestion untested in the summary. No equations, predictions, or falsifiable steps appear, so the piece does not advance a result that could be checked independently. This kind of paper is aimed at readers already working at the edge of philosophy of science and applied fields, such as space physicists or medical researchers who want a short conceptual check on their methods. It would not move core research in either domain but could fit a methods or foundations discussion. I would send it for peer review in a venue like Studies in History and Philosophy of Science or a space-physics methods journal, because the authors have the right mix of expertise and the topic is relevant, even though the manuscript will need more detailed examples to carry its weight.

Referee Report

2 major / 1 minor

Summary. The paper claims that the rapid rise of data-intensive statistical modeling over the past two decades has diminished the role of applied mathematics in causal scientific inquiry. It argues that distinguishing mechanistic causality from difference-making causality, illustrated through examples in space physics and medical sciences, can help explain discrepant or erroneous results in some cases and guide better selection of analytical methods to improve rigor and reliability.

Significance. If the distinction between mechanistic and difference-making causality can be shown to reliably inform method choice in the cited domains, the manuscript could encourage more careful integration of applied mathematics with statistical techniques in causal studies, potentially reducing mismatches between research questions and tools in fields like space physics and medicine.

major comments (2)

[Abstract and illustrative examples] The abstract and subsequent discussion reference examples from space physics and medical sciences to demonstrate how applied mathematical versus statistical methods differentiate mechanistic from difference-making causality, yet these examples are presented at a high level without detailed case analyses, specific equations, data, or derivations showing how one approach produces discrepant results and the other resolves them. This leaves the central claim about explanatory power and improved method selection without sufficient grounding to evaluate its applicability.
[Introduction] The opening premise that statistical modeling has diminished the use of applied mathematics in causal inquiry is stated without supporting evidence such as citation trends, publication statistics, or field-specific surveys; while not strictly circular, this assumption is load-bearing for the paper's motivation and would benefit from concrete documentation to avoid appearing as an untested assertion.

minor comments (1)

The terms 'mechanistic causality' and 'difference-making causality' are used without explicit definitions or pointers to the specific philosophical sources (e.g., Woodward or other standard references) from which the distinction is drawn, which could reduce accessibility for readers outside philosophy of science.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments and recommendation for major revision. We address each point below and describe the revisions we will make to strengthen the manuscript's grounding and motivation.

read point-by-point responses

Referee: [Abstract and illustrative examples] The abstract and subsequent discussion reference examples from space physics and medical sciences to demonstrate how applied mathematical versus statistical methods differentiate mechanistic from difference-making causality, yet these examples are presented at a high level without detailed case analyses, specific equations, data, or derivations showing how one approach produces discrepant results and the other resolves them. This leaves the central claim about explanatory power and improved method selection without sufficient grounding to evaluate its applicability.

Authors: We agree that the examples, while intended to illustrate the conceptual distinction, would benefit from greater specificity to demonstrate the claimed explanatory power. In the revised version we will expand the relevant sections with additional detail on the space physics and medical sciences cases, including selected equations, brief derivations, and references to concrete results or data sets that show how the two methodological approaches can yield discrepant causal inferences. revision: yes
Referee: [Introduction] The opening premise that statistical modeling has diminished the use of applied mathematics in causal inquiry is stated without supporting evidence such as citation trends, publication statistics, or field-specific surveys; while not strictly circular, this assumption is load-bearing for the paper's motivation and would benefit from concrete documentation to avoid appearing as an untested assertion.

Authors: The referee correctly identifies that the premise requires empirical support to avoid appearing unsubstantiated. We will revise the introduction to incorporate concrete documentation, drawing on available publication trend data, citation analyses, or relevant field surveys that illustrate the increased emphasis on data-intensive statistical modeling in causal research over the past two decades. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The manuscript is a purely conceptual and illustrative discussion paper with no equations, derivations, fitted parameters, quantitative predictions, or formal chains of reasoning. Its claims rest on established philosophical distinctions (mechanistic vs. difference-making causality) applied to examples from space physics and medical sciences, without any reduction of a result to its own inputs or to self-citations. No load-bearing steps exist that could be circular by construction, self-definition, or imported uniqueness; the argument is self-contained and does not derive new content from prior work by the same authors in a manner that would require external verification within the paper itself.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central position rests on the unproven premise that applied mathematics is being neglected and that the two facets of causality are the right lens for resolving discrepancies; no free parameters or new entities are introduced.

axioms (2)

domain assumption Causality has two fundamental facets: mechanistic and difference-making.
Stated directly in the abstract as the basis for differentiating analytical methods.
ad hoc to paper Neglecting applied mathematics in favor of statistical modeling alone creates hazards for causal inquiry.
Presented as the motivating premise without independent justification in the provided abstract.

pith-pipeline@v0.9.0 · 5491 in / 1274 out tokens · 31845 ms · 2026-05-13T01:00:00.933046+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear
We examine causal questions to demonstrate how applied mathematical and statistical methods may differentiate between two fundamental facets of causality, i.e., mechanistic and difference-making.
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
In mathematical modelling, the non-linear differential equations... capture the mechanistic structure of a dynamical system

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

[1]

toy problems,

that connects cause to its effect, and if we intend to find out via our intervention what it is for that cause to bring about its effect, then we use the intervention itself to do so. Woodward (2003) has put forward a much more polished concept of intervention that does not refer to the relationship between the variable intervened on and its effects and a...

work page 2003
[2]

as causal. The disagreement, controversy, and aftermath, i.e., the dominance of Keys hypothesis for three decades, are part of the social dimensions of science and how they affect scientific impact and reputation outside the scope of this paper. We can confidently draw a parallel between the methodologies the two scientists used and the two concepts of ca...

work page 2002
[3]

causation

defended his approach by stating that his cohort selection, though non-random, was based on practicality and dietary variation. (Keys, 1970, 1980; Kromhout and Menotti, 1994). 16 Here, we aim to look deeper at the impact of causality and focus on the conceptual conflations surrounding causal scientific questions that can give rise to a dominance of the us...

work page 1970
[4]

big data

made significant contributions to public health and medicine by addressing existing gaps and formalizing the concept of causality through statistical methodologies grounded in the potential outcomes framework. Their work extensively utilized large observational databases5, commonly referred to as big data. In comparison to the approach employed by Keys, H...

work page 2016
[5]

Darden (1986,

Two Case Studies from Mathematical Biology and Astrophysics: An Analytical Perspective This section examines how scientists validate causal hypotheses using mathematical tools to identify mechanisms and how the scientific community accepts these hypotheses. Darden (1986,

work page 1986
[6]

dynamic mechanistic explanations

argues that searching for mechanisms is essential to scientific discovery. Historical evidence shows that scientists use strategies to uncover causal pathways. Machamer et al. (2000) assert that both the philosophy of science and biology should focus on how scientists structure their research around mechanistic explanations. Philosophers Bechtel & Abraham...

work page 2000
[7]

uses equations tied directly to biophysical quantities such as membrane potential and ion channel conductance. In both cases, mathematical formalism functions not simply as a predictive tool but as a representational medium that encodes causal interactions over time within structured systems. Therefore, by distinguishing between these two modeling traditi...

work page 2016
[8]

or realistic/physiology-based models (H. F. Brown et al., 1984; Courtemanche Table

work page 1984
[9]

20 et al., 1998; Fulcher et al., 2014a; Ten Tusscher et al., 2003), and they correspond to the mechanistic approach to causality

Fundamental facets of causality and scientific modeling. 20 et al., 1998; Fulcher et al., 2014a; Ten Tusscher et al., 2003), and they correspond to the mechanistic approach to causality. These models provide a valuable quantitative and mechanistic approach to understanding the inherently complex structure of biological systems, including nonlinearities of...

work page 1998
[10]

and cardiac electrophysiology (H. F. Brown et al., 1984; Courtemanche et al., 1998; DiFrancesco & Noble, 1985; Luo & Rudy, 1991). In cardiac electrophysiology, models are regularly used to investigate mechanisms and treatments for cardiac arrhythmias (Callaghan et al., 2019; Pullinger & Gong, 2019). These models have identified the roles of specific ionic...

work page 1984
[11]

used a conceptual model of a human cardiac cell to investigate one of the potential mechanisms underlying a common irregular heart rhythm called atrial fibrillation, hence contributing to the understanding of the dynamical behavior of heart cells. In situations where existing data and methods limit researchers' ability to empirically test systemic perturb...

work page 2018
[12]

(iii) investigate the potential effectiveness of new or currently available drugs that may affect disease progression (Manga & Jha, 2017; Siepmann & Peppas,

work page 2017
[13]

For example, in the field of sleep and circadian rhythms, Phillips et al

(iv) develop testable hypotheses and design experiments for future therapies, (v) provide tools to reconcile contradictory results (Forger & Peskin, 2004; Gallego et al., 2006; Klerman et al., 1996), and (vi) propose new experiments based on model predictions or resolve debates. For example, in the field of sleep and circadian rhythms, Phillips et al. (20...

work page 2004
[14]

Their results suggested that chronotype is influenced by both circadian rhythm and noncircadian factors

and the circadian pacemaker (endogenous 24-hour cycles that regulate 21 physiological and psychological processes; Forger et al., 1999), and examined the interactions between these systems. Their results suggested that chronotype is influenced by both circadian rhythm and noncircadian factors. Their modeling was an example of a mechanistic approach that h...

work page 1999
[15]

These correlations and associations can be helpful in identifying patterns in the data, generating hypotheses, and designing new experiments

and identify correlations and associations between variables in a dataset. These correlations and associations can be helpful in identifying patterns in the data, generating hypotheses, and designing new experiments. However, they do not explain the causal mechanism connecting correlated variables, and thus by measuring the effects of a cause, their empha...

work page 2020
[16]

Asgari-Targhi et al., 2019)

(A) Fe XVIII spectroscopic intensity Ihot of the brightest loops as a function of total magnetic flux for the active regions is color-coded according to flux emergence or cancellation parameters (EC) (M. Asgari-Targhi et al., 2019). (B) The Kolmogorov-Smirnov plot compares the probability distributions of observations with no flux emergence (EC=0) versus ...

work page 2019
[17]

Asgari-Targhi et al., 2014)

where some of the model outputs have been validated with current solar observations(M. Asgari-Targhi et al., 2014). Furthermore, the models have been used to predict potentially detectable variations in other parameters (van Ballegooijen et al., 2017); these predictions are subject to validation by future NASA instruments with sufficiently high spatial re...

work page 2014
[18]

We began by making the distinction between scientific knowledge and scientific data

Discussion and Conclusion This paper offers a cautionary reminder of the scientific community’s attraction to big data and the use of statistical modeling techniques. We began by making the distinction between scientific knowledge and scientific data. We illustrated that statistical analyses often produce the latter rather than knowledge with predictive p...

work page doi:10.3109/07420528.2012.719971 1983
[19]

https://doi.org/10.1016/j.socscimed.2017.12.005 DiFrancesco, D., & Noble, D. (1985). A model of cardiac electrical activity incorporating ionic pumps and concentration changes. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 307(1133), 353–398. https://doi.org/10.1098/rstb.1985.0001 Facer-Childs, E. R., Boiling, S...

work page doi:10.1016/j.socscimed.2017.12.005 2017
[20]

https://doi.org/10.1186/s12916-015-0437-x Hall, N. (2004). Two concepts of causation. In J. Collins, N. Hall, & L. A. Paul (Eds.), Causation and counterfactuals (pp. 225–276). MIT Press. Harvey, K. L., Jones, H. P., Schrijver, C. J., & Penn, M. J. (1999). Does Magnetic Flux Submerge at Flux Cancellation Sites? Solar Physics, 190(1/2), 35–44. https://doi.o...

work page doi:10.1186/s12916-015-0437-x 2004
[21]

Law and Employment: Lessons from Latin American and the Caribbean

Heckman, J. (2000). Causal parameters and policy analysis in economics: A twentieth century retrospective. Journal of Economics, 115(1), 45–97. https://academic.oup.com/qje/article/115/1/45/1842259 Heckman, J. J. (2005). The Scientific Model of Causality. Sociological Methodology, 35(1), 1–97. https://doi.org/10.1111/j.0081-1750.2006.00164.x Heckman, J. J...

work page doi:10.1111/j.0081-1750.2006.00164.x 2000
[22]

https://doi.org/10.2307/2025310 Lewis, D. (1973b). Counterfactuals. Blackwell. Li, L. (2010). Dimension reduction for high-dimensional data. Methods in Molecular Biology (Clifton, N.J.), 620, 417–434. https://doi.org/10.1007/978-1-60761-580-4_14 Lim, S. S., Vos, T., Flaxman, A. D., Danaei, G., Shibuya, K., Adair-Rohani, H., AlMazroa, M. A., Amann, M., And...

work page doi:10.2307/2025310 2010
[23]

https://doi.org/10.1016/S0140-6736(12)61766-8 Luo, C

The Lancet, 380(9859), 2224–2260. https://doi.org/10.1016/S0140-6736(12)61766-8 Luo, C. H., & Rudy, Y. (1991). A model of the ventricular cardiac action potential. Depolarization, repolarization, and their interaction. Circulation Research, 68(6), 1501–1526. https://doi.org/10.1161/01.res.68.6.1501 Machamer, P., Darden, L., & Craver, C. F. (2000). Thinkin...

work page doi:10.1016/s0140-6736(12)61766-8 1991
[24]

Pariat, E., Aulanier, G., Schmieder, B., Georgoulis, M

Statistical Science, 5, 465–480. Pariat, E., Aulanier, G., Schmieder, B., Georgoulis, M. K., Rust, D. M., & Bernasconi, P. N. (2004). Resistive Emergence of Undulatory Flux Tubes. The Astrophysical Journal, 614(2). https://doi.org/10.1086/423891 Parker, E. N. (1972). Topological Dissipation and the Small-Scale Fields in Turbulent Gases. The Astrophysical Journal,

work page doi:10.1086/423891 2004
[25]

https://doi.org/10.1086/151512 Parker, E. N. (1983). Magnetic Neutral Sheets in Evolving Fields - Part Two - Formation of the Solar Corona. The Astrophysical Journal,

work page doi:10.1086/151512 1983
[26]

https://doi.org/10.1086/160637 Parker, E. N. (1994). Spontaneous current sheets in magnetic fields : with applications to stellar x-rays

work page doi:10.1086/160637 1994
[27]

Paul, L., & Hall, N. (2013). Causation: A User’s Guide. Oxford University Press. 34 Pearl, J. (2000). Causality: Models, reasoning, and inference. In Causality: Models, reasoning, and inference. Cambridge University Press. Pearl, J. (2002). Reasoning with cause and effect. In R. Trappl (Ed.), Cybernetics and systems 2002 (Vol. 1, pp. 103–110). Austrian So...

work page doi:10.1177/0748730410369208 2013
[28]

relaxation-oscillations

International Journal of Epidemiology, 14(4), 510–520. https://doi.org/10.1093/ije/14.4.510 Türkoğlu, G., & Selvi, Y. (2020). The relationship between chronotype, sleep disturbance, severity of fibromyalgia, and quality of life in patients with fibromyalgia. Chronobiology International, 37(1). https://doi.org/10.1080/07420528.2019.1684314 Taylor, T. W., G...

work page doi:10.1093/ije/14.4.510 2020
[29]

https://doi.org/10.1016/S0140-6736(74)91337-3 Yudkin, J., & Morland, J. (1967). Sugar Intake and Myocardial Infarction. The American Journal of Clinical Nutrition, 20(5), 503–506. https://doi.org/10.1093/ajcn/20.5.503 Zank, G. P., Dosch, A., Hunana, P., Florinski, V., Matthaeus, W. H., & Webb, G. M. (2012). The Transport Of Low-Frequency Turbulence In Ast...

work page doi:10.1016/s0140-6736(74)91337-3 1967
[30]

https://doi.org/10.1088/0004-637X/745/1/35 Zwaan, C. (1985). The emergence of magnetic flux. Solar Physics, 100(1–2). https://doi.org/10.1007/BF00158438

work page doi:10.1088/0004-637x/745/1/35 1985