Department of Earth Sciences, Palaeobiology, Uppsala University, Villavägen 16, Uppsala, SE 752 36, Sweden.
Department of Statistics, School of Mathematics, University of Leeds, Leeds LS2 9JT, United Kingdom.
Evolution. 2018 Nov;72(11):2276-2291. doi: 10.1111/evo.13593. Epub 2018 Sep 26.
Survivorship biases can generate remarkable apparent rate heterogeneities through time in otherwise homogeneous birth-death models of phylogenies. They are a potential explanation for many striking patterns seen in the fossil record and molecular phylogenies. One such bias is the "push of the past": clades that survived a substantial length of time are likely to have experienced a high rate of early diversification. This creates the illusion of a secular rate slow-down through time that is, rather, a reversion to the mean. An extra effect increasing early rates of lineage generation is also seen in large clades. These biases are important but relatively neglected influences on many aspects of diversification patterns in the fossil record and elsewhere, such as diversification spikes after mass extinctions and at the origins of clades; they also influence rates of fossilization, changes in rates of phenotypic evolution and even molecular clocks. These inevitable features of surviving and/or large clades should thus not be generalized to the diversification process as a whole without additional study of small and extinct clades, and raise questions about many of the traditional explanations of the patterns seen in the fossil record.
生存偏差会在原本同质的系统发育发生-绝灭模型中随时间产生显著的明显速率异质性。它们是化石记录和分子系统发育中许多惊人模式的潜在解释之一。这种偏差之一是“过去的推动”:幸存了相当长时间的分支很可能经历了早期多样化的高速率。这就产生了一种随着时间的推移而趋于稳定的错觉,而实际上这是一种向均值的回归。在大的分支中还可以看到增加谱系生成早期速率的额外效应。这些偏差对化石记录和其他地方的多样化模式的许多方面都有重要但相对被忽视的影响,例如大灭绝后和分支起源时的多样化爆发;它们还影响着化石化的速度、表型进化速度的变化,甚至分子钟。因此,如果没有对小的和灭绝的分支进行额外的研究,就不应该将这些生存和/或大分支的必然特征推广到整个多样化过程,这也对化石记录中所看到的许多传统解释提出了质疑。