Many articles this month, ranging from genuinely exciting stuff to just cool news.
This most significant piece of news comes late, being a claim of the first human gene-edited babies being born. As this comes from China there are plenty of skeptics but it seems sound enough. The claim is that fertilized embryos were edited to disable a gene known as CCR5 which can act as a protein pathway for HIV infection. The idea is that this can help to improve resistance to the disease. The most important bit seems to me is that the edited embryos were implanted back into the mother and successfully carried to term. The science isn’t that new but the ethical implications are more important.
Next up is another piece of news that is also more interesting due to how it came about than the actual result. In this case, a deep learning algorithm was used to analyze images of the retina, and it found a correlation between the images and cardiovascular risk factors. In essence, it might be possible to predict such risks by looking at patients’ retinas, an association that most likely would not occur to doctors and could not have been found using means other than an AI trawling through a huge trove of data.
Next is a depressing finding about how mobs nominate members of a group to be victims. They discovered from a simulated mobbing game that even in exchange for very modest gains, members will not only single out individuals to be victimized but will coordinate with one another to ensure that only a single person is targeted at a time as per the rules. There is no sense of pity as the group will repeatedly target the same victim if that is what the group agrees on and fear of being the next victim does not seem to dissuade the group.
Moving on to lighter stuff, here’s a cool post speculating about how the unknown interstellar object that has entered our solar system could be a piece of a solar sail from an alien civilization. The object called Oumuamua is thought to be more than just a piece of space rock because its movement shows signs of acceleration that is not due to gravitational forces plus it has no signs of emitting a tail that would indicate a chemical reaction creating that acceleration. Since it doesn’t appear to be a comet, it may be that the acceleration is due to solar radiation, suggesting that it may be a piece of solar sail due to having the right physical properties. Of course, it’s all pure speculation but it’s fun to think about.
Finally here’s an article about how science-fiction author Greg Egan contributed a partial solution to a mathematics puzzle. It’s an entertaining read because apart from Egan’s minor celebrity status there’s the fact that the puzzle are originally posed in 2011 as a question of the ordering of episodes of a television show. Shortly after that an anonymous poster submitted a lower bound to the solution but it was until recently that Egan offered an upper bound. The puzzle and the solution itself, part of something that Egan calls superpermutations, is of interest only to mathematicians, but I love the whole story of how this came about.
I’d thought it would be a quiet month on the science front due to the Nobel Prize announcements but there’s a decent mix of stuff.
Probably the bit of news that has been most talked about this month in academic circles is the attempt to replicate the famous Sokal hoax. Three people collaborated on a year-long project to write 21 bogus papers and submitted them for publication. Seven of these were accepted. The papers were deliberately outlandish, such as one about dog on dog sexual assault, and the idea is to mock the preposterousness of some fields of humanities. Yet as this article that I chose to highlight lays out, the joke seems more on the three than the field in general and their political motivations are obvious, such that this needn’t be taken seriously at all.
To balance out the above, next is a respectable paper on gender studies with a result that I find legitimate but may annoy some feminists. Based on survey data, the authors find that contrary to some predictions, the more that women have equal opportunities, the more skewed their participation in different occupations may be because they have more freedom to enter fields that they prefer. The data showed that the more developed the country, the more women expressed different preferences from men in their willingness to perform different jobs.
Staying on the topic of gender but in a different way, how could I avoid covering the news about baby mice being born to two female parents. The team used the CRISPR technique to ensure that the full set of genes that comes from a female mouse are switched on and injected the edited stem cells into the egg of a second female mouse to form an embryo. The resulting animals were born healthy and were even able to have babies of their own in the normal way. Of course, this is an artificial, invasive technique so it’s not as if males are completely obsolete just yet.
The next paper we cover has a pretty obscure title: voluntary arousing negative experiences (VANE) but its subtitle says it all: it purports to explain why people like to be scared. The researchers surveyed visitors to a “haunted” attraction and measured their EEG reactivity to cognitive and emotional tasks. They found that going through the scary experience improved the mood of the participants, especially if they were feeling tired, bored or stressed beforehand, and could help them cope with stress afterwards.
Finally, here’s a longer read, a retrospective of the Spanish flu pandemic that the article argues was the worst catastrophe of the 20th century, killing at least 50 million people and perhaps as many as 100 million. Yet it is almost forgotten today except by historians and health experts. The article covers its history and notes that the virus could not have evolved as it did without the trench warfare conditions of the First World War that, due to so many young men forced to live so close together in unsanitary, dangerous conditions, made it the perfect incubation environment to create a uniquely virulent form of the flu.
It’s that time of the year again when I do a round-up of the winners of the scientific Nobel Prizes. If nothing else, writing these posts helps me remember and understand what they are for. This task is a little more difficult this year as the prizes for so many categories are split into two fields that aren’t really closely related. But let’s get on with it.
We’ll begin with the prize for Physiology or Medicine as it’s closest to being the same research topic. It goes to James P. Allison and Tasuku Honjo for for their discovery of cancer therapy by inhibition of negative immune regulation. Researchers have long known the immune system, in particular a type of white blood cells known as T-cells, are regulated by accelerators and inhibitors to ensure that they attack only foreign microorganisms. Allison studied the T-cell protein CTLA-4 which serves as a brake and wondered if using an antibody to block its function would release the immune system to attack cancer cells. Indeed it worked and would eventually be developed into a new treatment for skin cancer.
Not much of note this month. We’ll start with the exciting space news and no, it’s not the one about the tourist going for Elon Musk’s moonshot.
Instead, it’s the strangely under-reported announcement by JAXA, Japan’s Space Agency that two rovers launched by its Hayabusa2 spacecraft have successfully landed on an asteroid named Ryugu. The rovers themselves are only 18 cm in diameter each and Ryugu is a rock with a diameter of about 1 km Yet the interception took place about 300 million km away from Earth and to do so Hayabusa2 made a circuitous journey of over 3,200 million km over four years, making this an incredible achievement in precision and control.
As always, drawing general conclusions about life outcomes from data should always be seen with a skeptical eye but this study about predicting income is sure made for the headlines. Crunching through a trove of data that covers nearly three thousand participants, it purports to show that delay discounting predicts income better than other factors such as age, ethnicity or height. This refers to the ability to delay immediate gratification in exchange for a bigger future reward. The result is as expected but I notice that this ability in the present and not when the participants were children so I’m not sure if it means that much.
The last article is a very speculative one about a new take on the well known Schrödinger’s cat thought experiment. Instead of a cat, this scenario uses two physicists, each of whom performs an experiment on a friend that they keep in a box. One friend in a box can toss a coin and send a message to the other friend in a box who can then guess the result of the toss. Each experimenter can then open each box to conclude which side the coin landed on and yet in some circumstances the results will be inconsistent, meaning that the model is wrong or reality itself is somehow inconsistent. Apparently the physics community is still divided over what to make of this experiment though it is currently impossible to actually carry it out.
A huge haul this month, so much so that I’ve had to be a bit more selective than usual. Most of it is once again in biology.
Most of us know that epigenetic inheritance isn’t a thing and that genes are passed down to the next generation. This article talks about how this isn’t necessarily the case and how fathers may pass down information that isn’t encoded in DNA to offspring through a mechanism that makes use of small RNA. The small DNAs appear to influenced by the state of the father’s health and has the ability to conceal genetic information from the cellular machinery, effectively cancelling out some genes.
Next we have one about how what we see is distorted by what we expect. This isn’t exactly news these days but it’s a pretty cool experiment that demonstrates how people naturally expect goal-directed, efficient actions and interpret what they see based on those assumptions even when that this is not what they actually saw.
A lighter bit of news is this paper about measuring how much cows value access to a grooming brush. The brush in question was an automatically rotating brush and the researchers mounted it in an area that required the cows to push against an obstacle to access the area. By altering the energy needed for the push, they found that the cows were willing to work about as hard to get to brush as to a supply of feed. But they were not willing to push to get to an empty enclosure.
Probably one of the more important papers this month is about how personality affects salaries. This particular study is based on school data from California and covers only the smartest students. Using the standard Big Five measure of personality, it found that high conscientiousness and extraversion are correlated with much higher lifetime earnings while high agreeableness are negatively correlated with earnings.
I like this simple story about an NGO working on providing reading glasses to people in poor countries. It’s absolutely trivial, basic stuff but it’s a good example of how much of a difference such work on the ground makes.
Studies about the benefits or harm of different types of foods always seem to swing back and forth so this is likely just another data point to add to the pile. This one is about alcohol and it posits that there is no health level of alcohol consumption. The study found that although there are some benefits from low levels of consumption particularly in protecting against heart disease, these are still more than outweighed by health risks such as its role in causing cancers.
Finally just to have a bit of variety, here’s a bit of news on the materials science front. The claim is that they’ve made the most wear-resistant metal in the world made out of a platinum-gold alloy which sounds incredibly expensive to make. To me, the interesting bit is how it works. When being worn down, the material generated a black film on its surface that has the effect of acting as a lubricant, protecting it from further erosion. The engineers identified it as a diamond-like carbon, so in a sense this metal generates diamonds when you rub it hard enough.
A whole boatload of stuff this month plus they’re from a nice range of domains too. I’ll try to be brief with each of them.
The big economics article of the lot is the much talked about one about how secularisation appears to precede economic development but only when it is accompanied by increased respect for individual rights. As the authors note, this doesn’t prove causation but you can see why it’s been a hot topic.
Here’s another that touches on the topic of sunk costs which has long been known to affect human judgments of value. According to new research, this sensitivity affects mice and rats as well, leading them to continue performing an apparently unproductive activity after they have already invested time and effort in it.
If you’ve paid attention to the news, you’d have already heard about the discovery of a large body of liquid water beneath the surface of Mars, sparking the usual excitement about finding life on the planet.
A corollary to that might be this announcement about how roundworms that have been frozen for 40,000 years in the Arctic permafrost have revived after being dug up by Russian scientists and simply placed into Petri dishes with a nutrient medium. There’s still a chance that they came from contamination and not the samples but if it pans out, it’s by far the longest period organisms have been revived from cryogenic preservation.
The next article presents a real time view of natural selection based on a population of anole lizards in the Caribbean after a devastating hurricane season. Examining the population before and after the hurricanes, the researchers found that the average measurements of the lizards’ toe pads and lengths of their front and back legs had changed simply because the storms had killed off the lizards with legs and feet less adapted to surviving those conditions. Going forward it can be presumed that breeding will lead to a lasting change of this type in future generations.
Moving on to computer science, this difficult to read paper talks about how quantum computers may be less in thrall to the arrow of time than traditional computers. By this, it is meant that certain computation operations have a natural order in which it is easier to perform the operations in one direction but much more difficult or impossible to do it in the reverse direction. Ths paper claims that switching the same computation to a quantum mechanics model and performing them with a quantum computer eliminates or at least reduces the overhead of going in the opposite direction.
My favorite paper of the lot however is this one which takes direct aim at all of the excitement around machine learning and AI. It argues that neural net models directly correspond to the polynomial regression models already well known in mathematics and that solving the same problems with the latter is both easier and more precise.
Once again, biology dominates the cool science news this past month.
We start with a report on an attempt to replicate the marshmallow test, a frequently cited study about how a child’s ability to delay gratification is correlated with academic achievement. The new experiment however found that the correlations were smaller than previously reported and were further reduced after controlling for factors like family background and home environment. Their conclusion is that associations between delay time and measures of behavioral outcome by the age 15 are statistically insignificant.
Probably the most anatomically distinctive feature of humans is our disproportionately large brains. A couple of studies have zeroed in on the cause of this as being a gene known as NOTCH2NL. The presence of this gene leads the brain to develop more neurons and hence a larger brain and the researchers were able to estimate that it became active roughly three to four million years ago which was when the brains of our ancestors started expanding.
The Flynn Effect is the well known phenomenon of human IQ scores steadily improving throughout the 20th century. This article talks about the effect seems to have ended and perhaps even gone into reverse in more recent years. Looking at results from three decades of Norwegian conscripted into the military, they found IQ scores falling over time, even between members of the same family tree. While there are plenty of offered suggestions as what could be causing this, all of this is speculation for now.
Next is a sad story of how many baobab trees seem to be either dead or are dying. These are some of the oldest trees in the world, aged between 1,100 and 2,500 and are known to be able to live to some 3,000 years, all in southern Africa. It is not known why they are dying but the cause is suspected to be related to climate change.
Finally just for the pleasure of reading, here’s an in-depth overview of the underwater tunnels that connect the various Faroe Islands halfway between Iceland and Denmark. It seems strange given the isolation of the islands and their very low population but the tunnels have been a success, both from an engineering perspective and economically as well. This article talks about the history of these tunnels, how they made it all work and ends with the current proposal to construct a huge underwater tunnel to connect the southernmost island to the mainland.