How To Use Video Games To Solve Scientific Problems

When was the last time you heard someone use video games to solve scientific problems? Probably never. The cliché is that video games are for mindless entertainment. And scientific problems are for scientists to solve.

So, how can these two totally unrelated entities come together productively and practically?

Well, that’s what this essay is all about. I will begin by covering a real-life case study of how scientists actually solved a long-standing scientific problem using a video game. Following this, I will touch upon certain latest advancements in the world of video games.

Finally, I’ll propose my case for how these advanced mechanics can be leveraged to solve complex scientific problems efficiently in the future. This will be fun. Without any further ado, let us begin.

This essay is supported by Generatebg

A Lesson From History — Use Video Games to Solve Scientific Problems

When I was a child, people did not see playing video games as a positive hobby. As a video game-loving child, I could not understand this stream of logic. Even though I primarily played video games for entertainment, I could see the practical utility in them; I just couldn’t prove it.

Fast forward to 2011, researcher Firas Khatib and his colleagues at the University of Washington managed to prove my intuition. Khatib worked together with computer scientist Seth Cooper (also at University of Washington) to develop a collaborative online game called Foldit.

The game looked like a curly Tinkertoy set on the surface. But underneath the playful optics, the game modelled these play mechanics based on real molecular physics. Players were rewarded for lower energy states of their molecular “folds” and penalised for higher energy states.

The monkey-virus puzzle was one of those long-standing mysteries that scientists could not solve until this point. Having tried to solve it using Rosetta (a protein-solving program), the researchers threw the baton to Foldit as a last ditch hope.

The Foldit online community collaborated together to solve this puzzle in under 10 days! The solution to this puzzle enabled the researchers to move one (small) step towards a permanent cure for AIDS (Acquired Immune Deficiency Syndrome).

What’s more interesting is the fact that the best Foldit players were neither scientific researchers nor software professionals. Seth Cooper had the following to say about the project:

“People have spatial reasoning skills, something computers are not yet good at…Games provide a framework for bringing together the strengths of computers and humans.”

— Seth Cooper (lead developer and designer of Foldit).

The Reality Today

This story happened in 2011. I am writing this essay at the end of 2022. Why haven’t we seen the trend of video game-based optimisation solutions pick up? Well, there is a rather specific reason for that.

If you read the story carefully, you would have noticed that Foldit basically outsourced the ‘grunt work’ of optimising for a low-energy state to the video game players. These players need not be good at understanding physics or science to solve such problems.

Today’s state of the art does not use such video games or online communities to solve such problems. It uses Artificial Intelligence instead. The task is similar, except that almost no human beings are involved. It’s just so much more efficient. Or is it?

We will explore the answers to this question as we progress in this essay. But for now, let us discuss some interesting advancements in the world of video games.

---

Speedrunners — The Untold Source of Hyper-intelligence in Gaming

Speedrunning refers to a video game playing style that aims to complete video games as fast as possible with little attention to any qualitative and stylistic details.

Most recently, I saw a montage of videos from a speedrunning community in Hitman 3 (a popular third-person stealth-action game). An established speedrunner/streamer had designed a “contract” such that the Hitman had to take out 5 targets in the same map as fast as possible.

He had designed the level in such a way that each target was in a different part of the map. One target was underground, whereas another one was on top of a hill deep inside a church, etc.

He put a bounty/award of $500 to the fastest speedrunner 5 days from the announcement. Finally, he streamed the best entries (all in video format).

Right from the offset, I was blown away by these speedrunners’ intricate understanding of the game’s physics and world. It was clear that they were ‘experts’ at what they were doing.

Not only did they have vastly detailed knowledge of the game, but they were also able to come up with ‘out-of-the-box’ creative ways of accomplishing their mission.

The last three entries involved professional speedrunners. At this point, I was convinced that the developers of this video game never intended their designs to be used the way these players were using them. These players were rebellious geniuses in their own right.

As I experienced these speedrunners push the human limits, I realised how much intelligence one has to possess to be good at speedrunning. At the end of the whole show, as the entire community was celebrating their best, one (professional) speedrunner walked away with $500. And that got me thinking!

Surely, the best effort deserves better than $500. And even more certainly, the rest of the brilliant speedrunners deserve better than nothing. Quite naturally, the following question popped into my head:

“Can we utilise the intelligence of these speedrunners to solve complex scientific problems?”

Before we answer this question, let us discuss why we would benefit from this venture in the first place.

Why Should we Use Video Games to Solve Scientific Problems?

The truth is that most complex scientific problems can be outsourced to experts, Artificial Intelligence, and other algorithms. However, why am I making the case for video game players here?

Well, one of the primary reasons is Artificial Intelligence. With the advent of systems such as ChatGPT, Copy AI, etc., we see professionals such as writers, programmers, artists, etc., being challenged by Artificial Intelligence.

In other words, there might soon come a time when most of humanity’s workforce is rendered obsolete by Artificial Intelligence systems. So, my proposal is to offer a dignified solution to our “retiring” workforce: play video games and solve humanity’s toughest problems in the process.

Besides, there still exists a chunk of complex problems that are beyond artificial intelligence and other algorithmic systems. So, human ingenuity still has a valid and necessary place in science.

Using video games gives us access to intelligence that we would otherwise never have. Let us see how to do this in the next section.

---

How to Use Video Games to Solve Scientific Problems?

Let us say that we have a complex scientific problem to solve. Here is a step-wise algorithm (as an example) of how to get video game players to solve this problem:

Step 1: Transform each factor/variable in the problem into a video game element.

Step 2: Model the physics/rules of the video game such that they represent a one-to-one relationship between the real-world equivalent factors/variables.

Step 3: Treat the actual video game as the variable and keep iterating until players find the problem fun to solve.

Step 4: Figure out a real-world reward system that benefits the players while they solve the scientific problem.

All of a sudden, while speedrunners try to take out targets as fast as possible, they solve for the fastest real-time virus tracking in reality. I’ll let your imagination go wild with the possibilities of such as system.

Of course, the algorithm that I have just proposed over-simplifies the task. This task is far more complicated than what you and I can currently imagine.

The point that I am trying to make is that this framework is very much possible. The question, then, becomes whether this is feasible.

Final Thoughts

One of the major challenges with such projects is the cost involved. We are essentially multiplying the costs involved by creating a newer layer of abstraction in the form of a video game. So, the whole venture has to be profitable considering the outcome in the end.

The fact is that there are seriously underutilised geniuses playing such video games. They would never apply for these research positions nor would they be motivated by the “boring” science. Without gamifying complex problems, we lose access to this class of intelligence altogether.

So, all in all, even considering the limitations, I think this approach is well worth exploring. What’s more, it appears as if it’s just a matter of time before the rapid development of artificial intelligence dethrones the majority of humanity’s workforce.

If that happens, solving (general) problems using video games might become more of a necessity rather than just an option!

---

If you’d like to get notified when interesting content gets published here, consider subscribing.

Further reading that might interest you:

• A Career Based On Passion Is Risky. Here’s Why.
• How To Really Use Mathematical Induction?
• The Three Prisoners Puzzle: How To Really Solve It?

If you would like to support me as an author, consider contributing on Patreon.

References: Cooper, Khatib, et al. and Atrioc.