Previous: The world’s largest waterfall is underwater. #shorts #science #SciShow
Next: Why Marine Mammals Suck



View count:320
Last sync:2023-01-06 23:15
Even if the reputation of video games has grown tarnished since a nuclear physicist first invented them back in the 50’s, they are still helping us advance scientific discovery.

Hosted by: Hank Green (he/him)
Support SciShow by becoming a patron on Patreon:
Huge thanks go to the following Patreon supporters for helping us keep SciShow free for everyone forever:

Matt Curls, Alisa Sherbow, Dr. Melvin Sanicas, Harrison Mills, Adam Brainard, Chris Peters, charles george, Piya Shedden, Alex Hackman, Christopher R, Boucher, Jeffrey Mckishen, Ash, Silas Emrys, Eric Jensen, Kevin Bealer, Jason A Saslow, Tom Mosner, Tomás Lagos González, Jacob, Christoph Schwanke, Sam Lutfi, Bryan Cloer
Looking for SciShow elsewhere on the internet?
SciShow Tangents Podcast:

#SciShow #science #education #learning #complexly

Image Sources:
[♪ INTRO] You may have seen headlines recently about a team of scientists who taught neurons in a Petri dish to play Pong.

As outrageous as that sounds, scientists aren’t exactly noobs when it comes to using video games in the pursuit of knowledge. In fact, it was a nuclear physicist who invented the first video game back in 1958.

Think of William Higinbotham as the original Player 1. He was working at the Brookhaven National Laboratory and wanted to demonstrate the trajectory of objects in a more engaging way at the institute’s public visitor’s day. Using a five inch oscilloscope with two controllers, he created a sort of Pong-predecessor called Tennis for Two to simulate how a tennis ball moves.

Voila - the age of electronic gaming was born. So from the early days, video games have been a tool for science and education. Today, they play a role in research ranging from microbiology to Artificial Intelligence.

So settle into your gaming chairs. Here are five ways scientists have used Ataris, Nintendo Wiis, and other systems to learn something new about our world. No genre of video game has generated more research over the years than violent ones.

More than a hundred journal articles have been written on the potential dangers of allowing little Johnny to play that first-person shooter game. Now, the data have been mixed, and there’s no conclusive evidence that violent video games promote antisocial behavior like aggression in children. After all, with the increase in popularity of online cooperative gaming, it’s hard to call battle-style games exclusively violent or non-violent.

Take Fortnite, which is an online game in which players compete to be the last survivor. On the surface, it’s kinda hard to imagine a game where players wield rocket launchers and axes against each other as being good for kids’ socialization. But in cooperative mode, players must also help their teammates survive.

One study had children ages 9-12 play Fortnite to see if cooperative game play could translate into pro-social behavior outside the game. The opposite of antisocial behavior, “pro-social” is defined as helping, sharing, and caring for others. In the experiment, some children played a game of pinball or a solo round of Fortnite, while another group played Fortnite in team-mode.

Afterward, experimenters asked everyone if they’d be willing to donate more of their time to future experiments. Children who engaged in co-op play were more likely to agree to give time to potential future experiments. And the team-playing kids agreed to donate larger sums of hypothetical prize money to charity.

Furthermore, cooperative game players also reported feeling more competent and more connected to their playmates, suggesting that co-playing video games could be important in promoting not just pro-social behaviors, but also psychological satisfaction and self-esteem. Studies with adults have showed the same kind of results: that playing even a violent video game cooperatively increases pro-social behaviors in the real world. Maybe your next company retreat should involve a team-building exercise with flamethrowers and crossbows on Fortnite.

Of course, video games are more than just explosions and button-mashing. They can also be used to teach some serious precision hand-eye coordination, too. Something surgeons really need.

A 2021 study showed that experience with first-person-shooter games, like Half-Life, boosted performance on a virtual reality surgical task. Despite the violence in such games, the precision and aim required seemed to translate to improvements in surgical skills. In fact, multiple studies have documented the beneficial effect that video game playing has on a surgeon’s abilities, particularly for robotic and laparoscopic surgery.

The 2021 meta-analysis evaluated all those results and found that medical students with a history of gaming more than three times per month had faster times to complete a simulated surgical procedure and better economy of motion, meaning they moved their hands and tools more efficiently. Shorter completion times means a shorter surgery for the patient, and hopefully less poking around inside their body. The same study reported that the Wii U game Underground, with its unique controller and inverted controls, was particularly great for medical training.

Underground is a game about a girl and her robot navigating through a maze of caverns. So, nothing to do with operating on bodies, but you do need two-handed coordination. And just like with robotic surgery controls, pushing up on the controller moves the character down and pushing right on the controller moves her left, etc.

Everything is switched. The game has been scientifically validated for how well it mimics and teaches laparoscopic skills. Because its controls so closely resemble the movements needed to perform surgery, skill at the game is a good predictor of skill in the O.

R. This sort of research suggests gaming could be incorporated into medical school curricula. Who knows?

The top surgeon at your nearest hospital may also be the one with the highest score at the local arcade. So far, we’ve been talking about clever ways scientists have used existing games. But sometimes, they develop a new one with a specific goal in mind.

In 2008, the University of Washington released a computer game called FoldIt. The initial objective of the game was to fold proteins into 3D structures that required the least amount of energy. Proteins do a lot of work in the body.

They make up your hair, nails, muscles, your digestion enzymes… But they only do their jobs right if the molecular chain they are made of is folded into the right shape. The problem is that if a protein requires too much energy to reach its final form, it will get stuck in the wrong, non-functional shape. It’s handy if getting into that just-right shape doesn’t take much effort.

Like sophisticated origami, figuring out the ideal shape can be tough. So scientists wanted to see if gamers could use FoldIt to identify it for various proteins they had already mapped. More than 100,000 people tried it out, and they got really good.

So good, in fact, that top FoldIt players produced ideal protein structures faster than even expert-trained AI. That means the players found shapes that satisfied the biological rules for low energy use while computers would get stuck on creating a partially-folded failure. Which is great news!

And now that players have figured the game out, they were ready for some real-world application. With FoldIt, scientists could provide the limited data they already had about proteins of interest and then crowdsource the rest to determine their 3D structures. Knowing the structure of a protein is critical to understanding its function, and in the case of virus proteins, how to design drugs that interact with them.

So for example, the online community identified the structure of a viral enzyme common in monkeys in three weeks that scientists had spent years trying to figure out with standard techniques. FoldIt-generated structures now appear in scientific journal articles every few years, and we’ve even begun to study the top players to better understand how they think through solving the puzzles. The goal is to teach AI better spatial reasoning skills so it can mimic the best players and help scientists solve structures faster for drug development research.

That protein-folding game actually looks pretty hard, but so is remembering the location of the warp zone in Super Mario Brothers. And as it turns out, that kind of challenge is a good thing for older people who could be facing cognitive decline. One region of the brain important for learning and memory is called the hippocampus.

And scientists believe that various kinds of mental enrichment keep the hippocampus functioning normally and its cells in tip-top shape. In a 2020 study, researchers set out to test if video games could offer this enrichment in seniors by giving adults 60-80 years old a Wii U and teaching them to play Super Mario Brothers or Angry Birds. Participants had to play about 30 minutes a day for 4 weeks.

That might not sound like a lot, but at the end of the study, those who had played the Wii fared better on all the memory tests that experimenters threw at them, compared to those who had just played solitaire. First, participants took a similarity matching test. Pictures of objects were presented for two seconds, and on the subsequent test, participants were re-shown pictures and had to answer if they’d seen the object before, or if it was a totally new image.

They also took a drawing test where they had to first copy, and then later draw from memory, a complex line drawing. Even though Mario Brothers and Angry Birds aren’t memory games like Simon is, the novelty of the new games and console seemed to be enough to provide stimulation to the hippocampus and boost some cognitive performance. The researchers believe games could be a potential intervention to help senior patients stave off memory problems.

So try inviting grandma to game night next time. Which brings us back to that weird example of neurons in a dish playing the Atari game Pong. First of all, no, the little brain cells were not lined up in a bar bouncing a ball back and forth across the dish.

They were playing virtually, if playing is even the right word for a collection of cells connected by electrodes to a computer. I’ll explain what I mean in a second, but first let’s talk about what the scientists were hoping to learn from this experiment. Neurons are pretty sophisticated.

Not only do they send electrochemical signals in a way no other cell does, but they can also learn from the experience and change their activity. Which makes them amazing! And also really tricky for scientists to study and understand.

If you look at the previous FoldIt example, it’s clear our brain cells are better at some things than computers are. That’s what inspired the Pong study. A team of scientists in Australia and in the UK wanted to better understand how neurons process inputs and generate outputs in order to create smarter computer circuits.

To test the hypothesis that experience can alter a neural circuit’s behavior, scientists cultured neurons in a dish and used electrodes to wire them to a computer. Using the electrodes, the scientists could send electrical stimulation to the cells and record outputs, too. Enter Pong.

The electrical output from the neurons in the dish controlled the game’s paddle in a simulated arcade. At first, the paddle intercepted the Pong ball only occasionally and completely by accident. Whenever it did, the computer sent nice, evenly distributed electrical stimulation back to the cells at predictable intervals.

When the paddle missed the ball, though, the electrical stimulus wired back to the dish was timed unpredictably. As it turns out, neurons vastly prefer predictable stimulation. So the neurons learned to produce an output that would make that happen, turning the dish of cells into star Pong-players with just five minutes of playtime.

The cells started firing in a new way timed with the arrival of the ball. Not only was the speed of learning impressive, it was also clear that feedback was key to making it happen. When a miss didn’t send a signal back into the dish, the cells didn't learn anything or change their behavior.

These results reinforce our hypothesis about neural plasticity, our nervous system’s ability to change its activity based on input. Over the years, video games have changed, too. We’ve come a long way since Tennis for Two and Pong.

But researchers continue to find new uses for even the oldest games, leading to discoveries about how the brain works, the structure of proteins, and even improving surgery. Science never put down the controller. For more video game science, you can watch our SciShow Psych episode explaining how video games hack your brain.

Thanks for watching this SciShow video. Thanks for sticking around to the end of it and thanks to our team for putting together such a cool episode. I learned so much and I hope that you did too. [♪ OUTRO]