Fight the four horsemen of irreproducibility, Nature

In this editorial, Dorothy Bishop explores how threats to reproducibility in science are recognized but not address for centuries, and how we might finally bring them under control. Surprisingly, not many scientists are concerned with the current scientific practice and persist in working in a way almost guaranteed not to deliver meaningful results. They ride on the four horsemen: publication bias, low statistical power, P-value hacking and HARKing (hypothesizing after results are known).

Publication bias refers to the fact that researchers are less likely to write up studies that show no effect, and journal editors are less likely to accept them. Consequently, no one can learn from them, and researchers waste time and resources on repeating experiments, redundantly. For example, if there are 20 studies of a drug and only one shows a benefit, but that is the one that is published, we get a distorted view of drug efficacy. This misleading effect is also shown by the growing use of meta-analyses, which combine results across studies.

Studies with low statistical power produce meaningless, unreliable data, yet researchers still produce them in large quantity. 

The problem of manipulating data set to produce significant p-value has been shown beautifully by an elegant, comic paper in which the authors crafted analyses to prove that listening to the Beatles could make undergraduates younger (J. P. Simmons et al. Psychol. Sci. 22, 1359–1366; 2011). “Undisclosed flexibility,” they wrote, “allows presenting anything as significant.”

HARKing is similar to P Hacking. Researchers would look at the result and then reverse engineer the question. However, p-values are meaningless when taken out of context of all the analyses performed to get them. 

"The problems are older than most junior faculty members, but new forces are reining in these four horsemen. First, the field of meta-science is blossoming, and with it, documentation and awareness of the issues. We can no longer dismiss concerns as purely theoretical. Second, social media enables criticisms to be raised and explored soon after publication. Third, more journals are adopting the ‘registered report’ format, in which editors evaluate the experimental question and study design before results are collected — a strategy that thwarts publication bias, P-hacking and HARKing. Finally, and most importantly, those who fund research have become more concerned, and more strict. They have introduced requirements that data and scripts be made open and methods be described fully."

How Did Our Diet Allow Us to Say F-words? Science

As we all know, one of the most popular words in English starts with an “f”. However, you might not have noticed that the ‘f’ sound is actually harder to pronounce than ‘a’s or ‘o’s. Consonants such as ‘f’ and ‘v’ are called labiodentals and articulated by the upper teeth and lower lip. That requires us to have an overbite, as shown on the right side of the picture above. Now, researchers have found that the development of our overbite was actually a result of our ancestors switching their diet from raw meat and plant fiber to processed food.
In 1985, American linguist Charles Hockett proposed that Chewing gritty, fibrous foods puts force on our growing jaw bone and wears down molars, which caused our lower jaw to grow larger, and our molars to erupt farther so that the upper and lower teeth align. That edge-to-edge bite makes it harder to push the upper jaw forward to touch the lower lip, which is required to pronounce labiodentals.
In their recently published paper, researchers from the University of Zurich tested that theory. First, they used computer modeling to show that with an overbite, producing labiodentals takes 29% less effort than with an edge-to-edge bite. Then, they scrutinized the world's languages and found that hunter-gatherer languages have only about one-fourth as many labiodentals as languages from farming societies. Furthermore, in ancient India and Rome, labiodentals might even have been a mark of status, signaling a softer diet and wealth. Finally, they looked at the relationships among languages, and found that labiodentals can spread quickly. The sounds went from being rare to common in just 8000 years since the widespread adoption of agriculture and new food processing methods such as grinding grain into flour, and today they appear in 76% of Indo-European languages.
Researchers concluded that it was the changes in diet attributable to food-processing technologies that modified the human bite from an edge-to-edge configuration to one that preserves adolescent overbite, and this change favored the emergence and maintenance of labiodentals. The findings also suggested that the evolution of language was shaped not only by the contingencies of its history, but also by culturally induced changes in human biology. However, our facility with f-words comes at a cost. As we lost our ancestral edge-to-edge bite, "we got new sounds but maybe it wasn't so great for us," says Moran, one of the coauthors of the Science paper, "Our lower jaws are shorter, we have impacted wisdom teeth, more crowding—and cavities."

Black Death Might Have Transformed Medieval Sub-Saharan African Societies, Afriques

In the 14th century, black death swept across Europe, killing more than half of the population in some areas. However, archaeologists and historians have assumed that the plague bacterium Yersinia pestis, carried by fleas infesting rodents, didn't make it across the Sahara Desert. Albeit the fact that European explorers and colonists frequently visited sub-Saharan Africa during the outbreak of black death, signs that disease might have spread out in Africa could not be found in the recordings of either European explorers or local civilizations.
Now, some researchers point to new evidence from archaeology, history, and genetics to argue that the Black Death likely did sow devastation in medieval sub-Saharan Africa. Historians and archaeologists have found that a number of sub-Saharan civilizations disappeared or significantly declined right around 14th century. One of them was Akrokrowa, a farming community founded around 700 C.E. But sometime in the late 1300s, Akrokrowa and all its settlements were abandoned. Another example was Kirikongo in Burkina Faso. The settlement was founded around 100 C.E. and had been steadily growing for more than 1000 years. Then, sometime in the second half of the 14th century, it suddenly shrank by half without any evidence of food stress, conflict, or migration. Moreover, historians have found previously unknown mentions of epidemics in Ethiopian texts from the 13th to the 15th centuries, including one that killed "such a large number of people that no one was left to bury the dead."
Researchers were looking for genetic evidences to support the historical findings. A 2016 study in Cell Host & Microbe revealed a distinct subgroup of Y. pestis now found only in East and Central Africa is a cousin of one of the strains that devastated Europe in the 14th century. Now, the silver bullet that scientists are dreaming of, is ancient Y. pestis DNA from human remains in sub-Saharan Africa. Although the region's heat and humidity quickly degrade DNA, researchers are turning their attention to DNA in human teeth, where Y. pestis DNA is most likely to be preserved.

ISP Sci. Rev. 18 (2019)
Editor: Shiwei Wang, Rossoneri Jing
Integrated Science Program
Northwestern University

This email was sent to <<Email Address>>
why did I get this?    unsubscribe from this list    update subscription preferences
ISP Science Review · 616 Noyes St · Evanston, IL 60208-0860 · USA

Email Marketing Powered by Mailchimp