(Intro)

Olivia: Scientists study a lot of complicated things, and when they're designing experiments, they usually need to do some creative problem solving.  Sometimes that can mean turning to decades old research in a completely different field for help, applying an odd method, or using clever stand-ins for rare or expensive materials, and without context, this kind of scientific research might come across as weird, useless, or a waste of money, but if you take a closer look, you'll usually find that seemingly silly science can have a huge, sometimes even life-saving effect on society, so here are six times that important scientific advancements have come from out of the box thinking.  

About a decade ago, a clip of a shrimp running on a treadmill went viral, and since then, it's been spread around on and offline as a poster child for wasteful spending.  Now, there was a National Science Foundation grant of over $425,000 that was awarded in 2002 to the team that made that video clip, but everyone crying 'what a waste!' is taking that number and the clip completely out of context.

The researchers weren't just plopping shrimp onto tiny treadmills to see how fast they could run.  This was part of a much larger study of how climate related changes in the ocean affect the health of commercial seafood species.  How these critters respond to stress affects their survival and ocean ecosystems, which is pretty important by itself, but at the very least, you probably care whether the shrimp going into your belly was able to fight off bacterial infections before it became food.

For these experiments, the researchers made their first treadmill entirely from scraps lying around the lab, for less than $50.  Shrimp are active creatures, so letting them lie around isn't a good model for how their bodies would respond and fight off an infection in the real world, so they used this equipment to compare the health and behavior of sick and healthy shrimp, in tanks that simulate stressful ocean conditions that are caused by climate change, like low oxygen or high acidity.  

That original grant led to four years of research and seven published papers on multiple species of crustaceans, not just the Pacific white shrimp from that video.  The scientists found that being sick seems to lower their activity and bad water conditions make it harder for them to move and fight off infections, and since then, the lead researchers have also been looking into how the side effects of climate change affect how crustaceans express certain genes that influence their health.

Speaking of climate science, there are traditional ways to study what Earth's climate was like thousands of years ago, and then there's crystallized pee.  The hyrax is a cave-dwelling mammal, a bit bigger than a guinea pig, found in the dry climates of sub-Saharan Africa in the Middle East.  Colonies of 50 or so creatures all use the exact same communal toilet, called a midden, over many generations.  One midden in South Africa has been building up for the past 55 thousand years.  

These middens can be found in caves or under rock ledges and grow to be several meters across.  Inside them, layers of goopy urine crystallize into an amber-like substance called hyraceum, which by the way, is used in some perfumes.  Dry regions lack the usual evidence used to study what the Earth used to look like, like layers of sediment that form in lake beds or peat bogs, but scientists can cut through a huge midden to see the individual layers and analyze the urine, poop, and dust blend, like how a climatologist uses ice cores or cross-sections of trees. 

Specifically, by looking at the ratios of different elements like carbon and nitrogen, they can see how much vegetation was available to eat and how dry it was, from hundreds to tens of thousands of years ago.  For example, some climate models suggest that because of slight shifts in how the Earth was orbiting the sun about 6,000 years ago, the Northern Hemisphere got drier while the Southern got humid, but some researchers found that hyrax middens tell a different story.  The Southern Hemisphere may have become drier, too.  

Unboiling an egg might sound like a cool science demo in a 'take that, entropy' kind of way, but it turns out the trick has some pretty important applications.  When you heat a raw egg up, some of the bonds in the egg white proteins break, so they change shape and get all knotted up together.  Unboiling an egg means untangling these proteins from each other, which gives them space to reform bonds and fold back into their original shape on their own, and a team of researchers found a way to do that more simply than ever before in 2015.  

First, they took a small piece of hard-boiled egg white and dissolved it in a solution of urea, the same compound found in urine.  So yeah, it becomes inedible really fast.  That breaks up the mass of tangled proteins into microscopic lumps, making them easier to manage.  Then, they put that solution inside a vortex fluid device, or VFD.  This spins it super fast at 5,000 times a minute, so a thin film flows up against the glass tube wall.  The closer the fluid is to the wall, the faster it's spinning.  The difference in speed creates a force that makes the proteins untangle from each other, and once they're separated, they can naturally re-fold. 

The researchers used chicken eggs in this experiment, mostly because they're cheap, but it also let them test a couple of different proteins, including a structural one called caveolin-1.  It takes four days for a conventional method to untangle it, but it took the VFD only minutes.  Improperly folded or tangled proteins are expensive waste in lots of industries, especially medical research and pharmaceuticals, so this technique means that non-functional proteins can be re-folded and hopefully re-used.  For example, some antibodies used in cancer treatments are developed inside expensive hamster ovary cells, but with VFDs, they can be grown in something cheaper that's more likely to lead to folding errors because the proteins can be fixed.  This would lower the cost to make them, which could be passed on to people who need these treatments to help save more lives.

Speaking of saving lives, another study that sounds silly but had a huge impact involved giving baby rats back rubs.  Back in 1979, researchers were trying to learn what factors could affect compounds found in baby rat growth. 

To avoid dealing with protective mama rats, the researchers separated them, but to their surprise, the compounds that they were trying to study stopped increasing as the separated pups got older.  Experiments ruled out other factors like nutrition, and they found that a mother's grooming and licking was important for healthy growth.  The researchers then tried taking a tiny brush, like the one you would use to clean a camera lens, and rubbed the pups' back, and the baby rats made normal amounts of those growth compounds again.

A psychologist working in pediatrics later learned of this study and applied it to her work with human infants born prematurely, and her study demonstrated that 15 minutes of gentle massage three times a day caused preemie babies to gain weight 47% faster than those left alone in incubators, which was standard practice.  Even though they were eating the same amount, the massaged babies' nervous systems also matured faster, so they became more active and alert and on average, the babies also spent six fewer days in the hospital.  

Infant massage therapy is now used in hospitals across the US and has saved an estimated 4.7 billion dollars a year in healthcare costs.  All these researchers were awarded a Golden Goose Award in 2014 for US federally funded research that seems unusual but has a major impact on society.

In 2016, a Golden Goose Award was given to researchers who took a study about honeybees and applied it to, of all things, the internet.  The research began in the 1980s, with studies modeling how honeybee colonies hunt for nectar.  This is harder than it might sound, because patches of flowers are different distances away, bloom at different times, and have different amounts of nectar, so their value to a bee changes over time.

Because these scientists monitored how each individual bee foraged in a colony, they had to glue tiny numbers onto over 4,000 of them by hand, and they were able to generate a computer model of the bees' decision making strategies.  In 2002, the model they created was adapted for something that might seem totally different: internet traffic management.

Instead of nectar, companies that owned web-hosting servers are like bees who want to collect the most amount of money from the websites and apps that request to use them, so you can find them online.  

Servers can use the honeybee algorithm.  It helps them modify how they distribute their time depending on changing supply and demand.  Like, when a bunch of people all decide to visit a science museum's website to buy a purple dinosaur hoodie featured on a hit Netflix show, that site needs more server capacity.  Basically, the honeybee algorithm works like an advertisement board that tells a company's servers about the current needs for a bunch of different websites.  

Ads for high-demand and high-paying websites pop up more often and stay up longer, to make sure more servers are dedicated to them, but the honeybee algorithm also helps us, by making the internet run more smoothly, to save time and hopefully keep suddenly popular pages from crashing.

And speaking of optimization, you might not think a brainless single-celled organism could grow up to become a city planner, but let me introduce you to Physarum polycephalums.  Researchers revealed in 2010 that it can grow into a structure that mimics the layout of real world transport systems.  They placed this slime mold on a 2D map of the Tokyo area with oat flakes for 36 nearby cities.  The slime mold started at Tokyo and sent out branches uniformly as it searched for food and as soon as it found the tasty flakes, the (?~9:11) between cities grow larger and the unnecessary branches shrunk away.  That way, it can sustain itself while keeping its body plan really efficient to save energy.

Over the course of 26 hours, it had formed a network of thin tubes that looked a lot like Japan's rail network.  Similar results happened across multiple tests.  Based on resources, the slime mold's network was a little more efficient than Japan's, but the real transport system had better resilience, meaning it could bounce back if any of the city connections failed.

The most important takeaway from this research is that the slime mold generated a really efficient network in way less time than we would, using a method that engineers couldn't do in real life.  It'd be like putting railways everywhere and then destroying the ones that didn't perform well.  Other researchers have run similar experiments, simulating different countries.  One group found that, of the 14 countries or regions they tested, Malaysia, Italy, and Canada's major transport networks matched the slime mold the best.

The USA, on the other hand, was one of the worst.  However, not all that research considers natural barriers like mountains or bodies of water.  Computer models derived from these studies could be used when we design more transport for people or resources, especially to help balance the cost of construction with the benefit of reaching lots of places.  That would put the slime molds out of work, of course, but maybe they'd still consult on especially tricky layouts.  

Studies like these show that you can't judge the value of a scientific experiment by how unusual the methods seem.  Something might seem funny at first glance, but maybe researchers are finding really clever solutions to complex problems.

Thanks for watching this episode of SciShow.  If you'd like to keep learning about all kinds of weird science with us, you can  go to YouTube.com/SciShow and subscribe.

(Endscreen)