【推荐1】Willpower might be key to getting off the couch to exercise, but bacteria may lend a helping hand. Studies in mice suggest microbes(微生物) in the gut(肠道) may be behind differences in the desire to work out.

To explore why some people like to exercise and others don’t, microbiologist Christoph Thaiss studied mice bred to have a lot of genetic and behavioral variation. His team found huge difference in how far the mice ran on wheels in their cages.

The active and lazy mice didn’t show any significant differences in their genetics. But the researchers did notice one clue: When treated with antibiotics(抗生素), mice that were normally highly energetic tended to exercise less. Follow-up studies showed the antibiotic treatment affected the brains of the formerly active mice. The activity of certain brain genes declined, along with levels of dopamine(多巴胺), a material linked to “runner’s high”—the sense of well-being that comes with more exercise.

The team also found that “bacteria-free” mice, which lack gut bacteria, become more active when given some of the gut microbes from energetic mice. It appears those bacteria send a signal that interferes with a material responsible for breaking down dopamine in the brain, causing the material to build up in the brain’s reward center.

Will the finding hold in people? Studies have found that marathoners have high levels of a particular gut microbe, suggesting an exercise connection in people. And much work has proved the key role dopamine plays in motivating overall behavior. As Christoph Thaiss puts it, “This reward system is such a central aspect of physiology that it’s something almost certainly true in other mammals”—including humans.

【推荐2】Road traffic accidents are caused by a number of factors which can be grouped into three categories: human factors, vehicle factors and road environment factors. Table 1 explains each factor and gives examples.

One factor alone or several factors together can cause an accident. However, human actions are by far the greatest causes of car accidents. Diagram 1 shows that human factors are involved in 95% of all car accidents. Although other factors are also involved in 28% of these accidents, in 67% of cases, human factors alone are the cause of the accident.

Although behavior such as not paying attention to road, swerving to avoid an animal and eating while driving causes accidents, most road deaths are caused by speeding, driving after drinking alcohol, driving when tried and not wearing seat belts. As shown in Diagram 2, speeding is involved in more deaths than any other behavior in both city (50%) and country accidents (32%). Drivers and / or passengers not wearing seat belts is the second most common cause of road deaths.

【推荐1】Can humans really understand what animals are saying, or are we just barking up the wrong tree?

“Artificial intelligence (AI) holds the key to unlocking the fascinating insights. Beyond creating chatbots that charm people, machine learning may soon make it possible to decode (解码) animal calls,” says Raskin, co-founder of the nonprofit Earth Species Project. It’s developing AI models that imitate a variety of species, aiming to have “conversations” with animals. Its team is collecting diverse data from various species and building machine-learning models for analysis. Project CETI (鲸类动物翻译计划) focuses on understanding a particular species, in this case the sperm whale.

Sperm whales have complex social groups. When a familiar young male rejoined his family, researchers seized the chance to record their sounds. For two decades, scientists documented two sperm whale groups, capturing their clicking sounds. After manually decoding some sounds, they turned to AI for faster translation. Using a neural (神经的) network, the team trained it to discern individual whales from a subset of sounds. Next, their ambitious goal is to train a computer to speak whale.

As tech advances, the door has been opened to using machine learning to decode unfamiliar languages. It’s found that AI tools have practical value beyond research. Translating animal sounds aids endangered species. Scientists study caged birds’ calls to grasp sound changes, explaining difficulties in reintroduction. Machine learning decodes pets’ signals like barks and facial expressions. Raskin’s AI model translates dogs’ expressions, revealing their surprising capabilities to owners. Additionally, it helps predict pig emotions based on their sounds, enhancing animal welfare. Advancements deepen our understanding of animals.

Now enthusiastic scientists are committed to open-source data and model sharing. “Every time you invent a technology, you also invent a responsibility,” Raskin says. “Designing a ‘whale chatbot’ demands we imagine an animal’s experience. The true value of any language is that it helps us relate to others.”

【推荐2】Ask anyone who lives in an apartment in a packed city and they'll likely tell you space is among their biggest problems. Those living in even smaller rooms know the headache all too well. Some even pay money each month for storage as a result just to make more room.

Boston-based company Ori is out to solve those problems with a system that uses robotics and apartment furniture to change small spaces in seconds.

The company's new system is also called Ori, which is named for the Japanese word "origami", the art of folding (折叠) paper. The new system makes your furniture change your small space from a living room into a bedroom and even a workspace with the push of a button (按钮).

Ori came from MIT's Media Lab, which works to help companies to create a better future for people. The company is working with developers Boston Properties, Skanska and Samuels Associates to put the furniture in high-end apartments in Boston. It is working with designer Yves Béhar on the furniture designs.

The Boston models are being rented (租用) on Airbnb, an online company offering people rentals, so Ori can collect information about what users think of the system and improve it in the future. The system will be put on the market in the near future, said Larrea, founder and CEO Hasier Larrea.

All of Ori's product (产品) is made in the US, and Larrea said the furniture is actually more expensive to make than the robotics. Assembly (装配) is simple — the unit can be put together in just a day, Larrea explained. The most difficult thing for this MIT team was making the system feel and look personal.

"We have created these amazing robots with amazing abilities," he said. "But when you bring a system into a home, it can't look like a robot."