Did Our Ancestors Face Oblivion 900,000 Years Ago?
Survival on the Brink or Statistical Mirage?
In 2023, a dramatic claim in Science1 suggested that human ancestors teetered on the edge of extinction around 900,000 years ago. The study, using a novel computational model called FitCoal, inferred that our ancestors experienced a catastrophic bottleneck, reducing the population to roughly 1,280 individuals. This bottleneck allegedly lasted for 117,000 years, with the survival of humanity hanging by a thread. While this idea captured headlines, a new study published in Genetics2 challenges the premise, arguing that the bottleneck may be a statistical artifact rather than historical reality.
The Original Claim: A Population Collapse in Deep Time
The 2023 study relied on advanced modeling to analyze genetic diversity in over 3,000 modern genomes, tracing mutations back to infer ancient population sizes. FitCoal revealed a striking bottleneck signature in African genomes but not in non-African populations. The authors suggested that this crisis coincided with climate upheavals during the mid-Pleistocene, potentially reshaping human evolution.
“The hypothesis is bold, yet intriguing,” the study's proponents argued.
They suggested the bottleneck could explain gaps in the African fossil record and speculated it marked the emergence of a "ghost lineage" in the human evolutionary tree.
Challenging the Bottleneck: A Statistical Artifact?
A 2024 study by Deng, Nielsen, and Song in Genetics systematically dismantles the bottleneck hypothesis, raising questions about FitCoal’s reliability. The authors argue that FitCoal’s findings likely stem from oversimplified assumptions about coalescence rates, the statistical process of tracing genetic lineages back to common ancestors.
“FitCoal artifactually tends to infer a sharp bottleneck when there in fact is none,” the study authors assert.
They explain that other computational models—such as multiple sequentially Markovian coalescent (MSMC) and mutation spectrum history inference (mushi)—failed to replicate the bottleneck, suggesting it might be a statistical illusion.
Limitations of Ancient Population Models
Dr. Aylwyn Scally, a geneticist at Cambridge University, also voiced skepticism about FitCoal. He highlights the inherent challenges in modeling ancient populations: “By the time you go back hundreds of thousands of years, most genetic differences coalesce into a single ancestor. Present-day genetic differences are very uninformative about anything beyond about 200,000 years ago.”
Scally argues that while computational tools like FitCoal can offer insights, they rely on simplifying assumptions that may obscure the messy reality of human evolution.
He adds, “If your method is the only one detecting a signal, you need to explain why. But the original study does not provide this clarity.”
Why the Bottleneck Likely Didn't Happen
Critics also point out that if a bottleneck as severe as proposed occurred, it would leave traces in all modern human genomes, including non-African populations. The absence of such evidence further undermines the claim.
A related issue lies in how FitCoal’s coalescence rates diverge from those of other models. Researchers speculate that FitCoal might overfit to specific genetic datasets, producing results that contradict broader trends.
Implications for Human Evolution
The debate over the 900,000-year bottleneck highlights the broader challenges of reconstructing human prehistory. Ancient population models, while powerful, are constrained by the complexity of evolutionary dynamics and limited genetic data from distant epochs. Despite these hurdles, advances in ancient DNA and computational tools continue to refine our understanding of humanity’s deep past.
“What's more probable for most people in the field is that their method [FitCoal] is responding to some aspect of the data in a particular or unusual way, which is probably an artifact rather than the truth,” concludes Scally.
Additional Related Research
Schiffels, S., & Durbin, R. (2014). Inferring human population size and separation history from multiple genome sequences. Nature Genetics, 46(8), 919–925.
DOI: 10.1038/ng.3015This study introduced MSMC, a method for reconstructing population history, offering insights into bottlenecks and expansions.
Prufer, K., et al. (2014). The complete genome sequence of a Neanderthal from the Altai Mountains. Nature, 505(7481), 43–49.
DOI: 10.1038/nature12886Highlights the use of ancient genomes to infer human and Neanderthal interactions, demonstrating the value of ancient DNA.
Li, H., & Durbin, R. (2011). Inference of human population history from individual whole-genome sequences. Nature, 475(7357), 493–496.
DOI: 10.1038/nature10231Introduced early methods for using genome-wide data to model demographic changes in humans.
Hu, X., et al. (2023). Ancestral population dynamics of modern humans inferred from whole-genome sequences. Science.
The original bottleneck hypothesis paper, laying the groundwork for the current debate.
Hu, W., Hao, Z., Du, P., Di Vincenzo, F., Manzi, G., Cui, J., Fu, Y.-X., Pan, Y.-H., & Li, H. (2023). Genomic inference of a severe human bottleneck during the Early to Middle Pleistocene transition. Science (New York, N.Y.), 381(6661), 979–984. https://doi.org/10.1126/science.abq7487
Deng, Y., Nielsen, R., & Song, Y. S. (2024). A previously reported bottleneck in human ancestry 900 kya is likely a statistical artifact. Genetics. https://doi.org/10.1093/genetics/iyae192