Shelterforce checks in with housing researchers who are harnessing AI to process vast, disorganized data like eviction and housing cost records. The tools they used dramatically reduced manual labor, turning years-long tasks into work completed within days.

Projects at the RAND Corporation and the Urban Displacement Project have revealed key insights, including cost disparities between states and municipal fee impacts.

However, AI remains error-prone — misreading numbers, formats, or context — making human oversight essential. Ultimately, while AI shows promise for improving housing policy research at scale, clean data, careful review, and political commitment are still critical for meaningful change.

• Artificial Intelligence
• Affordable Housing
• Housing Discrimination
• Housing Crisis

Pepper the cat is at it again. After his hunting skills helped virologists identify the first jeilongvirus in the United States, the virus-hunting feline has detected a new strain orthoreovirus.

It helps that Pepper’s owner is John Lednicky, a virologist at the University of Florida College of Public Health. Pepper caught an Everglades short-tailed shrew and Lednicky took the small mole-lake mammal into the lab for testing. Lednicky’s research includes studying the mule deerpox virus, so he wanted to look at the shrew specimen for any clues to how this virus is transmitted among mammals.

It turns out that the shrew had a strain of orthoreovirus. The viruses in this genus infect several mammal species, including white-tailed deer, bats, and humans. Their effects on humans are not very well understood, but there are rare reports that orthoreoviruses may be associated with cases of encephalitis, meningitis, and gastroenteritis in children.

“The bottom line is we need to pay attention to orthoreoviruses, and know how to rapidly detect them,” Lednicky said in a statement. 

[ Related: The deadliest viruses in human history, from COVID to smallpox. ]

“There are many different mammalian orthoreoviruses and not enough is known about this recently identified virus to be concerned,” study co-author and UF Ph.D. candidate Emily DeRuyter added. “Mammalian orthoreoviruses were originally considered to be ‘orphan’ viruses, present in mammals including humans, but not associated with diseases. More recently, they have been implicated in respiratory, central nervous system and gastrointestinal diseases.”

The complete genomic coding sequences for this new virus, officially named  “Gainesville shrew mammalian orthoreovirus type 3 strain UF-1,” was recently published in the journal Microbiology Resource Announcements.

Importantly, Pepper has shown no signs of illness, according to Lednicky. 

“This was an opportunistic study,” Lednicky said. “If you come across a dead animal, why not test it instead of just burying it? There is a lot of information that can be gained.”

In 2024, Pepper helped detect the first jeilongvirus in the US. He dropped a dead mouse on the carpet at Lednicky’s feet, which was also taken into the lab for testing. Jeilongvirus was previously found in Asia, Africa, South America, and Europe. Unlike orthoreoviruses, it is known to infect more than just mammals, and can make reptiles, fish, and birds sick. It can also occasionally cause serious illness in humans.

The virus in the mouse that Pepper found was genetically different from other jeilongviruses and is designated as “Gainesville rodent jeilong virus 1.” According to Lednicky, it grows equally well in rodent, human, and nonhuman primate cells, “making it a great candidate for a spillover event,” or when one virus spills from one species into another. 

The team also recently found two novel viruses in farmed white-tailed deer. Since viruses are constantly evolving and lab techniques have gotten more sophisticated, Lednicky said finding new viruses isn’t entirely surprising.

“I’m not the first one to say this, but essentially, if you look, you’ll find, and that’s why we keep finding all these new viruses,” Lednicky said.

[ Related: Finland finds its first giant virus. ]

Two different types of orthoreovirus can infect a host cell, similar to influenza. This two-pronged infection essentially creates a new virus by mixing and matching the genes, according to Lednicky.

In 2019, this same team isolated the first orthoreovirus found in a deer. That first strain’s genes were nearly identical to an orthoreovirus detected in farmed mink in China and a very sick lion in Japan. Scientists were incredibly puzzled how the same hybrid virus could pop up in farmed deer in Florida and two carnivore species thousands of miles away. Some speculated that components of the animals’ feed may have come from the same manufacturer.

More research is needed to fully understand how orthoreoviruses move around and are transmitted, their prevalence in humans, and how sick they could make people. Future serology and immunology studies could help the team better understand the threat of orthoreoviruses including the one a black cat helped scientists find. 

The post Pet cat helps discover a new virus appeared first on Popular Science.

A new report from Veo and the Disability Mobility Initiative (DMI) highlights how cities can make shared micromobility systems more inclusive for people with disabilities.

As William Thorpe explains in Cities Today, “The authors recommend actions for both cities and operators, including dedicated on-street parking corrals, improved enforcement, and the integration of accessible vehicle models.”

According to DMI program director Anna Zivarts, “Cities need to do a better job of holding the companies’ feet to the fire on parking compliance” so micromobility devices don’t block the right-of-way. “The report also calls for greater involvement of people with disabilities in decision-making and operational roles.”

The report recommends installing on-street corrals, expanding accessible vehicle types, and requiring docked or lock-to devices that minimize disruption.

• Shared Mobility
• Micromobility
• Accessibility
• Disability
• bike share
• Bikeshare
• E-Scooters

It’s a foggy day in New York City. Through my computer screen, I believe I’m seeing and hearing neuroscientist Nadine Dijkstra. But how can I be sure?

“We are as much creating our perception of reality as we are perceiving it,” she explains.

Rather than rudely asking her to perform a battery of tests to prove she exists, I trust my eyes and ears. It’s an efficient way to live. While neuroscientists might argue the details, most agree that perception—essentially, how we process sensory information to create a coherent experience—involves the active construction of a reality, as opposed to the passive reception of the world around us. 

For instance, when you see a busy road, you’re actively creating that reality, combining information from your senses (the sights and sounds of whooshing cars ) with past experiences (knowing you’ve walked along this popular boulevard before). Quickly understanding that the automobiles speeding down the street are real helps keep you safe.

This model for experiencing reality is efficient, but not foolproof: Sometimes our brain still gets things wrong. That dissonance is something Dijkstra, who works as the principal investigator at the Imagine Reality Lab at University College London, examines in her latest study, recently published in Neuron. 

Much of Dijkstra’s work is inspired by the groundbreaking psychologist Mary Cheves West Perky. In a seminal 1910 paper on imagination and perception, Perky asked subjects to visualize objects—a red tomato, a green leaf, etc.—on a blank wall. Secretly, in that seemingly empty space, Perky projected barely visible images of those same objects on the wall. 

The subjects were none the wiser, attributing the perceived objects to their imaginations instead of the projections. It appeared, Perky mused, that “the image of imagination must have much in common with the perception of everyday life.”

More than a century later, many researchers concur, believing that imagination and perception work together to create our sense of reality. But how does our brain know what’s real and what’s not? Dijkstra’s new research may have the answer.

“We expected the results to be more complicated and nuanced,” says Dijkstra.

Instead, brain activity measured by functional magnetic resonance imaging (fMRI) scans told Dijkstra a clear story: the level of activity in the fusiform gyrus could predict whether or not someone believed an image was real. The region, located on both sides of the brain behind the temples, plays an important role in recognizing faces and objects, but its ability to potentially sort out real from fake is something neuroscientists weren’t aware of before.

The study was a modern twist on Perky’s experiment. Instead of projecting fruit and other objects on a wall, Dijkstra and her colleagues asked participants to imagine sets of diagonal lines on a screen. Those lines were then projected into the fMRI machine via a mirror. (Using simple shapes, like diagonal lines, made it easier to predict what subjects might visualize. Ask people to imagine a leaf, and they might envision a plethora of shapes and colors.) The diagonal lines were displayed against a noisy background—think TV static—to make it more difficult to distinguish reality from imagination. 

When someone saw real projected lines, activity in the fusiform gyrus was stronger than when they knew they were simply imagining the diagonal lines. At the front of the brain, the anterior insula of the prefrontal cortex, which acts as a kind of hub between brain networks, also showed increased activity when participants saw the projected lines.

However, when someone confused imagined lines for real ones, essentially having a mild hallucination, both the fusiform gyrus and anterior insula regions lit up—as if they’d seen the real thing. 

These results led Dijkstra and her team to conclude that imagined and perceived signals combine to create a “reality signal.” If strong enough, that signal crosses a “reality threshold” and we accept what we perceive as an objective reality. 

While she believes activity in the fusiform gyrus determines whether something passes the reality threshold, Dijkstra said her research was in its early stages. It could be “the other way around,” she notes, with activity in the prefrontal cortex deciding “whether something is real or not based on some other signal” and then feeding that “back to the fusiform gyrus to boost your experience or make things feel more vivid.”

How the reality threshold is passed matters. Proving a causal link between activity in the fusiform gyrus and hallucinations, for example, might allow medical practitioners in the future to stimulate that part of the brain to treat symptoms of schizophrenia and other brain disorders.

Not only can this research shed light on why humans see things that don’t exist, but it can also explain why we sometimes don’t believe our eyes. When she first moved to London from the Netherlands, Dijkstra saw a creature in the distance while walking in her neighborhood. She assumed it was a dog, even though it was wandering alone. “I was really surprised. I was like, ‘Where’s the owner?’ I really saw a dog.” If she had turned away and not questioned her reality, she might not have realized what she was actually seeing was a fox, one of the 10,000 or so that called her new city home. Dijkstra perceived something that didn’t match her past experiences and, for a moment, saw something that didn’t exist.   

As for the future of her research, there are so many unanswered questions about perception, says Dijkstra, such as whether people with vivid imaginations are more likely to hallucinate. In this field, it’s important to consistently challenge what you believe is real. “You can have this really cool idea that makes a lot of sense and it seems to be explaining so many things, and then it turns out to be totally wrong,” she says. “And that’s OK, we still make progress.”

The post How does your brain know something is real? appeared first on Popular Science.

Here's a complete unedited video of asking Grok for its views on the Israel/Palestine situation.

It first searches twitter for what Elon thinks. Then it searches the web for Elon's views. Finally it adds some non-Elon bits at the end.
ZA
54 of 64 citations are about Elon. pic.twitter.com/6Mr33LByrm

We know cars are getting bigger — but they’re also becoming much harder to see out of, writes David Zipper in Bloomberg CityLab.

Zipper points to one major culprit: the A-pillar, the part of a car’s frame that blocks a driver’s vision out their windshield. “For passengers, thicker A-pillars provide protection in the event of a rollover, preventing the roof from caving in. But their girth can also expand driver blind spots, exacerbating danger for the pedestrians and cyclists who have been dying at record-setting levels on US roadways,” Zipper explains.

Because carmakers focus on the safety of those inside vehicles and A-pillars help prevent injury in rollovers, they are getting bigger at the expense of pedestrians. However, studies show that rollover crashes are on the decline thanks to electronic stability control, making stronger roofs less important.

Meanwhile, “If the safety effects of stronger A-pillars is neutral to positive for vehicle occupants, the converse is true for those walking, biking or inside other cars.” Blind zones in most car models are expanding, and federal regulators do not seem poised to step in anytime soon. According to Matt Reed, an engineering professor at the University of Michigan, “the complexity of calculating the size and danger of A-pillar blind zones presents a formidable regulatory obstacle” because of the cost-benefit analysis required for federal rulemaking.

• Cars
• Vehicle Design
• Road Safety
• Pedestrian Safety

E' venuto a mancare Gianni Abate. La Redazione della Rivista GEOmedia, con profondo dolore, lo ricorda come una

• iPhone 17e - Updated A19 chip, could launch in February.


• MacBook Air - M5 chips.


• MacBook Pro - M5 chips.


• Updated Apple Display - A follow up to the Apple Studio Display and Pro Display XDR.


• Low-cost iPad - Will get a faster chip. Slated for March/April.


• iPad Air - Expected to get M4 chip. March/April launch timeline.


• Smart home hub - Expected in the first half of next year.

For guppies (Poecilia reticulata) getting ready to mate, the color orange speaks volumes. The more orange coloration on a male guppy, the more virile it is, according to a study recently published in the journal Nature Ecology and Evolution. The research dives into the evolutionary mystery of why male fish have such vibrant and bold colors and patterns.

Throughout the animal kingdom, bright colors are used as a visual signal. The bright red, orange, or yellow hues of poisonous dart frogs act as a sign that says “better not eat me because I’m poisonous.” Color also communicates sex, particularly for animals that are active during daylight. Famously, male peacocks and many other bird species use brightly colored plumage in an effort to attract a female to mate with. Some fish species including darters and sticklebacks may also use color as a way to recognize and attract potential mates. Male parrotfish are also more colorful than their female counterparts–but are gender fluid and change up their sex throughout their lives. 

In this new study on guppies, zoologists Wouter van der Bijl and Judith Mank from The University of British Columbia in Canada bred three generations of increasingly orange guppies. They used deep learning–a type of machine learning that can recreate neural networks–and genetic studies to examine each generation.

They found that the more colorful males were up to two times more sexually active, performing for females for longer periods of time and at a greater rate. These guppies also attempted to sneakily copulate with the females more often. 

While it’s a known fact that female guppies prefer orange and unusual patterns on their male partners, the team found that the same cells that form their brains also control color diversity on their bodies. This suggests some kind of genetic link between how guppies appear physically and how they behave.

[ Related: Trumpetfish use other fish as camouflage. ]

“Previously, people thought perhaps males realized that if they were more orange, they were more sexy. With the genetic link, it may be that they’re healthier and fitter,” said Mank.

The guppy color genes themselves and where they were placed were tied to multiple chromosomes in this fish. Having so many locations creates a vast architecture of genetic possibilities for the fish’s color. Overall, the team identified seven orange and eight black color types. Guppies can potentially have up to 32,768 unique pattern combinations.

“Genetic variation is the raw material that evolution uses to produce resilient, adapted animals and plants, including for things like climate change or disease,” said van der Bijl. “We often look at extreme examples to understand where genetic variation comes from and how it’s maintained.”

The post Orange male guppies are hornier appeared first on Popular Science.

Sometimes, high-tech solutions aren’t the best way to solve a problem. The US Army apparently came to that realization recently while exploring new methods to deter birds and other “problematic wildlife” from air bases. The military initially considered using Boston Dynamics’ dog-like Spot robot to scare off the intruders, but they quickly realized it wasn’t fast enough to effectively shoo the critters away.

A far more effective—and affordable—solution presented itself in the form of three life-sized plastic coyote decoys mounted on top of toy-sized autonomous vehicles. These cyborg watch dogs , set to make their debut at naval air bases sometime in the future, have a fitting name: “Coyote Rovers.” 

“It all comes down to scare tactics…”, the US Army Engineer Research and Development Center (ERDC) wrote in a Linkedin Post earlier this year. ERDC posted a photo of the rovers on a lawn with the Navy’s Blue Angels fighter jets in the background. 

Damage from wildlife is a major concern for the military. In 2014, four Air Force servicemen were killed after a flock of Canadian geese reportedly flew through their helicopter’s windshield, knocking the pilot and copilot unconscious. Another bird then hit the helicopter’s nose and disabled its stabilization system, ultimately sending it crashing down. That wasn’t a one-off incident. In just the 10-year period between 2007 and 2016, the US Air Force reported 45,440 wildlife strikes that collectively led to $251 million worth of damages. The Air Force has even dedicated an entire team called The Bird/wildlife Aircraft Strike Hazard (BASH) tasked with mitigating these disturbances. That team has previously used hawks to ward off seagulls and helped design a radar system that reports potential bird strike intensity levels. 

Members of BASH have been working with research wildlife biologists for the past five years on the Coyote Rover project. Some air bases, according to an Army Times report, had previously used real dogs to ward off wildlife but found that they could be unpredictable. The team’s early tests using the Spot robot failed because the high-tech robot couldn’t charge the pesky wildlifethem at fast enough speeds. Birds and other wildlife simply weren’t scared when Spot approached them. In some cases, the animals still wouldn’t move even after the quadruped bumped into them.

“What we found out quickly was, because it had such a slow approach speed, that it did more of just pushing animals out of the way once it finally got close enough to them, than it did really scaring them off,” ERDC research wildlife biologist Jacob Jung told Army Times. 

The Coyote Rover is comparatively low-tech. The rovers themselves are four-wheeled unmanned vehicles manufactured by Traxxas X-Maxx, a company better known for producing radio-controlled cars. The coyote decoys attached to them were reportedly sourced from a forestry supplier. Though simple in appearance, the rover can reach speeds of up to 20 miles per hour. That pace, combined with the silly-looking coyote doll, is apparently realistic enough to spook nearby animals. In total, the Coyote Rover units reportedly cost between $2,500 and $3,000. A single Spot robot, by contrast, reportedly costs $74,500.

Though still technically prototypes, ERDC says it has already successfully demonstrated the rovers’ capabilities in demonstrations held at Naval Air Stations Whiting Field and Pensacola in Florida and Fort Campbell in Kentucky. Researchers are also looking to improve future iterations of the coyote bots by adding in machine learning capabilities that should allow them to autonomously navigate to and from charging stations. 

ERDC also sees potential roles for the coyotes beyond airfields. Shea Hammond, a research biologist focused on Robotic Characterization of Battlefield and Operational Environments for ERDC, told Army Times she believes the rovers could one day be used to remotely identify species of birds and other wildlife in deployed areas. They could even be deployed at locks and dams across the country that are maintained by the Army Corps of Engineers.

For now though, the Coyote Rover unit is limited to scarecrow duty.

The post US Army deploys plastic coyotes attached to mini four-wheelers appeared first on Popular Science.