1 . Borders, departments, or issue areas all represent what systems analysts call system boundaries. System boundaries divide the big, messy, interconnected world into smaller subsystems. This is useful, even necessary. Our minds and our collective governance systems would be stuck if we had to always consider all the connections of everything to everything else. But dividing systems into subsystems can sometimes break a natural cooperativity. For instance, a decarbonizing country will spend money in its energy and transportation sectors and save money in its health system.

Decarbonization could be a win for the whole, but it might be experienced as a bother for particular subsystems.

Donella Meadows, the early systems modeler, wrote that system boundaries are “lines in the mind, not in the world.” And that is actually good news. If departments, and disciplines are just ideas, then there is nothing immovable about them. We can make these borders less obvious and conduct partnerships across them. We can even redraw them to include more of what matters in a single project or investment. That’s the premise of multisolving — using one investment of time or effort to achieve several goals at once.

For instance, Warm Up New Zealand (WUNZ) upgraded the energy efficiency of residential buildings and provided jobs in the building sector after a financial downturn. The project resulted in better health for residents, as well. That translated into health systems savings. Taken together, a 2011 study estimated that across all these benefits, the project saved $3.90 for every $1 invested.

Multisolving seems possible everywhere and like an obvious choice. Yet, it is very much the exception, not the rule. Why is multisolving still so rare when it has the power to boost progress on some of the most urgent issues we face?

Unfamiliarity stands in the way, as does an often-unexamined assumption that making issues smaller makes them easier to address. We often hear the viewpoint, “I already work on poverty (or climate, etc.) and that’s hard enough. Why should I add biodiversity or pollution to the mix?” Fundraising for crossing borders can be a struggle too. Funders want the “visible results” shown, but they don’t always see crossing borders as an essential part of achieving those results.

It is easy to devalue and underemphasize connection-building. After all, it can be delicate and not always visible. But to realize goals in today’s world, from equitable policies and low-carbon facilities to values like cooperation and fairness, we do need deep shifts, and we need them soon. And facilitating the flow of ideas back and forth across borders is one way to speed change.

2 . Whenever anyone asks me what tech I’d like to see invented, I always say the universal translator, which lets you understand and speak any language.

Meta AI recently announced the start of the universal speech translator (UST) project, which aims to create AI systems that enable real-time speech-to-speech translation across all languages, even those that are spoken but not commonly written. Meta says that today’s AI translation models are focused on widely-used written languages, and that more than 40% of primarily spoken languages are not covered by such translation technologies.

According to Meta, the model is the first AI-powered speech translation system for the unwritten language Hokkien (闽南语), a Chinese language spoken in southeastern China. The system allows Hokkien speakers to hold conversations with English speakers, a significant step toward bringing people together wherever they are located.

To build UST, Meta AI focused on overcoming three important translation system challenges. It addressed data scarcity by getting more training data in more languages and finding new ways to use the data it had found. It solved the modeling problems that arise as models grow to serve many more languages. And it sought new ways to improve on its results.

Meta AI claims that the techniques it pioneered with Hokkien can be extended to many other unwritten languages—and eventually work in real time. For this purpose, Meta has released the Speech Matrix, a large collection of speech-to-speech translations, which enables other research teams to create translation models for other languages.

Artificial (人工的) speech translation could play a significant role in our world. For interactions, it will enable people from around the world to communicate with each other more smoothly, making the social net more interconnected.   For content, using artificial speech translation allows you to easily localize content.

Yashar Behzadi, CEO and founder of Synthesis AI, believes that technology needs to enable more natural experiences if the digital world is to succeed.   He says that one of the current challenges for UST models is the computationally expensive training that’s needed because of the wide range and very slight differences in meaning or sound of languages. Also, to train strong AI models requires vast amounts of typical data. A significant bottleneck to building these AI models in the near future will be to ensure training data collect the privacy in agreement with rules and law.

4 . Think of the words in your head: that tasteless joke you wisely kept to yourself at dinner; your unvoiced impression of your best friend’s new partner. Now imagine that someone could listen in.

Recently, scientists from the University of Texas, have made another step in that direction. In a study published in Neuroscience, the team showed it was possible to read people’s thoughts with a non-invasive brain scanner called fMRI and large language models (LLMs) built with GPT.

The study centered on three subjects, who lay in an fMRI scanner recording their brain activity by detecting changes in blood flow in parts of their brains while they listened to online stories. By integrating this information with the LLMs’ ability to understand how words relate to one another, the researchers developed an encoded (编码的) map of how each individual’s brain responds to different words. Then, the team worked backward. They recorded the fMRI activity while the participants listened to a new story. Using a combination of the patterns previously encoded for each individual and LLMs, the researchers attempted to translate this new brain activity.

While many of the sentences it produced were inaccurate, the decoder generated sentences that got the main idea of what the person was thinking. For instance, when a person heard, “I don’t have my driver’s license yet,” the decoder spat out, “She has not even started to learn to drive yet.” Alex Huth from the university said, “We were shocked and impressed that this worked as well as it does.”

The researchers also found that the technology isn’t one-size-fits-all. Each decoder was quite personalized and worked only for the person whose brain data had helped build it. Additionally, a person had to voluntarily cooperate for the decoder to identify ideas. If a person wasn’t paying attention to an audio story, the decoder couldn’t pick that story up from brain signals.

While the technology was still far from perfect, the result could ultimately lead to seamless devices that help people who can’t talk or otherwise communicate easily. However, the research also raises privacy concerns about unwelcome neural overhearing. The team said the potential of the technology was such that policymakers should proactively address how it can be legally used. Jerry Tang from the team said, “Nobody’s brain should be decoded without their permission. If one day it does become possible to get accurate decoding without a person’s will, we’ll have a regulatory foundation in place.”

5 . The need for clarity extends beyond how we communicate science to how we evaluate it. Who can really define stock phrases such as ‘a significant contribution to research’? Or understand what ‘high impact’ or ‘world-class’ mean? Scientists demand that institutions should be clear about their criteria and consider all scholarly outputs—preprints, code, data, peer review, teaching, mentoring and so on.

My view about the practices in research assessment is that most assessment guidelines permit sliding standards: instead of clearly defined terms, they give us feel-good slogans that lack any fixed meaning. Facing the problem will get us much of the way towards a solution.

Broad language increases room for misunderstanding. ‘High impact’ can be code for where research is published. Or it can mean the effect that research has had on its field, or on society locally or globally—often very different things. Yet confusion is the least of the problems. Words such as ‘world-class’ and ‘excellent’ allow assessors to vary comparisons depending on whose work they are assessing. Academia(学术界) cannot be a fair and reasonable system if standards change depending on whom we are evaluating. Unconscious bias(偏见) associated with factors such as a researcher’s gender, ethnic origin and social background helps the academic injustice continue. It was only with double-blind review of research proposals that women finally got fair access to the Hubble Space Telescope.

Many strategies exist to improve fairness in academia, but conceptual clarity is paramount. Being clear about how specific qualities are valued leads assessors to think critically about whether those qualities are truly being considered. Achieving that conceptual clarity requires discussion with faculties, staff and students: hours and hours of it. The University Medical Center Utrecht in the Netherlands, for example, held a series of conversations, each involving 20-60 researchers, and then spent another year revising its research assessment policies to recognize social impacts.

Frank conversations about what is valued in a particular context, or at a specific institution, are an essential first step in developing concrete recommendations. Although ambiguous(模棱两可的) terms, for instance ‘world-class’ and ‘significant’, are a barrier when performing assessments, university administrators have said that they rely on flexible language to make room to reward a variety of contributions. So it makes sense that more specific language in review and promotion must be able to accommodate varied outputs, outcomes and impacts of scholarly work.

Setting specific standards will be tough. It will be inviting to fall back on the misleading standards such as impact factors, or on ambiguous terms that can be agreed to by everyone but applied wisely by no one. It is too early to know what those standards will be or how much they will vary, but the right discussions are starting to happen. They must continue.

6 . In the story A Christmas Carol, the wealthy miser (吝啬鬼) Ebenezer Scrooge has a magical, life-changing epiphany (顿悟). Scrooge’s eyes are opened as to how his behavior affects other people — and he goes from a selfish grump to a generous benefactor overnight, thanks to visits from ghosts.

Scrooge’s transformation comes down to knowledge. But do people really want to know how their actions affect others? As behavioral scientists, we wanted to understand just how common willful ignorance is — as well as why people engage in it.

Experiments were carried out to find answers. Researchers asked one member of each pair to choose between two options (选择) in one of two settings, determining the earnings for themselves and their partner.

In the transparent setting, if they chose $5 for themselves, they knew their partner would also receive $5. If, however, they chose $6 for themselves, they knew their partner would receive only $1 in return.

In the ambiguous setting, there were two possible situations. In one, if the decision-maker selected $6 for themselves, their partner would receive $1, and if the decision-maker chose $5, their partner would receive $5. But in the other situation, the decision-maker could pick $6 and their partner would receive $5, or the decision-maker could select $5 and their partner would receive $1. The decision-maker knew these two systems — but they were not initially aware of which situation they were in. Interestingly, the decision-maker had the opportunity to resolve that ambiguity by clicking a button.

Across all studies, we found in the transparent setting 55% chose the altruistic option. In the ambiguous setting, however, 40% of participants chose to remain ignorant. 60% of people in the ignorant group chose a higher personal payout in situations where this choice came at the expense of their partner. Among those who requested more information, 36% knowingly kept a higher payout at a cost to their partner. Only 39% of people in the ambiguous setting made the choice that ultimately benefited their partner — a significant drop from 55% in the transparent condition.

But how do we know if ignorance in the ambiguous setting was willful? We conducted a second analysis focused on what motivates people to seek information.

In this analysis we looked at how people who obtained additional information behaved in comparison with those who were given information. We found that people who chose to receive information in the ambiguous setting were seven percentage points more likely to make the altruistic choice than people in the transparent setting. By the same token, the finding also suggests ignorance prevents people from knowing how their actions harm others.

If we can avoid putting a strong moral emphasis on decisions, it may make people feel less threatened and, as a result, less willfully ignorant. We may not have Dickensian ghosts to guide us — but there are still steps we can take.

7 . Supermarkets have long been suffering as one of the thinnest-margined businesses in existence and one of the least-looked-forward-to places to work or visit. For more than a decade, they have been under attack from e-commerce giants, blamed for making Americans fat, and accused of contributing to climate change.

Supermarkets can technically be defined as giants housing 15,000 to 60,000 different products. The revolutionary idea of a self-service grocery, where people could hunt and gather food from aisles rather than asking a clerk to fetch items from behind a counter, first came about in America. There is some debate about which was the very first, but over the years a consensus has built around King Kullen Supermarket, founded in New York in 1930.

For some 300 years, Americans had fed themselves from small stores and public markets. Shopping for food involved mud, noisy chickens, clouds of flies, nasty smells, bargaining, and getting short-changed. The supermarket imitated the Fordist factory, with its emphasis on efficiency and standardization, and reimagined it as a place to buy food. Supermarkets may not feel cutting-edge now, but they were a revolution in distribution at the time. They were such strange marvels that, on her first official state visit to the United States in 1957, Queen Elizabeth Ⅱ insisted on an impromptu (即兴的) tour of a suburban-Maryland Giant Food.

The typical supermarket layout has barely changed over the past 90 years. Most stores open with flowers, fruit and vegetables at the front as a breath of freshness to arouse our appetite. Meanwhile, they keep the milk, eggs, and other daily basics all the way back so you’ll travel through as much of the store as possible, and be tempted along the way.

In the early days, as the supermarket multiplied, so did our suspicion of it. We have long feared that this “revolution in distribution” uses corporate black magic on our appetite. The book The Hidden Persuaders, published in 1957, warned that supermarkets were putting women in a “hypnoidal trance (催眠恍惚状态),” causing them to wander aisles, bumping into boxes and “picking things off shelves at random.”

8 . We often think of fingerprints as the tiny arched patterns on the tip of each finger. They are regarded as special markers of human identity, even more individualized than DNA. But new research suggests our brains have “fingerprints” that are equally unchanging and unique to each person.

With modern neuroimaging (神经影像) techniques, scientists can track your brain’s distinct signature composed of tens of thousands of electrical signals that communicate across the brain. The final product is a picture of brain’s electrical activity that is detailed, distinct and difficult to change. According to Zack Y. Shan, head of the neuroimaging platform at the Thompson Institute, “The brain is a symphony orchestra (交响乐团).” Each region plays a unique instrument and adapts to work with nearby tunes at the same time. This cooperation leads to our thoughts and actions. “And no two symphonies sound exactly alike,” Shan adds.

A recent study published in Sleep maps the extent of this neurodiversity through EEG snapshots (脑电图快照), which describe the sleeping brain’s electrical activity as wavy lines. Led by Michael Prerau, a professor of medicine at Harvard Medical School, the researchers analyzed brainwave data of sleep spindles, one to two seconds evident neural activity associated with our ability to turn short-term memories into long-term memories.

For Dara S. Manoach, a co-author of the Sleep study and professor of psychology at Harvard Medical School, the sleeping brain is a “new frontier” for studying the treatment of memory problems in neurological disorders. She notes that lack of sleep spindle activities has been linked to different mental diseases like epilepsy and Alzheimer’s disease. So, the researchers compared two nights of sleep recordings from healthy participants to those with Alzheimer’s disease. Their analysis revealed that their broadened approach to analyzing sleep spindles also could unearth new biological indicators for Alzheimer’s disease. “It’s a first step to better understanding how the disorder operates and developing targeted treatments,” Manoach says.

Moreover, brain fingerprints may offer inspirations that traditional therapeutic (治疗性的) observation can’t. Patients with different diseases can have similar symptoms. “That’s where neuroimaging comes into play. Brain fingerprints are windows into distinguishing between two patients who seem identical,” explains Dan Hermens, a professor of neurobiology at the Thompson Institute.

In the wake of increasing reports of mental disorders, there is therapeutic promise. Brain fingerprinting could offer a potential way out of the dark chapter and provide new possibilities for the mental health crisis using best evidence-based practices to overcome it.

9 . Frozen in time, a 125-million-year-old mammal attacking a dinosaur. A 39-million-year-old whale, the heaviest animal that ever lived. The oldest known jellyfish, from 505 million years ago. Paleontology (古生物学) produces newsworthy discoveries.

Fossils (化石), moreover, provide direct evidence for the long history of life, allowing paleontologists to test hypotheses (假设) about evolution with data only they provide. They allow investigation of present and past life on Earth. Flows of biological diversity, appearances of new life forms and the extinctions of long existing ones, would go undiscovered without these efforts. But the headlines over exciting new fossils greatly underestimate the true importance of paleontology. Its real significance lies in how such discoveries brighten the grand history of life on Earth. From its beginnings, more than three billion years ago, to the present day, fossils record how life adapted or disappeared in the face of major environmental challenges.

Paleontologists provide us with a unique vantage on modern climate change. They play an essential role in interpreting ancient environments, in reconstructing ancient oceans, continents and climates. Fossils provide key limitation on the climate models that are essential for predicting future climate change. And the fossil record gives important insights into how life will respond to predicted future climate conditions, because these have occurred before in Earth’s history.

In addition, paleontology has provided a fundamental contribution to human thought: the reality of species extinction and thus of a world that has dramatically changed over time. In documenting the history of life, paleontologists recognized that many extinction episodes could occur suddenly, such as the event 66 million years ago that ended the dinosaurs. The search for the causes of past mass extinctions started pioneering studies from across the scientific spectrum (科学界), focusing on potential future threats to humanity.

Not only do paleontologists know what happens to life when things go bad, they also know how long it takes for ecosystems and biodiversity to recover from these disasters, which can take far longer than modern humans have existed.

Paleontologists thus provide a unique perspective on the nature and future long-term ecological impact of the current human-produced biodiversity crisis, the so-called Sixth Extinction, and therefore the importance of protecting modern biodiversity. The very concept of a Sixth Extinction would not exist without paleontologists documenting the first five.

Paleontologists know that understanding life’s past is critical to anticipating and adapting to life’s and humanity’s future. Paleontology is important because it brings its unique and critical perspective to current challenges in climate change, biodiversity loss and the environment. Paleontologists can predict the future because they know the past.

We’ve all seen the little kids—losing temper in the toy store, screaming in restaurants and generally making a scene in public. For their parents, giving in to a kid’s monstrous behavior helps to end the mess and gain some peace and quiet, and this is where overindulgence begins.

According to Kathy Webb, a psychotherapist in Brunswick, overindulgence is the result of parents’ beliefs. A lot of overindulging parents believe their children should be happy all the time, which is the reason why the parents try to avoid conflict at all costs. They also hold that overindulgence equals love, but experts insist that being overly permissive and indulgent is not a healthy kind of love.

Overindulgence can create a myriad of wrong attitudes and behaviors in children. When overindulged, children develop unrealistic expectations which do not serve them as they grow. Healthy parenting, Webb said, means giving children unconditional love, quality time, healthy discipline and respect for what is appropriate. “Healthy parents promote good values and use everyday life experiences to teach their children,” she said.

Many parents don’t intentionally overindulge, but fail to follow through on consequences.   “Consequences help children develop their own self-guidance skills,” Webb said. “Without consequences, children never learn to discipline themselves.”

But parents who have overindulged shouldn’t just throw up their hands and walk away from the situation, feeling like failures. There are steps that can be taken to turn family life around. “All is not lost,” Webb said. “You just have to take it step by step.” Identifying problems and what can be done to correct them are the first steps. This is where parents might want to invest in some counseling or parent coaching and refer to some parenting books.

Such parents should also manage to regain proper parental power. In some families, overindulging parents are acting like peers, not parents. Consistency is also important—children often imitate the behavior they see from their parents. “Say what you mean and mean what you say,” Webb said. It seems that all it takes is one raised eyebrow and a very stern look to stop the misbehaviour. But Webb believes what really grounded her children is their inclusion in the family’s life, not just the fun stuff like vacations and outings. They should also be taught to clean house, cook, and do laundry.