NEW YORK (GenomeWeb) – In the year 2032, a virus has been unleashed on an unsuspecting populace and it is spreading fast. In the midst of this pandemic, an advanced research laboratory has recruited a team of scientists to investigate the origins of the virulent virus and figure out how to stop the outbreak. They are armed with a "gun-like bubble launcher" which they will use to extract DNA from potentially infected animals and compare it to a database of sequences using an algorithm developed, quite possibly, well before some of these scientists were born. Its name is the Basic Local Alignment Search Tool or more commonly, Blast.
There's a little more to it than that but that's the basic set up for Max Five, a new game slated for release this fall that's being developed by researchers in the Scientific Collaboration and Creativity Laboratory (SCCL) at the University of Washington. The project is funded by the National Science Foundation and is led by Cecilia Aragon, a UW associate professor and director of the SCCL. The developers also worked with the Northwest Association for Biomedical Research (NWABR), an organization that works to promote biology in classrooms and has developed a bioinformatics curriculum specifically for use in high schools.
The focus of the UW project was to introduce high school students to cyberinfrastructure and bioinformatics concepts and tools through, among other methods, educational games that teach important concepts in a manner that is both appealing and interesting enough to compete with the visually impressive graphics and compelling narratives that characterize popular commercial games, said Daniel Perry, a doctoral candidate in UW's College of Engineering and one of Max Five's developers. Similarly, members of the life science community are starting to develop and use online games to convey complex concepts in simple and interesting ways and to encourage broader participation in science from the general public.
"There are a growing number of science games out there and some really exciting developments in that realm, [but] at the same time there is this big disconnect when you talk to high school students about the commercial games they are playing as opposed to learning and educational games," he told BioInform. Not surprisingly, the latter are played much less often outside traditional classroom settings. "The grant was not just [about] how we integrate learning concepts into a game but how do we … make it engaging and appealing to students?"
To tackle this, the development team went to the source and incorporated student feedback and ideas from the onset of the project. They recruited seven students from local Seattle high schools as co-designers and gave them some basic bioinformatics and programming training and workshops. These students then worked alongside UW graduate students for a little more than a year and a half providing feedback and ideas about what to include in the game and contributing both code and artwork. The developers also took Max Five to seven high school classrooms with students from various socioeconomic backgrounds in Seattle and let students there play the game and give feedback. In total, they surveyed more than 80 students who played the game and interviewed another 25 students from various socioeconomic backgrounds
Essentially, what the surveys and interviews showed is that there really is no one-size-fits-all approach when it comes to developing science learning games, Perry told BioInform. Educational games are sometimes hampered by the misconception that simply because teenagers like to play games, virtually any game that's developed will captivate them, but "what we really found in our process is that students are engaged by diverse aspects of game play," he said.
For example, when playing Max Five, some students are driven by how many DNA samples they can accumulate and by how successful they are at using Blast — they use an imitation version of Blast developed specifically for the game and not the actual resource. Other students enjoy the science but find the game's visual environment far more impressive and enjoy exploring it. The first environment that students encounter when they enter the game is the lab where they receive their mission instructions. After that, they are taken to a desert environment where potentially infected animals reside. Other environments, such as new laboratories, get unlocked as the game progresses.
A third group of students is drawn to the logical parts of the game — at one point, players find out that a bioterrorist organization is developing new viruses and they have to solve puzzles to find samples. Finally, a fourth group is animated by the social and collaborative nature of the game — students play in pairs and can share clues and information with each other to unlock new stages of the game.
Games with a Purpose
Last month at the annual Intelligent Systems for Molecular Biology (ISMB) conference in Boston, Benjamin Good, a senior staff scientist at Scripps and one of the developers of the bioinformatics game The Cure and other games, gave a presentation focused on online games and how they could be effective tools for teaching bioinformatics concepts in interesting ways.
In his talk, he discussed how games can serve the two-fold purpose of getting people involved and actually interested in learning and using bioinformatics; and also engaging and holding their attention and, as a side effect, increasing their knowledge of a field. An example of the former is a role-playing game called Distributed Annotation Systems (DAS), which was developed to teach the basics of data integration to biologists. In 2012, Maria Victoria Schneider, now the head of scientific training, education, and learning at The Genome Analysis Center in the UK, and Rafael Jimenez, now chief technical officer and a technical coordinator for ELIXIR, published a paper in PLoS Computational Biology describing the game in detail.
Perhaps one of the best known examples of the latter is Foldit, a protein-folding game, played by scientists and non-scientists alike, that was created in 2008 by researchers in the computer science and engineering and biochemistry departments at UW. In 2011, members of the Foldit community reportedly solved the structure of a protein-cutting enzyme from an AIDS-like virus that had eluded researchers for more than a decade.
Another example of a game in this category that's also focused on folding — but as it pertains to RNA — is EteRNA, which came online in 2010. It was developed by researchers at Carnegie Mellon and Stanford Universities. A study published earlier this year in the Proceeding of the National Academy of Sciences about EteRNA by researchers from those universities and Seoul National University demonstrated that combining gaming with experimental testing yielded more accurate synthetic RNA structures than was possible with current in silico approaches.
Other genetics-centric games include Phylo, a sequence alignment game that was developed by researchers at McGill University. Players are expected to find the best possible alignment for sequences that are represented as opposing rows of colored squares that gamers move around within a grid. Meanwhile, researchers at the Scripps Research Institute have developed and released at least three games — called Combo, Dizeez, and GenESP — to improve current knowledge about gene-disease links and combinations of genes associated with particular phenotypes. Recently, they published a paper in the biology archive about The Cure — which used to be called Combo — that discusses the results of using the game to identify genes to predict breast cancer survival.
Learning through Play
"The holy grail to me … would be a game that you go into and you start playing for fun but was deep enough and complex enough that by the time you won the game, you also had mastered some advanced concepts in whatever it was we are trying to teach you," Scripps' Good said during his conversation with BioInform. But, "it's very difficult to get there," he added.
Games can serve as a bridge between subject semantics and backgrounds of the intended audience — for instance, explaining programming concepts to biologists — helping to demystify subjects that at first glance can seem complex and scary, Schneider told BioInform. But developing effective and interesting games based on complex concepts takes time.
Developers need to think about things like what goals they want to accomplish, how long it takes to play the game, and who the target audience is, she said. Also, games don't have to be complex creations. It's true that some of the more successful games have involved partnerships with game development companies and access to the resources needed to make the games attractive and inviting, she said, but they can also be simple activities done in a classroom setting with paper like DAS.
Good also noted that while there is evidence that games can help capture the interest of non-scientists, it's much harder to find empirical evidence that they succeed in actually teaching concepts. For one thing, he said, it's expensive to run these kinds of thorough evaluation studies. Also, at least in the bioinformatics field, there are very few actual teaching games.
But there is evidence from related subject areas that suggest that education games could be an effective means of helping students master difficult and often abstract scientific concepts. One of these is a game called Immune Attack, a first person shooter game that was developed to teach immunological concepts to high school students — the game was developed through the Federation of American Scientists' learning technology program. In the game, students navigate a three dimensional environment of blood vessels and connective tissue with a spaceship-like nanobot on a mission to save a sick patient by retraining non-functional immune cells.
Melanie Stegman is a biochemist and game developer who was hired by FAS in 2008 as the project manager for Immune Attack. She has founded her own company called Molecular Jig Games, is working on a new game called Immune Defense, that she hopes to market to high schools. As the name implies, it's another immunology game but it has a slightly different approach than Immune Attack and is a little harder to play, she said. Stegman and her team are currently testing a 2D prototype of the game in high school classrooms and are working on a 3D version that they hope to release in December at the American Society for Cell Biology meeting in Philadelphia.
A second set of games focuses on concepts from organic chemistry. These games are being developed by Julia Winter, CEO and founder of a company called OChemPrep. Winter, who is a chemist by training, is working on computerized versions of games she used to play with her students in the classroom that graphically and interactively teach organic chemistry concepts. One of the games she developed is called CHAIRS that is slated for release in September — it's a puzzle game focused on the reactions undergone by cyclohexane chairs. She is also working on a more involved game called Cyclo6 which teaches students how chemical bonds are formed and broken in reactions.
In developing their games, both Stegman and Winter, like Perry and his colleagues at UW, consulted with and incorporated input from the students they hope will play them. In her interactions with students, Stegman found that they were not only able to grasp basic immunological concepts but they also became interested in more of the details. "[They were] asking me really hard questions about molecular biology and proteins and receptors and how many receptors are on a neuron and what happens to neurotransmitters after they bind to the receptor," she said. "But they also realize [that sometimes] I don't know the answer. And they are floored," she added. But that helps drive home the point that "science is a search for answers …not a collection of answers." Often the next question they ask is "how are we going to find out?"
Games are also a way to make abstract concepts more of a reality, Stegman continued. For example, when grade school children play the game Angry Birds, they learn a little bit about how parabolas work and that forms a foundation that teachers can build on. ImmuneAttack and ImmuneDefense do something similar for molecules and proteins. They show students how cells and proteins interact, diffuse, and move in a random fashion to accomplish their roles in the body and that gives them a frame of reference for when they actually learn these concepts in class, she said.
But are these games truly effective? In the case of Immune Attack, the evidence seems to suggest that students' understanding of basic immunology increased. Stegman found that students who played the game performed better than a control group on an immunology test taken afterwards and were able to recall facts and concepts. For example, after playing with monocytes in the game, students could remember that it is a type of white blood cell and were also able to recall what it is used for in the body, she told BioInform. Details of her findings were published in Faraday Discussions.
For Winter, evidence of the success of her organic chemistry games is mostly anecdotal but "it's really good solid anecdotal evidence" based on 10 years of experience, she said. In addition to raising funds to launch her games, Winter is now working with professors at Wayne State University and the University of Illinois, Springfield to design protocols to more concretely measure the effectiveness of her games.