Could the National Institutes of Health (NIH) fund your next game?

During the European Citizen Science conference last June, together with the Human Computation Institute, we organised a workshop and a networking event about games that help scientific research. At the workshop, Dr Jennifer Couch, the National Institutes of Health (NIH) citizen science coordinator talked about games that engage the public in biomedical research. Jennifer was also the keynote speaker at the networking event, and she presented examples of successful partnerships between game designers and researchers. This was my second encounter with ‘the feds’. I had met with Dr Jennifer Couch and with Dr David Miller at GDC in March, where we were attending the panel Scientific Research Using Games. The NIH sees the potential of citizen science games and understands the importance of collaboration with the games industry. I decided to find out more about their vision, their program and the financial support they can give to games. Jennifer and David accepted to answer a first round of questions by email and to follow up with a call. This interview aims at showing that games and game technology are needed to help solve complicated scientific problems and if related to health, can be funded by this government agency.

Background information about the NIH and the mission of program directors

The NIH is the largest supporter of biomedical research in the U.S. and one of the world's foremost medical research centers. They support research worldwide, research that spans from developing basic enabling technologies and data science, understanding the biological mechanisms that underlie health and disease, as well as clinical, epidemiological, behavioral research and the impact of the environment on health and disease.

NIH Program Directors work together to support and enable scientific research. They assess scientific projects and make recommendations about what projects to financially support, but also assess the barriers and challenges that are slowing down the pace of scientific research. They constantly seek opportunities to bring in methods that have been successful in other areas, to engage problems solvers and to build partnerships with other fields. They encourage researchers to use these new methods and to collaborate across fields and disciplines.

Jennifer and David’s area of expertise is in cancer research. Both see the enormous potential for the use of games themselves, as well as the technologies, design, and thinking behind games to accelerate cancer research, spark new discoveries, bring in creative problem solving from a broader audience, and enable researchers to interact in a more effective way with the vast amounts of cancer data being generated. One of the most recent game that benefited from the support of the NIH is the simulation game Cancer Crusade, a mobile game based on a mathematical model to understand cancer progression. Gameplay data is sent to the Moffitt Cancer Center and analyzed to improve real therapeutic schedules.

Games and technology are used to crowdsource research, but can also be used to do the research. In some experiments, scientists deal with complex, high dimensional data, that they try to render into models, for example, tumor progression, anatomical structures. Rapid rendering, ability to visualise, multiscale, manipulate and rotate things around, is important when they are looking at their own data, and that's something games do well. Technology that comes from the game space can help researchers even when they are not building a game.

Screenshots from Cancer Crusade, treatment simulator that let players design treatments for virtual cancer patients.

The NIH held a think tank which revealed great potential for collaboration between game and science

"The game development world is vast and diverse, with talented game designers focusing on all sorts of interesting areas. The need for sophisticated software, thoughtful design, and community building inherent in game design and development all have parallels in the scientific research world. It struck us a few years ago that while a handful of scientific researchers had successfully utilized games to engage the public in solving complicated scientific puzzles, there must still be significant barriers to the broader use of games as a method of engaging the public in scientific research. At the same time, we recognized that the public is very interested in contributing to scientific research for a variety of reasons and they are often highly motivated to contribute in all sorts of ways, including lending their creativity and problem solving capability to analyzing data, if provided with the right tools and the right environment.

To understand the potential for developing more scientific research games (games that focus on either analyzing data or solving research questions as opposed to games that are strictly for entertainment or education), we decided to hold a think tank. We were lucky enough to identify two amazing co-chairs, Ben Sawyer and Markus Covert, representing the game space and the biomedical research space, who worked with us to develop and run the think tank. We brought together a group of game designers and a group of biomedical researchers. From that think tank we gained a much better understanding about the way the two communities think (and where they think differently) as they explained to each other their motivations, struggles, and methods for developing games or for carrying out a scientific research project. It was eye-opening for the participants as well, both in terms of the potential for working together and for just how challenging and sophisticated both these fields can be. The participants identified barriers and opportunities related to collaboration, they made some recommendations and their findings can be found in the Big Data to Knowledge Think Tank: Game Developers and Biomedical Researchers 2 pages report.

The results of that think tank, together with our own insights from attending meeting such as GDC and scientific conferences, led us to do an experiment: we issued a special funding opportunity that explicitly asked for proposals bringing game design, crowdsourcing, and scientific research together. A wide range of projects, scientific questions, and crowdsourcing methods came in through that funding opportunity and we funded ten of them (the list of projects can be found under Big Data to Knowledge (BD2K) Advancing Biomedical Science Using Crowdsourcing and Interactive Digital Media). Watching the evolution of the projects we funded through that program as well as the ever changing landscape of scientific research and game design, we are now better able to showcase to the scientific research world the potential for using games, game design, and game technologies to crowdsource or otherwise enable their scientific research. We are very excited about where this will go in the future. What does seem clear is that building a successful citizen science game takes talent and consideration and usually a strong multi-disciplinary partnership that can take time to solidify. Games are not the answer to all tough research questions and it is still a bit early to know what research questions work well with games. However, we see game designers and biomedical researchers using them in a wide variety of inspiring ways to engage the public in solving challenging research questions."

Quote from the think tank report: https://datascience.nih.gov/sites/default/files/bd2k/docs/BD2K_GDBRsummary.pdf

Contact the NIH program directors, bring a team together and write your research plan

"At any given time, the NIH has many grant opportunities that can fund crowdsourced and biomedical research using games. If nothing matches exactly your idea, you can also submit your proposal under the any NIH's investigator-initiated funding opportunities, such as the R01 grant mechanism which is the most common submission path for NIH grants of any kind. Your proposal will be looked at and reviewed by someone with the right skill and expertise. The process of applying for a NIH grant can be daunting, long and complicated, but it pays off. The best thing to do is to contact and to have a conversation with the program directors. We will understand what you want to develop, will help you determine if our grant is the best option for you and whether your project is something that could be funded. We will help you navigate through the system and advise you on how and where to find the right partner. What we cannot do is helping find a project, a partner or data, and we won't help you build the project, as in the end, we are the ones recommending what projects should be funded.

When submitting a proposal, it's important to be clear on how your project is going to have impact on the area you are trying to have impact on. Citizen science games have a unique challenge and are innovating by nature. Therefore, you don't need to spend too much energy explaining why your idea is innovative. It will be more important to detail why using citizen science or crowdsourcing is the best methodology to solve the problem you are trying to solve. If a game developer wants to get into the biomedical space, the best way is to partner with a biomedical researcher. They know to to write NIH grants, they have officers who can help write, submit grants and reports. Working with biomedical data is hard and requires a lot of legal and ethical considerations. It is also technically challenging to guarantee privacy and security of the data. Once again, researchers and universities have programs in place, secured databases and consent forms. Successful proposals are ones that come with interesting problems, propose a strong plan to solve it and have the expertise to solve it. When proposing a games with scientific purpose, it is important for a game designer to bring their expertise and to contribute actively to the proposal. Partnership is the key."

Spectrum of Biomedical Research Opportunities, from Dr Jennifer Couch talk at ECSA.
 

Finding the right partners and the right question

"Finding partners is probably one of the most difficult parts of this equation. Where does a motivated game designer go to find a scientific researcher with a good question to work on? Where does a motivated researcher go to find a game designer with the patience, adaptability and motivation to work in the sometimes messy world of research data? There’s often a third partner, someone who understands crowdsourcing or citizen science, someone who understands the citizen science audience. These different people rarely stumble into each other at meetings or trade shows or conferences.

As more and more game design schools focus on, or offer options in designing serious games or games for good, it seems there might be some opportunity there. If a game designer is interested in pursuing a citizen science type game, a first step could be to attend a conference like the Citizen Science Association where many citizen science practitioners would be. This is a great place to meet with crowdsourcing or citizen science experts. If a game designer is interested in working with biomedical researchers or doing something in that space, another option might be to reach out to disease or community advocacy organizations. Advocates often have ties to researchers who are more motivated to work with the public and to try different approaches, such as games. Advocates sometimes also have ties to organizations, graduate or medical students who are similarly motivated.

Citizen science games make (and publish) scientific breakthroughs, and the research world is increasingly interested in reaching out, adapting, and working together. Scientists are becoming more aware that this methodology can be successful, thanks to Foldit, Eyewire, EteRNA publications in peer reviewed journals, which is scientists currency. We are seeing an increasing number of sessions at scientific conferences where crowdsourcing and/or games are highlighted.

Finding partners/collaborators, finding the right opportunity or the right question for the right skills and method are always tough questions when one innovates or steps outside their own field of expertise, and this is something we would really like to explore more. Conferences and meetings like Games for Change and the Citizen Science Association strive to reach out and build bridges across these fields. It is also striking to see how the field has expanded. A few years ago at GDC, there were only few small indie projects linked to the theme of cancer. Today, there are games on cancer treatment, diagnosis, games that educate or help deal with anxiety. There are sometimes hackathons and jams developed around scientific research or biomedical questions, these could be a bit of a foot in the door also."

Great example of collaboration for Project Discovery in EVE Online. Can’t see on the picture? watch the detail of the collaboration on Youtube - From 7’48

Let’s talk about return on investment, for-profit and success

"Scientific research works at a much different pace than game design and it is often carefully planned and carried out over the course of several years at great expense. While biomedical researchers also adapt and iterate as their research progresses, it can be costly and sometimes technically or ethically impossible to adjust the course of research in a way that would make designing games around that data simpler. I’m sure this is frustrating and can be a challenge to an individual or company accustomed to working in rapid sprints or being funded all at once for a project (not usually the case in scientific research). We’ve seen enough instances of success in this space to know it can be done, but it likely is not for everyone. Thinking a bit more broadly than financial reward, the return on investment here is the opportunity to accelerate medical breakthroughs, to enable the next great “aha moment” in scientific research, to work with the public on these challenging problems. Citizen science games is a relatively new area with great untapped potential; it seems that there might be real appeal in being innovators in this technically challenging space.

Citizen science participants have a number of reasons for dedicating time to projects. But, I imagine much of the player base for citizen science find and play these games because they are motivated to help. Until we see more examples of embedding scientific problems or citizen science into existing entertainment games (successful but few examples so far), it’s not entirely clear if it would be difficult getting people to help if the project was run by a for-profit company, but we don’t see the particular business structure of the company as a barrier. Tesla crowdsources aspects of their driving AI via an opt-in data-sharing program with some of their customers (short anonymized driving video snippets are shared with Tesla to help with training their autonomous driving AI models). If, hypothetically, the company turned some of that video and image analysis into a mobile game of identifying signs and road markings, I’m sure lots of citizen science participants would sign up, as many would want to be a part of helping to create the next generation of autonomous vehicles. The fact that one of Tesla’s goals is turning a profit probably wouldn’t, in my opinion, enter into it for most players. Recently, Project Discovery in EVE Online has done a tremendous service to citizen science by demonstrating one model for how citizen science projects can be placed in a context of a commercial game. In the past, Cancer Research UK did something similar when they incorporated research data in a bonus level of the mobile game "The Impossible Line".

Successful citizen science games all do some things in common: they make it interesting or compelling, they keep an element of fun, they work a lot on building and maintaining the player community, and they pick the right question or topic to make a game around. For us, what is exciting is that we are able to achieve things other projects were not able to achieve, and they were able to do it because they used a game. This is not about free labor and free data, this is about creativity, a way of approaching a problem differently. Games are good at that, they helped solve problems that could not be solve with other methods."

Because citizen science game makers are often asked about competition with mainstream games, and fun

"There are so many different flavors, styles and genres of games out there, and so many players with so many different motivations and preferences. Game genres don’t really compete with one another. Are driving simulators competing with action-adventure games? While it’s true that all activities are competing for your time, it’s ultimately up to the player to decide how to spend their time and what games to play. For instance, I’m playing Phylo, Reigns: Her Majesty, and Zach Gage’s Flipflop Solitaire game quite a lot at the moment. I wouldn’t say that they are competing for my time and attention, rather I drop into a play session of whichever game I’m in the mood for. Similarly the potential research question space is large and diverse.

Fun is simply what happens during a rewarding activity, and what is considered rewarding is highly variable from player to player. Years ago, in Demon’s Souls on PS3 I ended up repeating a series of character steps over and over for hours in order to find a “pure bladestone” which had a very low drop rate. It wasn’t the sort of delighted “fun” you would think of for a game, but it was very very rewarding once I finally got that item and earned one of the rarest trophies on that console. I think a lot of citizens science game players have this view, and while they do enjoy the play activity of the game, the fun of the game is multi-faceted and naturally includes the greater context of the game’s science project. Game designers can strike a good balance between fun and meaningful activity as it appeals to their different potential players. It seems worth considering how the game will be rolled out and who the target audience is. Pick a compelling problem, one for which we really do not have a solution, one for which the solution will make a real difference; develop a game around it in a way that entices, enables, and motivates players to lend their creativity; and listen to the players." 

To sum up

The NIH can help fund games related to biomedical research. To be funded, one must have an idea and a problem to solve, must demonstrate that using game elements is the best approach for this problem, must be clear on the impact the project will have, and show that the team has the skills required to solve that problem. Finding the right partners is not easy, but there are evidence that the gap between science and games is reducing, and there are great games showing that collaboration can lead to astonishing results. The most recent example is the publication in Nature Biotech, using data analysed by EVE Online players.

It was really easy to talk with Jennifer and David, they are really knowledgeable and willing to bring more games to science (thank you both of you). Don't hesitate to discuss your ideas with them.

More about Jennifer and David

Jennifer Couch oversees a unit within the National Cancer Institute (one of the 27 Institutes that makes up the NIH) that focuses on development of tools, methods, and foundational data that enable cancer research including areas like data science, mathematical modeling, bioengineering, biophysics and open innovation methods.

Dave Miller is a Program Director at the NIH's National Cancer Institute, the federal government’s principal agency for funding cancer research, training, and technology development. Dave has led multiple NIH programs supporting the development of data science software tools, and he manages a broad portfolio of biomedical software and computational modeling research grants in areas including data visualization, science games, interactive media, and crowdsourcing methodologies.

Contact : https://www.cancer.gov/about-nci/organization/dcb/about/contact

For regular news about citizen science games, follow Claire on Twitter: @claire_csg