Ok, no more kids’ stuff this week! My past two reviews (TeachMe: Kindergarten & The Backyardigans Space Adventure) have already been on educational games, both of which had curriculum targeted at early learners’ mastery of rudimentary reading and math concepts. So, what about the other, very far end of the spectrum? Can the game paradigm be applied to teaching highly complex scientific concepts? Amazingly, the answer is yes. There are even a handful of examples in computer science that take that idea a step further and actually have the primary goal of contributing to the greater good of humanity. GWAPs, or a ‘game with a purpose’, “is a game played on a computer that serves some purpose for the people setting up the game by harnessing human abilities in an entertaining setting.” (Wikipedia, n.d.) Some of these software applications typically leverage a computer science technique known as human-based computation, “in which a computational process performs its function by outsourcing certain steps to humans. This approach uses differences in abilities and alternative costs between humans and computer agents to achieve symbiotic human-computer interaction.” (Wikipedia, n.d.) There are certain tasks, like image recognition, that humans perform innately with ease, yet computers “are almost incapable of”.
Foldit is a stellar example of a GWAP that actually enables its players to contribute to important scientific research and help find cures for debilitating diseases like Alzheimer’s and Parkinson’s. It was developed at the University of Washington to help biochemistry researchers apply crowdsourcing and distributed computing techniques towards the “computationally demanding” process of predicting protein structures.
"Foldit is an attempt to apply the human brain’s natural three-dimensional pattern matching abilities to this problem. Current puzzles are based on well-understood proteins; by analysing the ways in which humans intuitively approach these puzzles, researchers hope to improve the algorithms employed by existing protein-folding software."
According to, Zoran Popovic, one of the UoW professors and creators of Foldit, players are “predicting the actual structure that proteins can take in real life. And because structure is fundamental to how proteins interact and function and do things in real-life cells, whoever knows the structure of these proteins will know the secret of life.”
Players manipulate three-dimensional models of proteins and simulate protein folding, “the physical process by which a polypeptide folds into its characteristic and functional three-dimensional structure from random coil”. (Wikipedia, n.d.)
The two primary biological “big problems” that Foldit seeks to address are:
- Protein structure prediction: …knowing the structure of a protein is key to understanding how it works and to targeting it with drugs. A small proteins can consist of 100 amino acids, while some human proteins can be huge (1000 amino acids). The number of different ways even a small protein can fold is astronomical because there are so many degrees of freedom.Figuring out which of the many, many possible structures is the best one is regarded as one of the hardest problems in biology today and current methods take a lot of money and time, even for computers. Foldit attempts to predict the structure of a protein by taking advantage of humans’ puzzle-solving intuitions and having people play competitively to fold the best proteins.
- Protein design: Since proteins are part of so many diseases, they can also be part of the cure. Players can design brand new proteins that could help prevent or treat important diseases.
Simulation/Education Game Review
from University of Washington Computer Science & Engineering and Biochemistry Departments
systems supported: Windows XP & Vista, Mac OS X 10.4 and up, Linux
Foldit is available for free, directly from the developers at http://fold.it. Gameplay requires the creation of an account, which is also integrated with the website where progress/achievement is tracked, and each player is ranked. The website also provides several social communication methods to connect with not only the game developers, but also the growing community of players.
The game’s main screen is broken down simply into three main sections: introductory puzzles (of which there are 27), more complex science puzzles (based upon well-understood proteins) and contests, where presumably the bulk of the advancement of protein-folding is conducted (as a Foldit ‘n00b’ with a global soloist rank of 6798, I avoided that area for the time being). Just as in a traditional video game, introductory puzzles are broken down into levels, starting the player with relatively simple concepts like “sidechains”. As the player graduates from level to level, they are introduced to increasingly complex scenarios and objectives, which build upon previously learned protein-folding skills.
The interface is simple, almost spartan, depicting a three-dimensional model of a protein in the middle of the game’s canvas, along with a small collection of menu palettes to assist the player and/or connect them with the community. Players can rotate and maneuver the protein and its various components within the 3D space by clicking and dragging them. There is no audio instruction, nor avatar (one would expect a goofy character like Mr. DNA from the film Jurassic Park) to help guide beginners. Instead, the game mainly relies upon basic pop up bubbles to coach the player through gameplay. Audio cues help to provide immediate feedback and signify success or failure. Successful completion of one level automatically advances the player to the next.
The player’s score, number of moves and time spent on each level is communicated upon successful completion. The menu system gives the player complete control over each level (reset and replay, undo step-wise actions, set/restore best (discovered) solutions, etc), as well as connect with the community especially for solution assistance.
There is virtually no bias with respect to gender, ethnicity and stereotype - proof, I suppose, that at a biological level, we really are truly equal. In fact, the game has a sub-project that aims to translate Foldit into as many languages as possible.
The game is almost exclusively constructivist in its implementation, giving the player only a sparse amount of information to complete each level. The player discovers (and in most cases interprets) solutions through exploration, trial and error. This is perhaps an intentional decision by the game’s developers, so as not to cloud the player with any scientific misconceptions, or bog them down in biochemical minutiae. Because of this, it is almost impossible to quantify the physical fidelity of protein-folding simulation. The protein body, and the components (like sidechains) connected to it rotate, bend, attach and behave according to a set of apparent physical rules. Though they are introduced in the very first level, the game doesn’t even bother to explain what a sidechain actually is. Further, whether or not sidechains actually chemically behave in the manner they are represented in the atomic ‘real world’ is practically unknown to most. Players (especially those with no background in biochemistry) have no choice but to take a leap of faith and simply must accept these observed behaviors as accurate. Scoring is just as ambiguous. It can be difficult to identify (especially at higher levels) what the exact scoring rubric is. The player is given only a score for which to aim, and a set of objectives to complete.
At the very beginning of this post, I did mention that I wouldn’t be reviewing a kids’ game. However, the true beauty of Foldit is that it breaks down the science of protein folding in such a way that even a child could play. By completely obfuscating its biological and biochemical underpinnings, the game developers in effect have widened Foldit’s access and broadened its appeal far beyond the tiny biochemist community. In a recent NPR interview, Professor Popovic was asked, “…so you’re not just looking for biochemists out there or even computer scientists. I mean, you’re just looking for people who have a lot of time to get addicted to something like this?” Popovic responds, “That’s exactly right. I mean, it’s actually still not clear what kind of people are very good at this. I flew a huge number of them to Seattle to just - we call them protein savants. And even when I observe them directly as they’re trying to solve problems, you know, they see stuff that I just don’t see at all. One of the early outcomes of this was that we realized that regular people playing all over the world are way better than biochemists at this.”