Paper check · Magnetic fields and aging
A team led by the University of Nottingham kept healthy fruit flies and a Parkinson’s-model strain inside a magnetic shield for 50 to 60 days [1 pp. 1316–1317, 1319]. The paper’s abstract reports 20% longer lives in the Parkinson’s model and calls low-field exposure a potential non-invasive approach to neurodegenerative disease [1 p. 1316]. We checked six of its claims against its own statistics, keeping what the paper reports separate from our assessment.
Our analysis marks our own reasoning, and the verdicts are our assessment. Everything else comes from the numbered sources; for the paper [1], p. gives the journal’s page number and Fig. a figure or one of its panels.
0.50
Hazard ratio for Pink1 flies moved into the shield at day 20: a significantly lower risk of death than controls (p = 0.003) [1 Fig. 1F]
0.80
for Pink1 flies moved in at day 10, which spent longer in the shield: not significant (p = 0.284) [1 Fig. 1F]
40–60
flies per group, all male, in one shield with no sham enclosure [1 pp. 1317–1318]
The paper’s claims
The paper’s abstract [1 p. 1316]
Pink1 flies moved into the shield at day 20 survived significantly better than controls (hazard ratio 0.50, p = 0.003), but flies moved in at day 10, which spent longer in the shield, did not (0.80, p = 0.284) [1 Fig. 1F]. The paper doesn’t say how it calculated the 20% [1 p. 1316]. See section 2.
The abstract and discussion [1 pp. 1316, 1324]
The paper’s own tests found no significant change (hazard ratios 1.42 and 1.38; p = 0.225 and 0.273), and its results section says lifespan was not significantly altered [1 Fig. 1E][1 p. 1319]. The risk leaned higher, as it also did in the authors’ short pilot study, whose groups were not age-matched [1 Suppl. Fig. 1]. But approximate 95% intervals of about 0.8–2.5 fit no effect as well as a large one. See section 2.
The abstract and results [1 pp. 1316, 1321]
In this one-shield experiment, wild-type flies climbed better at days 20, 35 and 40 and Pink1 flies worse at days 20–35, with significant age-by-treatment interactions in both (p = 0.0026 and 0.0013) [1 Fig. 2]. The controls’ own scores swung widely between tests, so some single-day differences may reflect test-day noise [1 Fig. 2]. There was no sham enclosure, so what was tested is the shield as a whole, not necessarily the low field; blinding isn’t reported. See section 3.
The paper’s abstract [1 p. 1316]
Complex II-linked respiration was higher in shielded wild-type flies (p = 0.027, 4 chambers of 3 flies per group) and in a small supplementary test of wild-type flies shielded from day 20 (p = 0.050, 3 per group), where it rose because complex I respiration fell [1 Fig. 3C][1 Suppl. Fig. 3]. In Pink1 flies it was not significantly higher (p = 0.275) [1 Fig. 3D]. See section 2.
The paper’s abstract [1 p. 1316]
The diamond sensors’ free-radical signal did rise (overall treatment effect p = 0.0015) [1 p. 1319]. But the probe, TEMPOL, isn’t specific to superoxide: it also reacts with other radicals, and cellular reductants such as ascorbate, and mitochondria, turn it into the same non-magnetic form that the sensors count as scavenged radicals [1 p. 1318][9 §1.2, §2][10]. A change in the signal is shown; a rise in superoxide is not. See section 3.
The abstract and discussion [1 pp. 1316, 1325]
“Potential” hedges the claim, but the basis is one study in male flies, with no sham enclosure, no mitochondrial measurements in the group that lived longer and no tests in mammals [1 pp. 1317–1319]. The same treatment worsened the disease-related climbing deficit in Pink1 flies, as the authors note [1 p. 1321]. See section 4.
1 · The experiment
Two groups of flies were moved into a magnetic shield 10 or 20 days after emerging as adults and stayed there; controls stayed in Earth’s field. Survival was followed to day 70 [1 pp. 1317, 1319].
Only male flies were used, because the Pink1 mutation is on the X chromosome. Control flies of the laboratory strain w1118, which the paper calls wild type, were compared with Pink1B9 mutants [1 p. 1317].
Losing Pink1 gives flies features of early-onset Parkinson’s disease: faulty mitochondria, poor climbing, loss of dopamine neurons and shorter lives [1 p. 1317][4].
The flies were kept in a passive MuMetal benchtop shield (Magnetic Shields Ltd, model GA4); the paper points to a published description of the shield’s design [1 pp. 1316–1317][3].
The field inside, measured before and after the experiment, averaged 5.6 nT, give or take 0.3 nT; the paper doesn’t say what the margin represents [1 p. 1317]. The abstract rounds this to 5 nT [1 p. 1316]. The shield’s lid was opened to take flies out, change their food (every 2–3 days) and run tests [1 pp. 1317–1318].
Days since the flies emerged as adults; survival was followed to day 70
Logarithmic scale: each labeled step is ten times stronger than the one before
At Earth’s surface
23–67 µT
the field’s approximate range around the globe [2 Table 1]. The paper gives 25–60 µT [1 p. 1316].
Earth’s field is stronger by
Our analysis4,000–12,000×
roughly: 23,000 nT ÷ 5.6 nT ≈ 4,100; 67,000 nT ÷ 5.6 nT ≈ 12,000. Each time the lid was opened, the flies were briefly exposed to a stronger field.
Climbing
4–5
replicate tests per group at each age, 20 flies to a vial, every 5 days from day 15 to day 45. The paper doesn’t make clear whether these were separate vials or repeat tests of the same flies; its Figure 2 legend says “20 individuals were used per group” [1 p. 1318][1 Fig. 2].
Mitochondrial respiration
4
chambers per group, each with 3 pooled flies, from the groups shielded from day 10 [1 p. 1319]. A supplementary figure adds wild-type flies shielded from day 20, measured at day 72 (3 per group) [1 Suppl. Fig. 3].
Free radicals
5
replicates per group, read with diamond quantum sensors (nitrogen-vacancy centers in nanodiamonds) and the spin probe TEMPOL [1 p. 1318][1 Fig. 3]. The method was recently described in a preprint by the same group [8].
2 · What they found
Shielded wild-type flies climbed better and showed a non-significant trend toward shorter lives. Shielded Pink1 flies climbed worse, and those shielded from day 20 lived longer [1 Figs. 1–2].
| Measure | Wild-type flies | Pink1 mutant flies |
|---|---|---|
| Survival | No significant change Hazard ratios 1.42 and 1.38; p = 0.225 and 0.273 [1 Fig. 1E] |
Longer, day-20 group only Hazard ratio 0.50, p = 0.003; for the day-10 group, 0.80, p = 0.284 [1 Fig. 1F] |
| Climbing | Better At days 20, 35 and 40 in Figure 2A; the results text says days 20, 30 and 40 [1 Fig. 2A][1 p. 1321] |
Worse At days 20 to 35 in Figure 2B; the results text also lists day 40, where the figure shows no significant difference [1 Fig. 2B][1 p. 1321] |
| Complex II respiration | Higher p = 0.027, 4 chambers per group [1 Fig. 3C] |
No significant change p = 0.275, although combined complex I and II respiration was higher (p = 0.035) [1 Fig. 3B, 3D] |
| Free-radical signal | Higher during active respiration p = 0.025; not significantly at rest (p = 0.069) [1 Fig. 3E] |
Higher at rest and during active respiration p = 0.026 and 0.037, with a smaller rise between the two (group means only; no test reported) [1 Fig. 3E–F] |
Climbing was analyzed with mixed models that allow for repeated tests of the same vials. The age-by-treatment interaction was significant in both genotypes: wild-type F(10,61) = 3.16, p = 0.0026; Pink1 F(10,61) = 3.44, p = 0.0013 [1 p. 1319][1 Fig. 2]. Complex I-linked respiration did not differ significantly in either genotype (p = 0.694 and 0.378) [1 Fig. 3C–D]. “At rest” is the paper’s routine respiration and “active respiration” its combined complex I and II OXPHOS state [1 Fig. 3E].
The shield’s effect on survival differed between the genotypes (likelihood-ratio χ² = 8.13, df = 2, p = 0.017) [1 Fig. 1]. A hazard ratio below 1 means a lower risk of death than in the controls.
Hazard ratio, log scale
The dashed whiskers are approximate 95% intervals we back-calculated from each hazard ratio and p-value: wild-type day 10, 0.81–2.50; wild-type day 20, 0.78–2.45; Pink1 day 10, 0.53–1.20; Pink1 day 20, 0.32–0.79. The paper shows the exact intervals only as a chart, and ours match it to within about 0.02 [1 Fig. 1E–F].
| Shielded group | Hazard ratio | p | Our interval | Paper’s chart |
|---|---|---|---|---|
| Wild-type, from day 10 | 1.42 | 0.225 | 0.81–2.50 | 0.79–2.52 |
| Wild-type, from day 20 | 1.38 | 0.273 | 0.78–2.45 | 0.77–2.46 |
| Pink1, from day 10 | 0.80 | 0.284 | 0.53–1.20 | 0.51–1.22 |
| Pink1, from day 20 | 0.50 | 0.003 | 0.32–0.79 | 0.30–0.80 |
Hazard ratios and p-values from the paper [1 Fig. 1E–F]. Our intervals take z from the two-sided p-value and set the standard error of ln HR to |ln HR| ÷ z, then go 1.96 standard errors either side. The paper’s intervals are read from its chart, to within about 0.02.
The abstract calls this a 20% lifespan increase in Pink1 flies [1 p. 1316], and the university’s news release says “around 20%” [16], but the paper doesn’t say how the figure was calculated.
Pink1 controls
Our analysis≈44 days
median survival, read from the paper’s survival curves [1 Fig. 1B]
Pink1, shielded from day 20
Our analysis≈54 days
roughly 20–25% longer than the controls [1 Fig. 1B]
Shielding raised complex II-linked respiration in wild-type flies and raised the free-radical signal in both genotypes [1 Fig. 3C, 3E]. The authors read the complex II rise as a response to the stress of the low field and compare it with low oxygen [1 p. 1324].
In healthy flies, they suggest, the extra complex II activity places unnecessary demands on mitochondria and shortens life [1 p. 1324], although their own survival analysis found no significant change in wild-type lifespan [1 Fig. 1E]. In Pink1 flies, whose complex I works poorly, extra complex II activity could compensate, they propose [1 p. 1324].
They also propose that free radicals act as signals that switch on protective pathways [1 p. 1324], and they describe the pairing of longer life with worse climbing in Pink1 flies as paradoxical [1 p. 1316].
The study the authors cite for this found that complex II supplies heart cells’ reserve respiratory capacity and that activating it with drugs helped the cells survive low oxygen, but that low oxygen itself used up that reserve rather than raising it [11].
A second study, which they cite for oxygen’s ability to “activate or inactivate” complex II, found that severe low oxygen reduced complex II respiration in human blood cells [1 p. 1324][12].
3 · Headline claims
Headline claims, checked against the paper’s own numbers. The left and middle columns come from the paper; the verdicts on the right are our assessment.
HMF is the paper’s abbreviation for the hypomagnetic field inside the shield, GMF for Earth’s geomagnetic field and WT for wild-type; 10D and 20D are the groups shielded from day 10 and day 20.
“HMF exposure increased lifespan in Pink1⁻ D. melanogaster by 20%, with a paradoxical reduction in climbing ability. WT D. melanogaster had decreased lifespan and improved locomotor performance under HMF.”
“HMF exposure did not significantly alter overall lifespan compared to the same genotype in GMF (Figure 1A).”
“In WT flies, neither 10D HMF (HR=1.42, p=0.225) nor 20D HMF (HR=1.38, p=0.273) significantly altered mortality risk. In Pink1– flies, 20D HMF significantly reduced mortality risk relative to GMF (HR=0.50, p=0.003), while 10D HMF did not (HR=0.80, p=0.284).”
| The paper says (paraphrased) | What the paper’s statistics show | Our verdict |
|---|---|---|
| The shield increased Pink1 lifespan by 20%. | Significant for flies shielded from day 20 (HR 0.50, p = 0.003), not for flies shielded from day 10 (HR 0.80, p = 0.284) [1 Fig. 1F]. The 20% appears only in the abstract, with no method given [1 p. 1316]. | Partly trueHolds for one of the two shielded groups, and the number is unexplained. |
| Wild-type flies had shorter lives in the shield. | Neither shielded group differed significantly (HR 1.42, p = 0.225; HR 1.38, p = 0.273) [1 Fig. 1E], and the results section says lifespan was not significantly altered [1 p. 1319]. The abstract and discussion say it decreased [1 pp. 1316, 1324]. The hazard ratios lean toward higher risk, as did the authors’ short pilot study, whose groups were not age-matched [1 Suppl. Fig. 1]. | OverstatedThe abstract and the results disagree; the data fit no effect as well as a large one. |
| The shield improved wild-type climbing and worsened Pink1 climbing. | Significant at several ages in both genotypes (wild-type days 20, 35 and 40; Pink1 days 20 to 35), in mixed models that allow for repeated tests of the same vials [1 Fig. 2]. The controls’ own scores swung widely between tests (Pink1 controls about 53%, 95% and 56% at days 25, 30 and 35), so some single-day differences may reflect test-day noise [1 Fig. 2]. There was no sham enclosure, and blinding is not reported. | TrueIn this one-shield experiment. |
| The shield increased mitochondrial complex II activity. | Significant in wild-type flies (p = 0.027, 4 chambers per group; p = 0.050 in a supplementary comparison with 3 per group), not significant in Pink1 flies (p = 0.275) [1 Fig. 3C–D][1 Suppl. Fig. 3]. In the supplementary comparison, combined complex I and II respiration was about the same in both groups (about 60 and 57), and the complex II value rose because complex I respiration fell (from about 48 to 33) [1 Suppl. Fig. 3]. In the main wild-type comparison, combined respiration did not differ significantly (p = 0.089) [1 p. 1321]. | Partly trueWild-type flies only, on small samples. |
| The shield raised superoxide levels. | The free-radical signal rose under the shield (overall treatment effect p = 0.0015); the pairwise tests were not adjusted for multiple comparisons [1 p. 1319]. TEMPOL is not specific to superoxide. It also reacts with other radicals, such as hydroxyl, and inside cells it is reduced to a non-magnetic form, both in mitochondria and by reductants such as ascorbate; the sensors read that change the same way as scavenged radicals [1 p. 1318][9 §1.2, §2][10]. | OverstatedA change in the signal is shown; a rise in superoxide is not. |
| Low-field exposure is a potential non-invasive approach for neurodegeneration. | One study, in male flies only. The paper itself notes that shielding worsened the disease-related climbing deficit in Pink1 flies [1 p. 1321]. | Overstated“Potential” hedges it, but it still reaches well beyond the evidence. |
We rated each claim against the paper’s own statistics, on our standard scale. Because there was no sham enclosure, we rate each claim as a claim about the shield; the paper attributes the effects to the low field itself.
4 · Strengths and limitations
Our assessment of the study’s design, including what its three supplementary figures add [1 pp. 1329–1330].
5 · Reading the results with care
Shielded flies were compared with flies kept outside the shield, not with flies in an identical box that restored Earth’s field [1 p. 1318]. And a much larger, blinded study of brief exposures, which the paper cites only in passing, found no magnetic effect on fly climbing [5].
Without a sham enclosure, the box itself, not only the low field, could have changed the flies’ activity or metabolism, and that could affect both climbing and survival. Because activity does not reliably predict lifespan in fruit flies, this is a possibility, not a likely mechanism. The paper does not test it.
A simple trade-off between activity and lifespan is not established in fruit flies. Studies that tracked individual flies found that the less active ones were not the longer-lived [13][14], and a study of inbred fly lines found that metabolic rate did not predict lifespan [15].
The authors cite a large 2023 study [5] and a published comment disputing it [6], but only to say that other researchers “have considered whether MFs [magnetic fields] can influence D. melanogaster” [1 p. 1325].
The 2023 study, blinded and with sham controls, tested climbing in 10,960 flies and found no magnetic effect, including in a near-zero field; its authors stood by that finding in a reply [5][7]. Those flies were exposed for minutes, not weeks, so the study does not rule out the effects reported here.
“Under meticulously controlled conditions and with vast sample sizes, we have been unable to find evidence for magnetically sensitive behaviour in Drosophila.”
We read the full text and the three supplementary figures: a short pilot survival study whose groups were not age-matched [1 Suppl. Fig. 1], model checks, aging-rate estimates and a “dose-response” model [1 Suppl. Fig. 2], and a small respiration test of wild-type flies shielded from day 20 [1 Suppl. Fig. 3]. No raw data are public, so our median survival times are readings of the paper’s figures, not a re-analysis.
The whiskers on the survival chart are our approximation, not the paper’s, although they match the paper’s chart to within about 0.02 [1 Fig. 1E–F].
The enclosure may be better matched to the controls’ conditions than the methods describe [1 p. 1318].
We came to this paper while checking claims about magnetic-field catastrophes, which could prime us toward skepticism. To limit that, we rated each claim against the paper’s own statistics.
A sham shield: the same box with coils restoring Earth’s field.
Blinded scoring of climbing and deaths.
Several independent cohorts of a few hundred flies each.
Respiration and free radicals measured in the Pink1 flies shielded from day 20, with the normalization clearly stated; the methods and Figure 3 give different units [1 pp. 1318, 1323].
Both sexes.
Raw data shared openly.
These are our suggestions; the 2023 study shows what sham exposures with blinded experimenters look like [5]. A replication along these lines that again found longer survival in shielded Pink1 flies, with mitochondrial changes measured in the same flies, would firm up the lifespan and complex II claims. One that found no difference would weaken them.
6 · Conclusions
Hypomagnetic fields change mitochondrial physiology and free-radical production in fruit flies, with effects that depend on genotype: detrimental in healthy flies but life-extending in the Pink1 model. Low-field exposure, they suggest, could become a non-invasive, mitochondria-targeted approach for neurodegenerative disease [1 pp. 1316, 1325].
An interesting exploratory study that raises a fair question. Its strongest result, longer survival in Pink1 flies moved into the shield at day 20, rests on one shield, 40–60 flies per group and no sham control, and it came with worse climbing [1 p. 1318][1 Figs. 1–2]. Treat it as a lead worth replicating, not as evidence for a treatment or for Earth’s field shaping human aging.
Prepared by Fact-Checked Science from the open-access text and figures of Reed et al. (2026), Aging 18, 1316–1330, doi:10.18632/aging.206424, CC BY 4.0 [1]. The charts are redrawn from the paper’s reported numbers; the survival intervals are our approximation. The assessment is our own and has not been reviewed by the authors.
Bracketed numbers refer to this list. For the paper [1], p. and pp. are the journal’s page numbers (1316–1330), Fig. a figure or panel and Suppl. Fig. one of its three supplementary figures. Table 1 is in [2], and § marks a section of [9]. Links checked October 12, 2026.
Reed, J., Fromhold, M., Moisoi, N., Mather, M., and Chakrabarti, L. (2026). Hypomagnetic fields modulate lifespan, physical ability and mitochondrial metabolism in a Pink1 model of neurodegeneration. Aging 18(1), 1316–1330. Published September 23, 2026. Open access, CC BY 4.0.
Chulliat, A., Brown, W., Nair, M., Gomez Perez, N., Young, L.-Y., Watson, C., Boneh, N., Beggan, C., Meyer, B., and Paniccia, M. (2025). The US/UK World Magnetic Model for 2025–2030: Technical Report. NOAA National Centers for Environmental Information.
Hobson, P. J., Holmes, N., Patel, P., Styles, B., Chalmers, J., Morley, C., Davis, A., Packer, M., Smith, T. X., Raudonyte, S., et al. (2023). Benchtop magnetic shielding for benchmarking atomic magnetometers. IEEE Transactions on Instrumentation and Measurement 72, 1–9.
Park, J., Lee, S. B., Lee, S., Kim, Y., Song, S., Kim, S., Bae, E., Kim, J., Shong, M., Kim, J.-M., and Chung, J. (2006). Mitochondrial dysfunction in Drosophila PINK1 mutants is complemented by parkin. Nature 441(7097), 1157–1161.
Bassetto, M., Reichl, T., Kobylkov, D., Kattnig, D. R., Winklhofer, M., Hore, P. J., and Mouritsen, H. (2023). No evidence for magnetic field effects on the behaviour of Drosophila. Nature 620(7974), 595–599.
Kyriacou, C. P. (2024). Magnetic field responses in Drosophila. Nature 629(8010), E3–E5.
Bassetto, M., Reichl, T., Kobylkov, D., Kattnig, D. R., Winklhofer, M., Hore, P. J., and Mouritsen, H. (2024). Bassetto et al. reply. Nature 629(8010), E6–E7.
Menon, S., Reed, J., Ebanks, B., Shanmugham, M., Brookfield, A., Collison, D., Sow, M., James, V., Mongan, N. P., Moisoi, N., Jelezko, F., Chakrabarti, L., and Mather, M. L. (2025). Detecting mitochondrial free radicals with quantum sensors: from organelles to a D. melanogaster model of neurodegeneration. bioRxiv preprint 2025.11.29.691160, not peer reviewed.
Wilcox, C. S. (2010). Effects of tempol and redox-cycling nitroxides in models of oxidative stress. Pharmacology & Therapeutics 126(2), 119–145.
Bobko, A. A., Kirilyuk, I. A., Grigor’ev, I. A., Zweier, J. L., and Khramtsov, V. V. (2007). Reversible reduction of nitroxides to hydroxylamines: roles for ascorbate and glutathione. Free Radical Biology and Medicine 42(3), 404–412.
Pfleger, J., He, M., and Abdellatif, M. (2015). Mitochondrial complex II is a source of the reserve respiratory capacity that is regulated by metabolic sensors and promotes cell survival. Cell Death & Disease 6(7), e1835.
Riou, M., Charles, A.-L., Enache, I., Evrard, C., Pistea, C., Giannini, M., Charloux, A., and Geny, B. (2025). Acute severe hypoxia decreases mitochondrial chain complex II respiration in human peripheral blood mononuclear cells. International Journal of Molecular Sciences 26(2), 705.
Le Bourg, E. (1987). The rate of living theory. Spontaneous locomotor activity, aging and longevity in Drosophila melanogaster. Experimental Gerontology 22(5), 359–369.
Lints, F. A., Le Bourg, E., and Lints, C. V. (1984). Spontaneous locomotor activity and life span. A test of the rate of living theory in Drosophila melanogaster. Gerontology 30(6), 376–387.
Van Voorhies, W. A., Khazaeli, A. A., and Curtsinger, J. W. (2004). Testing the “rate of living” model: further evidence that longevity and metabolic rate are not inversely correlated in Drosophila melanogaster. Journal of Applied Physiology 97(5), 1915–1922.
University of Nottingham (September 23, 2026). Could Earth’s magnetic field influence how we age? News release.
Method
On October 12, 2026, we checked every statement, number, quotation and citation on this page against its source: the full paper and its three supplementary figures, its Figures 1–3 read directly from the images, and the 15 other sources listed here, about 77 items in all. Every hazard ratio and p-value held up, and so did the sample sizes as the paper states them, though the paper is unclear about its climbing replicates. The check changed the climbing days, which now follow the paper’s Figure 2 rather than its text; the first climbing results on the timeline (day 15, not day 5); and Earth’s surface field, which is about 23–67 µT in the World Magnetic Model report rather than 25–60 µT [1 Fig. 2][2 Table 1]. It also removed a statement that less active fruit flies tend to live longer, which studies of fly activity and metabolic rate contradict [13][14][15], and added what the supplementary figures show. A fresh reviewer then read the built page against the sources. That check corrected how we described the TEMPOL probe and the climbing replicates, made our account of the authors’ low-oxygen sources fairer, and added caveats to the climbing and complex II results. The change log lists every change.
The verdicts and the items tagged “Our analysis” are our own assessment, built on the cited sources. They include the approximate survival intervals, the field ratios, the median survival times we read from the paper’s Figure 1B, and our reading of how the paper’s statistics bear on each claim.
No raw data are public, so we re-ran only the arithmetic, and every number reproduced. The intervals we back-calculated from each hazard ratio and p-value match the intervals the paper plots in its Figure 1E–F to within about 0.02, and the paper’s χ² and F-test p-values come out as reported [1 Figs. 1–2]. Only the field ratio changed, with the new range for Earth’s field.
This page was researched and written with Claude, an AI assistant made by Anthropic, and then re-checked against its sources in a separate verification pass.
Corrections
If we get something wrong, we fix it and record the change below with the date. If a correction changes a verdict, the log says what the old verdict was.