Let the controversy begin! An analysis from the University of Pittsburgh and Oxford University examining decades of global scientific progress has concluded that long-distance collaborations between remote groups and individuals—the standard approach to modern science—is more likely to yield incremental advances at the sake of new, disruptive ideas. The researchers (who are separated by 3,600 miles and one ocean) looked at 20 million research articles published from 1960–2020 and four million patents submitted between 1976 and 2020 across the globe. They looked at how disruptive those research results were and compared collaborations between researchers co-located in the same city to ones where the scientists were spread out to two or more cities. “Although remote collaboration among specialized researchers permits more new combinations of knowledge, it also makes it harder for teams to integrate the pieces,” the researchers write, adding that true innovation often has a hometown. Nature

This map visualizes how geography matters to scientific discovery. The darkened waves represent 60 years of collaboration resulting in 20 million research articles across 3,562 cities. The bright lights show how ideas tend to fuse and light up locally, driven by co-located or nearby teams. Remote connections spread the illuminations, but they often fall short in sparking brilliance. Yiling Lin

One of the most famous results of mathematics in the last 60 years is the so-called butterfly effect—the idea that a tiny insect flapping its wings in Africa could influence the path of a tornado in Kansas weeks later. That concept, technically known as sensitive dependence on initial conditions, is the basis for chaos theory, which shows that even if a phenomenon like the weather is mathematically “deterministic” and can be accurately modeled, there are nevertheless hard limits on our ability to predict which way it will go. While many examples of chaos have been described in physics, their existence in biological systems has been less certain because of the inherent complexity of biology. But now researchers at Oxford University (none of whom collaborated with researchers in other cities) have demonstrated that populations of E. coli bacteria display gene regulation chaos in response to oxidative stress. Those chaotic dynamics help generate strong and variable responses to changing environments, the researchers write. Current Biology

Herman Melville’s Moby Dick is a classic example of the “sunk-cost fallacy,” how humans are ever more committed to chasing something after they’ve already spent blood, sweat, tears, and money on its pursuit—even if that hunt is self-destructive. Economists have described many examples of this irrational behavior in modern life, though neuroscientists have never been able to explain why we do it. Now based on studies in mice, researchers at Stanford University may have found the underlying neurological basis for the sunk-cost fallacy: Dopamine release in the brain’s striatum, a region which is essential for learning, habituation, motivation, and addiction. They showed that the brain responds to not only the size of a reward but also how hard it is to achieve. If mice are pursuing a goal, more dopamine is released in their brains when they pursue more difficult goals as well as ones with large payouts. Likely the same is true in humans. Neuron

Bias against artificial intelligence may be driven more by our affinity for creators who are our fellow humans than some strong aversion to AI-created work, according to a first-of-its-kind study from MIT and the University of California, Berkeley. The study found the generative AI tool ChatGPT-4 outperforms human professionals when tasked with creating ad content for campaigns aimed at enticing consumers to do things like purchase air fryers, projectors, and e-bikes. “Research like this can help better illuminate the ways people think about AI, which is critical for understanding its adoption, bias proliferation, and how we can design human-first AI tools,” one of the researchers said in a press statement. Judgment and Decision Making

Researchers at Uppsala University and the Karolinska Institute in Sweden have demonstrated that genetics influence whether infants prefer to look at the faces of people around them or instead to gaze at cars, phones, or other “non-social” bright shiny objects. Based on data from the Babytwins Study Sweden (BATSS), the research compared both identical and fraternal twins and found a strong genetic effect: Identical twins were alike in their gaze preference. This suggests there is a biological basis for what infants look at and seek to learn most about, the researchers write. They also found that the more infants preferred looking at faces rather than objects at five months of age, the larger their vocabulary in the second year of life. Nature Human Behavior

Developmental heart defects are the main cause of fetal death, but they are also hard to study because the key milestones in fetal heart development cannot be studied after 14 days, which is the longest human embryos are allowed to be kept alive. And neither animal models nor cell cultures quite cut it. Now researchers at the Institute of Molecular Biotechnology of the Austrian Academy of Sciences (IMBA) in Vienna have created an “organoid” based on human heart cells that reproduces the different aspects of the embryonic heart, like separate ventricle and atrial chambers. They use this new technology to unravel how mutations, drugs, and environmental factors affect heart function in each compartment and cause specific defects in the developing human heart. And in an unrelated business announcement, Philadelphia-based biotech Vivodyne announced the close of $38 million in seed funding for its work developing lab-grown human organoids for drug discovery, which shows how much the technology of organoids is really coming of age. Cell

While the biological gulf between people and penguins is greater even than, say, between humans and mice, perhaps there is something we can learn from those flightless, lovely little birds. According to researchers at the Korea Polar Research Institute in Incheon, South Korea, and the Lyon Institute of Nanotechnology in Villeurbanne, France, the Antarctic chinstrap penguin has a most remarkable ability to rest in fractured “microsleeps,” which is driven by the fact that they are almost constantly surrounded by danger from predators as well as aggression from other penguins. The price of eternal vigilance for these tuxedoed cuties? Their sleep. Were they to fall asleep too long, they would risk being attacked or having their eggs stolen. As a result, they have evolved the ability to take some 10,000 micro naps each day, each one lasting just a few seconds, so that at the end of the day they have racked up 11 hours of sleep while barely blinking. What does this mean for human health? Imagine if we find a way to force-fragment daytime naps into tiny bouts like the penguin does. We would reduce accidents, enhance mood and brain performance, and improve health! Science

Chinstrap penguin with two chicks. Won Young Lee