How a US teenager achieved the impossible and what his feat tells us about human capabilities

By Tom StaffordFeatures correspondent

A mind-boggling achievement in a classic video game reveals wider lessons about the extreme limits of human performance, says cognitive scientist Tom Stafford.

For decades, it was a feat that was considered impossible.

In the dying days of 2023, US teenager Willis Gibson – online handle "Blue Scuti" – "beat" the Nintendo Entertainment System version of the video game Tetris, which was first released in 1989.

The original Tetris designers thought it couldn't be done – the game is designed to play endlessly. The pieces fall faster and faster until a player is overwhelmed. To beat the game, a player has to achieve scores so high that the game's memory banks overload and it crashes. Victory is achieved because the computer simply cannot continue.

As a professor of cognitive science, I'm interested in how people acquire expertise, particularly in video games, so when Gibson performed his dizzying feat, it immediately caught my eye. How this 13-year-old did it tells us a lot about how the limits of human performance are changing in the digital age.

Previously, the NES version of Tetris had only been beaten by AI. A specially designed program was able to perceive, near-instantly, the state of the Tetris game and select actions as fast as the console could register them. It played tirelessly, never making an error – something that seemed far beyond the constraints of mere human performance.

At the time, 2021, the AI-Tetris player was able to show humans previously uncharted levels of the game. Like physics at the limits of a black hole, the reality of Tetris begins to bend at the higher levels. The speed suddenly doubles at level 29, a level few humans reach and fewer survive for long. When the score counter breaks 1 million, the digits begin to be replaced by letters, and then finally glyphs from the Tetris graphic set. Eventually, the colours of the blocks warp and change, some levels are all violent pink, others have blocks so dark you can hardly see them – especially at the speed you need to act to survive.

This is the context for the game Gibson streamed on 21 December 2023, in which he played the game at increasingly frenetic speeds for 40 minutes. In the process, he set new world records for high score, levels played and lines cleared. Eventually, he was rewarded with the crash that signals he had done the impossible – beaten the game.

You may also like:

The achievement is real, but the importance is far beyond the realm of classic arcade games and those who love them. What Gibson did, and how he did it, gives general lessons for how people learn, and how the limits of human performance are extended.

To understand why, consider that Tetris is a community, as well as a game. Gibson didn’t just pick up an old console, he joined a living tradition. There are thousands of players dedicated to Tetris on the NES, the platform on which it was first officially released in North America. As well as the players, there are streamers, bloggers, theorists discussing strategy, and record-chasers competing to outmatch each other. There is even the Classic Tetris World Championship, held annually in Portland, Oregon.

Tetris is famously compelling, but it is the community that ensures new players are drawn into the hobby, despite the proliferating number of alternatives out there.

Communities provide encouragement, and another crucial ingredient for accelerating human potential – inspiration. Communities are living laboratories of ideas and experimentation, where different people can try out new things, and successes can be passed along. Scientists study this social learning under the label "cultural evolution". Although some other animals do this, humans excel at it. It provides the seeds of culture, as different communities evolve different practices to suit their immediate environment.

Successful new skill techniques spread in learning communities. Take the Fosbury Flop, a high jumping technique popularised by Dick Fosbury when he won gold for the US in the 1968 Olympics. Fosbury’s gold winning jump was 2.24 metres (7ft 4.19in), but this height was achieved – and in many cases smashed – by every single competitor at the 2020 Olympics in Tokyo, all using the Flop technique. 

What Fosbury did for the high jump, the internet and AI are doing for nearly every skill imaginable. Recently, the internet has put rocket fuel in the engine of cultural transmission. Whether you want to learn to code, or just want to watch a video on how to fix your dishwasher, it's easier than ever to copy new skills. Since the advent of home chess engines there has been a massive generational improvement in chess skill, which sees today's players playing better chess than at any point in history. Even criminals employ social learning – one recent news story blamed Tiktok for a rise in a particular style of car thefts.

In Tetris, a crucial innovation has been a new way of holding the controller – called "rolling", where players drum the bottom of the pad into a finger or thumb hovering just above the keypad. It allows players to input commands faster than by pressing with a single finger. Despite the game's 34-year history, rolling only became popular in recent years, spreading through the community of streamers and competition Tetris players. As is often the way, it is the youngest members of a community who recognise useful innovation. Gibson began playing Tetris at age 11 and used the rolling technique in his record breaking run.

Lots of discussion around artificial intelligence focuses on domains where human skills might become obsolete, but it is a mistake to think that human performance is a stationary target.  As Gibson's record-breaking achievement showed, we constantly look to the outer limits and, by reaching, extend our grasp.

The moral is that pressing the frontiers of human skill is a consequence of continual collective innovation, as well as of remarkable individuals. Humans are a species defined by our ability to learn, and in the digital age there is more and more potential to push into uncharted territories of performance in all aspects of art, science and culture – Tetris included.

* Tom Stafford is professor of cognitive science at the University of Sheffield, UK. He writes a newsletter called Reasonable People.

--

If you liked this story, sign up for The Essential List newsletter – a handpicked selection of features, videos and can't-miss news delivered to your inbox every Friday.

Join one million Future fans by liking us on Facebook, or follow us on Twitter or Instagram.

ncG1vNJzZmivp6x7o67CZ5qopV%2BbwrXB0Z5mmqqknrCtsY5rZ2tsYGZ9dHnTnquroaNitbDDjJpkrqtdqbKmusCgnKtlkZi1qrHVnptmrJiaeqq5z6iqrKGSobI%3D