Why do some mammals live for decades while closely related species burn through their years in a fraction of the time? A new comparative study in Biogerontology pits two long-standing ideas against each other with pre-specified decision rules: the classical extrinsic-mortality prediction that safer niches permit longer life, and an information-turnover proposition holding that ancestral environmental variability should shorten captive lifespan once body mass, temperature and extrinsic mortality are controlled for. The verdict, drawn from thousands of species and a 4,100-species mammal phylogeny, favours niche safety and finds no detectable independent role for environmental fluctuation rate at the effect sizes the analysis could resolve.
The author, Daicun Li, assembled an unusually broad data foundation: 4,219 longevity records from AnAge, trait data from PanTHERIA, climate layers from WorldClim and CHELSA, and fossil occurrences from the Paleobiology Database spanning two mass extinctions. Rather than registering a third-party protocol, the study states its decision rules in advance, then runs a battery of phylogenetic generalized least squares models using Pagel’s lambda to tune how strongly species are treated as statistically non-independent because of shared ancestry.
The analysis first reproduces a classic benchmark: the lifespan–mass scaling law, with slopes of 0.158 for mammals and 0.212 for birds, at significance levels below 10 to the power of minus 140. The focal samples comprised 977 mammals and 1,063 birds for the scaling relationship, 798 species with climate data, 878 species with trait contrasts, 652 species in the joint phylogenetic model, and 349 species with basal metabolic rate measurements, 298 of them in the tree-matched model.
Against that backdrop, the information-turnover proposition, labelled P1 in the study, failed its most conservative tests. Weak proxy signals for environmental variability attenuated or reversed direction when lambda was fixed at 1, meaning full statistical dependence on the phylogeny, and under family fixed effects. The author explicitly cautions that the set of analyses is best read as overlapping sensitivity and robustness checks rather than independent falsifications, and that equivalence intervals for the climate coefficients exclude large, arboreality-sized effects but not small ones.
A parallel result emerged for metabolism. A basal-metabolic-rate correlate of lifespan, a recurring theme in ageing biology, likewise dissolved once lambda was set to 1. In other words, the apparent metabolic signature did not survive the strictest accounting for shared evolutionary history.
Turning the same pipeline to niche safety produced a sharper picture. Flight, arboreality, aquatic life and hibernation all carried positive mass-corrected coefficients in raw and maximum-likelihood-lambda contrasts. But in a joint model at lambda = 1 with 652 species, only multiply-origin arboreality remained significant, with a coefficient of +0.106 dex and a 95 percent confidence interval of 0.065 to 0.147, at P = 5.8 × 10−7. Flight and aquatic life were inconclusive, with wide confidence intervals that the study reports rather than resolves.
Ancestral reconstruction helps explain why arboreality stands out. The analysis recovered dozens of independent gains of arboreal living: 42 by parsimony, 45 to 50 by maximum likelihood, and a stochastic-mapping median of 70 to 81 with a 95 percent range of roughly 57 to 94. By contrast, flight or aquatic life evolved independently only one to a few times. That replication gives arboreality far more statistical leverage, and it suggests the trait repeatedly reshapes the mortality landscape species face.
The deep-time component extends the niche-safety logic to the fossil record. Geographic breadth in bivalves was associated with survival across both the K–Pg and end-Permian extinctions, although the effect weakened after controlling for sampling completeness. The effect weakened after controlling for sampling completeness, and the author frames it as a genus-level extinction phenomenon rather than a mechanism operating on individual lifespan. Still, the parallel is suggestive: lineages occupying broader, arguably safer geographic footprints persist when the world turns hostile.
Two further results anchor the extremes of the lifespan distribution. And human lifespan emerges as an extreme outlier on the scaling plane, sitting far above where a mammal of our body mass would be expected to fall.
The study is careful about what it does and does not claim. The comparative evidence is described as most consistent with a strong association between extrinsic-mortality regime and lifespan, and with no detectable independent role for environmental fluctuation rate at the effect sizes this analysis could resolve. The author reports where inference is inconclusive rather than claiming a definitive null effect or a decisive causal setting.
Subject of Research: A comparative phylogenetic analysis testing environmental variability and extrinsic mortality as correlates of mammalian lifespan.
Article Title: Environmental variability versus extrinsic mortality as correlates of mammalian lifespan: a pre-specified comparative analysis with phylogenetic and deep-time evidence
Article References: Li, D. (2026). Environmental variability versus extrinsic mortality as correlates of mammalian lifespan: a pre-specified comparative analysis with phylogenetic and deep-time evidence. Biogerontology, 27(6), Article 183. https://doi.org/10.1007/s10522-026-10531-z
Image Credits: AI Generated
DOI: 10.1007/s10522-026-10531-z
Keywords: lifespan, mammals, extrinsic mortality, phylogenetic comparative methods, arboreality, body mass scaling, mass extinctions, bivalves, turtles, basal metabolic rate, biogerontology, AnAge
News Source: Beatrice Stafford. (October 28, 2026). Niche Safety, Not Climate Swings, Tracks Mammalian Lifespan in Pre-Specified Comparative Test. Scienmag.
