stories which we suspect may be of more than passing significance
A hundred years ago the most exciting work being done-where the future was being shaped, if you like-was in quantum mechanics. Nowadays, we think the future is being shaped by those researching things like Artificial Intelligence, Complexity Theory, and the general need to cope with the huge quantities of data which we now possess. Nowhere is this clearer than in the world of Meta-science; that exciting world where researchers look at results across many studies, scaping out new significances and the precious gold of new knowledge. And one real opportunity from the new technologies is the potential for multiple statistical analyses of data sets. About time too, as this study in Nature makes clear (“one-note analysis can give false confidence”)
Restricting analyses to a single technique can blind researchers to an important aspect of uncertainty, making results seem more precise than they really are. For example, in 2020, the UK Scientific Pandemic Influenza Group on Modelling asked nine teams to calculate the reproduction number R for COVID-19 infections. The uncertainty across all the teams was considerably larger than the uncertainty within any one team — important knowledge for decision-makers. Meta-science researcher Balazs Aczel and statisticians Eric-Jan Wagenmakers and Alexandra Sarafoglou argue that data should be subject to multiple analyses, ideally by multiple teams — and that the extra work is worth it.Nature | 10 min read
Magic Mushrooms and depression The relentless persecution of recreational drug users may have led to a missed trick- a cure for the terrible affliction of depression. A more relaxed attitude could lead to some fruitful research into cures. Here’s Claire Tweedy for The Conversation:
China Crisis We have always profoundly admired Chinese Civilisation. It seemed only fair that they should take a rightful place in the world, although in recent decades their rise has been vertiginous-and slightly scary. Yet all is not necessarily well for this Asian Colossus as this piece by Nectar Gan for CNN, suggests we thank Peter Seymour for this link
Homage to a pioneer It wasn’t easy to be a woman in the nineteenth century and to be a woman scientist was many orders of magnitude more difficult. So we were heartened to see that a statue of paleontology pioneer Mary Anning is to be raised in her native Dorset. Surely no dinosaur eggs will be hurled at this one? Esther Addley for the Guardian:
Sports Algorithm Finally, as Saturday is a big sports day, here’s another use of new algorithmic techniques to predict sports injuries, says Sam Tonkin for the Mail
If we cleave to the exciting possibilities offered by new technologies, and have the courage to use our brains, the future can be better indeed. Drifting into boozy, nostalgic fantasies of lost national greatness offers no real future indeed. As more than one country is currently finding out.
#metascience #algorithms #AI #mental illness
Learning, Science and Society 