Holocene selection for genetic variants associated with general intelligence, or; you’ll miss it when it’s gone!
In this special post, Michael Woodley discusses the issues around holocene selection for genetic variants. The accompanying paper ‘Holocene Selection for Variants Associated With General Cognitive Ability: Comparing Ancient and Modern Genomes‘ was published in Volume 20, Issue 4, of Twin Research and Human Genetics. You can read it for free up until October 31st, 2017.
We all know that human evolution ceased around fifty thousand years ago, when we left its African cradle and became a global species. With the exception of the odd surely inconsequential trait, such as the ability to digest dairy products or the development of paler, hairier bodies among the ancestors of today’s Europeans, not much has happened to our species’ genetic makeup. Instead we have culture, and it is to this realm that evolution is now strictly confined. There is nary an anthropologist who would disagree with this summation — but everything I just wrote is wrong.
That human evolution, since our African exodus “has been recent, copious and regional”, to quote Nicholas Wade, is perhaps one of the worst kept secrets in science. At least 8% of the human genome is now known to have changed since we left Africa, and the rate of adaptive evolution has been accelerating – with change over the last 10,000 years (the period known as the Holocene) having been on the order of 100 times that observed in previous epochs. In their groundbreaking book on the subject, Henry Harpending and Gregory Cochran termed this the 10,000 year explosion – a simultaneous play on Stephen Jay Gould’s ‘Cambrian explosion’ and a clever counter to his zealous promotion of the orthodoxy of human evolutionary stasis.
Why then has human evolution sped up? Far from being passive agents of a seemingly spontaneously complexifying culture, which saw human populations progress in pseudo-Marxist fashion from hunting and gathering, to agriculture and beyond that to urbanization, it would appear that culture is in fact a major source of selection pressures shaping our genomes, adapting them to its ever-changing contours. Neither does cultural change arise in a vacuum, but is itself the product of human ingenuity passed through a Darwinian filter – a process that separates useful from useless and damaging cultural change. This co-evolutionary interplay between culture and our genomes is the engine of accelerated human evolution. Runaway increases in cultural complexity favoured the no less rapid evolution of increasingly complex humans, especially with respect to those aspects of our cognition that allow us to cope with complexity, e.g. general intelligence – the core problem-solving faculty tapped by virtually all measures of intelligence.
As both a major cause and consequence of rising cultural complexity it is reasonable therefore to expect that the general intelligence of moderns, with their cities, stock markets, rocket ships and fidget spinners, will be higher than the comparatively ‘primitive’ peoples of, say, the Bronze Age, with their daggers, halberds and flanged axes. But how to test this possibility? Whilst our ancient ancestors did not take IQ tests, they did leave behind their genomes, which have been dutifully harvested and sequenced by those interested in documenting the genetic pedigree of modern man. It is with respect to the trove of ancient Bronze and early Iron Age genomes in particular that my colleagues and I turned in answering the question; ‘Are we smarter than the ancients?’
General intelligence (g) is a heritable trait. But genetic variants can currently be used to predict only relatively small amounts of variation in g and related measures (e.g. educational attainment). Having some of the variation explained is however good enough as this means that we can directly compare the ancient and the modern genomes with respect to these variants. To this end my colleagues and I constructed three different genetic indices (called polygenic scores) of g, using different combinations of variants and found that in each case, modern populations (from Europe) outclassed their ancient ancestors (from Eurasia). Furthermore, as our ancient genomes are not uniformly ancient – rather, they spanned 3.35 thousand years – we were able to show via correlation that these genomes gradually became enriched for g-promoting variants over time, corroborating the results of the first analysis.
Whilst this seems like good news, sadly the molecular trend towards increasing g appears to have come to an end – at least in the West – probably beginning in the early or mid 19th century. Writing in the journal Proceedings of the National Academy of Sciences, Augustine Kong and colleagues have documented this tragic decline in genetic variants for g across eight decades of Icelandic birth cohorts, hazarding that IQ may have fallen at a rate of a third of an IQ point a decade in consequence. Why have our evolutionary fortunes been reversed? In a nutshell conditions are presently too good. Many babies live, who would otherwise have died, many more children mature into adults who go on to start families than was the case historically, and critically, many high-ability individuals find creative excuses to produce few or no offspring. The evolutionary advantage by definition goes to the fittest who in modernity are not necessarily the brightest.
So the moral of our research is best captured by the advice that stockbrokers offer their clients: past performance is not necessarily a guide to the future. The conditions of the past that led to rising g are no longer pertinent to understanding the current state of human evolution, which is continuing apace, albeit perhaps not in the desired direction. Or to put it another way: We now understand how we got here, but where we are going is an entirely different question.
Read the paper in full for free until 31st October 2017.