In early 2024, laboratory-grown neurospheres (i.e. clusters of neural stem cells) were sent to the International Space Station (ISS) to be exposed to microgravity. The return trip was delayed several times due to bad weather in the landing zone. As a result, the cells were outside controlled culture conditions for longer than anticipated—beyond the limit of what they were expected to survive.

When they finally arrived at the laboratory at the Uppsala Biomedical Center (BMC), no one knew whether it would be possible to complete the experiment.

"When the cells came back, they didn't look normal, but we were still cautiously optimistic. It looked as though there were only small fragments of neurospheres in some groups, but not all of them. Only the cells that had previously been in space on the sounding rocket had survived. So we changed the medium and waited. Then some of the cells recovered and began to grow, very slowly," explains Elena Kozlova.

A month later, the experiments could begin. The results have now been analyzed and are presented in an article in npj Microgravity.

The 3D structure protected the cells
Despite being subjected to severe stress, some of the cells survived. After recovery, they were still able to develop into nerve cells and glial cells and functioned normally in the researchers' tests.

A key finding is that cells placed in 3D-printed support structures fared particularly well across all groups. They survived and continued to divide.

"Thanks to the experiment being extended, we were able to see just how significant the 3D-printed structure was for the cells' survival...

It's well-known that regular exercise comes with health benefits, but exactly how those benefits manifest at the cellular level isn't clear. Now, researchers have identified a series of changes in metabolism and the immune system that differentiate regular exercisers from people with sedentary lifestyles.

The findings, published Sept. 18 in the journal Science Advances, point to potential pathways for managing chronic diseases, such as diabetes, heart disease and cancer, the scientists say.

The team of researchers analyzed data from the Chinese Immune Multi-omics Atlas (CIMA), which includes data from 400 adults who gave blood samples and answered an in-depth questionnaire about their lifestyles.

To capture the biggest possible effects of exercise, the team included only data from participants who did the most or least exercise. One group worked out at least three times a week, racking up more than 150 minutes of moderate exercise or 75 minutes of high-intensity exercise in that timeframe. The sedentary group said they did not work out at all.

The researchers then analyzed their volunteers' blood samples for physiological markers, such as blood sugar and fat levels. They found that the participants who exercised more had signs of boosted sugar and fat metabolism, stronger muscle metabolism, and higher levels of antioxidant molecules, which combat cell-damaging processes.

Many of these systems and their links to exercise were already known, said Dr. Michael Siebers, a researcher at the University of Duisburg-Essen who wasn't involved with the study. The study's real step forward was its subsequent cellular analysis, he said.

"It's a really good paper, which is showing, for the first time, what is really happening when you're doing sports on the intercellular level," he said...

“We used to think of the body as an isolated system that regulates itself,” said Professor Shir Atzil, a researcher at the Hebrew University of Jerusalem. “Our findings show that human physiology is inherently social. Being close to someone you love lowers the physical energy your body spends on regulating itself.”

“This gives us a direct metabolic benefit to being together and reveals a core evolutionary reason why humans form bonds in the first place: our bodies run better together.”

In one experiment, 108 adults ate a standardized carbohydrate meal — first alone, then with their romantic partner — while the researchers tracked their blood glucose for two hours.

Glucose levels rose less and returned to baseline faster when partners were present, producing about a 15% reduction in overall glucose exposure compared with the alone condition.

The scientists then repeated the design with two other physiological systems.

In a cold-exposure test with 116 participants, skin temperature dropped less and recovered more quickly when a partner was nearby.

In a mild-shock stress test with 115 adults, skin conductance — a marker of nervous-system arousal — likewise showed a smaller, shorter-lived response in the social condition.

A related pilot study of 56 mother-infant pairs found the same pattern: infants showed reduced stress responses to a loud noise when their mother was present in the room, compared with a separate-room condition.

Across all three adult experiments, a combined statistical analysis found a consistent effect, with roughly 8 to 12% lower cumulative physiological ‘cost’ in the social condition than when participants faced the same challenges alone.

Professor Atzil and colleagues argue the findings point to something they call Social Physiology, the idea that being near another person doesn’t just buffer stress or provide comfort, but can make routine bodily regulation itself more efficient, even in systems like glucose metabolism that aren’t usually thought of as socially responsive.

Notably, the glucose effect didn’t appear to be explained by reduced stress: markers of physiological arousal didn’t differ between the alone and social conditions during the eating experiment, and the effect held up even after statistically accounting for stress levels and physical touch between partners.

The authors suggest the benefit may instead involve the brain’s predictive regulation of the body — a process called allostasis, in which prior social experience helps the brain anticipate and manage physical demands before they fully unfold.

Sleeping less than five hours a night was associated with a higher risk of dozens of diseases in a large study that tracked nearly 100,000 adults for almost nine years. The analysis also found that people who spent more time in REM and deep sleep tended to have lower risks of several major conditions, suggesting that both sleep duration and sleep structure may be linked to long-term health.

Researchers analyzed 95,559 UK Biobank participants who wore wrist accelerometers continuously for seven days and nights. A deep learning algorithm estimated total sleep time, REM sleep, deep sleep, light sleep, wakefulness after sleep onset, and how much sleep patterns varied from night to night.

More REM Sleep Was Linked to Lower Disease Risk
Greater REM sleep was associated with lower rates of 83 diseases. Each additional 47.6 minutes of REM sleep was linked to a 26 percent lower rate of heart failure, a 46 percent lower rate of dementia, and an 80 percent lower rate of Parkinson’s disease during follow-up.

More deep sleep was associated with reduced risk of seven conditions, including type 2 diabetes and major depressive disorder. By contrast, greater night-to-night sleep irregularity and more wakefulness after falling asleep were associated with higher risks of several conditions, including anxiety and substance use disorders.

The study did not prove that these sleep patterns directly cause or prevent disease. Because the research was observational, other health, behavioral, or biological factors could contribute to the associations.

A Six to Eight Hour Window Stood Out
Total sleep time showed a nonlinear relationship with many health outcomes. For most of the conditions showing this pattern, the lowest risk was concentrated between six and eight hours of sleep per night.

People who slept less than five hours had the greatest overall vulnerability, with higher risks of 37 conditions. Across the full analysis, researchers identified 156 statistically significant associations between sleep patterns and newly diagnosed diseases after correcting for the large number of comparisons...

Margaret Hamilton, a groundbreaking computer scientist who led the team that created the onboard flight software for NASA's moon-bound Apollo program and later in life was celebrated as a powerful symbol of women in technology, has died. She was 90.

Hamilton died Sept. 30 at a hospital in Cambridge, Massachusetts, her daughter, Lauren Hamilton, told The Associated Press on Thursday. NASA Administrator Jared Isaacman said on the social platform X that her work helped make humanity's first journeys to the moon possible.

"To say Margaret Hamilton was a pioneer—to say she was ahead of her time—would be a dramatic understatement," Olivier de Weck, the Apollo program professor at the Massachusetts Institute of Technology and interim head of its aeronautics and astronautics department, told MIT News, a publication of the university's communications office.

"She was a software engineer at a time when that field was in its infancy," de Weck continued, "and she not only developed advanced code herself but also led a team in using that nascent technology to develop one of the most complex systems humanity had ever achieved"...

Could a psychedelic drug help the brain recover from mild injuries? New research suggests that psilocybin, the mind-bending ingredient in magic mushrooms, might protect the brain from the damage caused by minor head impacts.

Even mild head impacts that don't count as concussions can raise the risk of neurodegenerative disorders such as chronic traumatic encephalopathy (CTE), so finding an effective treatment for these blows could be useful for warding off debilitating conditions down the line.

The new research, published in the journal Communications Biology, included only lab rats, so there's more work to be done to translate the findings to humans. But the early results hint at a new application for psilocybin — a drug that's already been explored for a range of other conditions, including depression, PTSD and anorexia.

Live Science spoke with study co-author Argel Aguilar-Valles, an associate professor of neuroscience at Carleton University in Ontario, about the new study and how the research might eventually translate to treatments for repetitive head injuries.

If humanity intends to engage in interstellar exploration in the near future, lightsails are currently the best option. This technology, consisting of laser arrays pushing a reflective sail and tiny spacecraft to relativistic speeds (a fraction of the speed of light), is presently the only promising method for sending missions to the nearest stars within a human lifetime. Alas, several major hurdles need to be overcome and technological developments made before any such missions are possible.

These range from developing space-based laser systems and creating materials that can survive the journey through interstellar space to testing proof of concepts in space. According to recently published results, engineers at Cornell University's Space System Design Studio (SDSS) successfully deployed two lightsail experiments - Alpha CubeSat and Sailing to the Stars - from the International Space Station (ISS).

The missions were created as part of the Museum of Science Fiction’s 2016 CubeSat Competition, a design competition for next-generation spacecraft concepts using off-the-shelf technologies. The experiments were funded by NASA's CubeSat Launch Initiative, which launched to the ISS in late 2025 aboard the NG-23 and Crew-11 missions, respectively. Both consist of origami-style lightsails and chip satellites (ChipSat) that were folded for launch (to conserve room for other payloads) and unfurled upon deployment...

Everything that’s swept the world since the 2022 release of ChatGPT — “the AI economy”; data centers; chatbots acting as best friends, doctors, and mathematicians; talk of existential risk — it all starts, in some sense, with an intern in a University of Montreal lab named Dzmitry Bahdanau.

It was 2014. The 23-year-old Belarusian, who had been coding competitively since he was a kid but hadn’t done academic research before, was asked to help figure out how to get a computer to translate sentences from English to French. Bahdanau was working with artificial neural networks, which had already shown promise in pattern-recognition tasks such as deciphering handwriting and identifying objects in photos. The networks learn to recognize patterns by strengthening and weakening connections between nodes in response to training data. Bahdanau quietly doubted their potential to achieve anything like human-level intelligence, but his adviser, the computer scientist Yoshua Bengio, and the postdoc on the project, Kyunghyun Cho, were believers. And translation was widely seen as a formidable and interesting problem for neural networks to try to solve.

Translation is not as simple as swapping each English word for a French one. Languages have different grammatical structures and are riddled with idioms and connotations, all of which had long stymied attempts to automate the process. You can’t simply implement rules of translation because there are none. But neural networks seemed promising because they learn by example. They can be trained on huge databases of paired English and French sentences to, eventually, absorb the quasi-rules, exceptions, and subtleties.

Results had been mediocre, though; the longer the sentence, the worse neural networks performed. That’s because researchers were trying to cram each sentence’s data into a fixed amount of memory before decoding the sentence in the target language. It was a bottleneck that limited how complex and sophisticated translations could be. The network architecture just wasn’t right.

So Bahdanau tried something different. “It was just obviously the right thing to do — one of these things that happens once in a while in science where it’s just very clear to someone who is not very experienced in the field,” he recalled, speaking to me from Montreal, where he still works in AI. “And it just worked. From the very first attempt.”

Bahdanau’s trick, a mathematical operation that he, Cho, and Bengio referred to as “attention,” enabled their neural network to translate long sentences just as well as short ones. That meant the approach was scalable. Other AI researchers at the time quickly integrated attention into their neural network designs. Then, in 2017, a team at Google devised a new kind of neural network architecture called a transformer that supercharged the attention method and stripped away everything else. The team’s paper, “Attention Is All You Need,” reported that transformers achieved state-of-the-art translation while requiring a fraction of the computation needed to train prior models. The paper, which has since been cited more than 250,000 times, was a watershed moment in AI. Large language models such as ChatGPT and Claude followed.

Given the staggering power that’s been unleashed, it’s worth asking what this attention mechanism is exactly, why it is “all you need,” and why machine translation was the task that unlocked it all. Attention is a way for us to understand how a single AI model can converse with us, diagnose diseases, generate realistic videos, solve long-standing math problems, and more. Is it somehow at the heart of intelligence — artificial and otherwise?