Research

A study by MIT and Harvard researchers has found that sleep onset, the transitional period from a woozy but still awake state into sleep, has a strong effect on creativity, reports Alan Mozes for HealthDay.

Science reporter Sofia Moutinho spotlights how MIT researchers used a glove that tracks sleep stages to guide people’s dreams while they snoozed. Many participants who considered themselves “stuck and uncreative” were surprised at how inventive they could be in their dreams, explains postdoc Adam Haar Horowitz. “Most people don’t know that there’s a piece of themselves that is biologically designed to be totally unstuck, but they’re forgetting it every night.”

Principal Research Scientist Ana Jaklenec speaks with CBC host Bob McDonald about her work developing a mobile vaccine printer. The device “can be very important in certain scenarios when you’re trying to bring the ability to vaccinate in areas that might not have the right infrastructure to make vaccines or even to administer vaccines,” says Jaklenec, “so I think the portability is key here.” 

Researchers from MIT and elsewhere have found chance encounters among employees of different companies can kickstart innovation, reports Bart Ziegler for The Wall Street Journal. Researchers explained that such chance meetings “may spark a conversation that leads to a transfer of knowledge or a collaboration,” writes Ziegler.

A study conducted by graduate student Aspen Hopkins and colleagues trained a version of a GPT neural network on the board game Othello “by feeding in long sequences of move in text form”, reports George Musser for Scientific American. “Their model became a nearly perfect player,” writes Musser.

A new working paper by Prof. Christian Wolf and his colleagues explores a “mechanism by which a government could run deficits and never have to pay them,” reports Peter Coy for The New York Times. The researchers found that “‘deficits contribute to their own financing via two channels.’ First, they can accelerate economic growth, which generates more tax revenue. Second, they can cause inflation to rise, which shrinks the effective cost of debt.

MIT researchers have found that interactions between people from different economic backgrounds have dropped significantly since the start of the Covid-19 pandemic, reports David Scharfenberg for The Boston Globe. Scharfenberg notes the “the phenomenon could hurt low-income people in direct ways – they’ll lose connections to better-off people – and indirect ways.”

Prof. Jongyoon Han and research scientist Junghyo Yoon have developed a new portable desalination device that can deliver safe drinking water at the push of a button, reports Meghan Gunn and Kerri Anne Renzulli for Newsweek. The device “requires less power than a cell phone charger to run and produces clean drinking water that exceeds World Health Organization standards,” writes Gunn and Renzulli.

Prof. Benedetto Marelli and his colleagues have created “packaging that can react to changes in the food it contains to better indicate when it has gone bad,” reports Karmela Padavic-Callaghan for New Scientist. The biodegradable plastic-like wrap, which is made from silk, changes color when it is exposed to rotting foods and degrades quickly in soil. 

Prof. James Poterba, president of the National Bureau of Economic Research and a member on the Business Cycle Dating Committee, speaks with NBC Boston reporter Annie Nova about the practice of dating economic downturns. Poterba notes that examining economic fluctuations “helps to design policy going forward. It enables us to look back and say, for example, what are the consequences of interest rate increases?”

Wired reporter Caitlin Harrington writes that a study by researchers from MIT and Stanford highlights the impact of generative AI tools on workers and raises a “provocative new question: Should the top workers whose chats trained the bot be compensated?”

Researchers from MIT have found that “although women received higher performance ratings than their male colleagues, they received 8.3% lower ratings for potential than men,” reports Caroline Castrillon for Forbes. “Because those ratings strongly predict promotions, female employees were 14% less likely to be promoted than male ones,” writes Castrillon.

Soledad O’Brien spotlights how researchers from MIT and Massachusetts General Hospital developed a new artificial intelligence tool, called Sybil, that an accurately predict a patient’s risk of developing lung cancer. “Sybil predicted with 86 to 94 percent accuracy whether a patient would develop lung cancer within a year,” says O’Brien.

Astronomers from MIT and elsewhere have become the first to witness a star consume an entire planet, reports Elisha Sauers for Mashable. “The new study confirms that when a sun-like star nears the end of its life, it expands into a red giant, 100 to 1,000 times its original size, eventually overtaking nearby planets,” explains Sauers. “Such events are thought to be rare, occurring only a few times each year throughout the galaxy.”

Researchers from MIT have found that Twitter “bot detection tools can rely on funky, flawed data sets that replicate mistakes made within one another, rather than trying to accurately identify bots,” reports Chris Stokel-Walker for Fast Company. “We realized that this was a systemic issue in the data sets that are commonly used for bot detection,” says postdoc Zachary Schutzman.