The NSF-funded project conducted by Kevin Crowston at Syracuse University will investigate the capabilities and potential of social-computational support systems in the context of citizen science, defined as "partnerships between volunteers and scientists that answer real-world questions". The research will examine nature of the computational systems currently used in a number of citizen science projects and will use these insights to improve computational support for different kinds of citizen science projects. The project will focus on the following three goals: (1) developing a practical understanding of the conditions under which social computation can enhance science and education; (2) generating new research models of social-computational systems that support large-scale public participation in scientific research; and (3) developing and testing social-computational systems that incorporate explicit knowledge about human cognitive and social abilities.
The project will produce societal benefits by investigating how involving the public in scientific research can advance scientific goals while contributing to the science education of the volunteer participants, determining the conditions under which citizen science can prove beneficial for large-scale data collection and analysis, and providing guidelines for improving the design and implementation of computational support systems for citizen science projects.
In their research, scientists collect a large number of photographs and images. While computers can be used to help scientists analyze pictures, humans, however, are much better than computers at deciding what is in pictures. An important task of citizen science is about how to enable and motivate the general public to assist scientists in analyzing large data sets such as classifying images. Citizen science participants, generally, are enthusiasts who have intrinsic interest in science projects. However, since many projects are not inherently interesting to people, it may be difficult for these projects to attract enough enthusiast users to be viable, so it becomes important to study the potential of another group of people (a much larger group than the enthusiast one) who do not have particular interest in science but have interest in online games or entertainment. Citizen Sort, a set of citizen science games created at Syracuse University, enables online participants to produce valuable scientific data to aid scientists in their work while playing Happy Match or Forgotten Island to help classify images of animals, insects, and plants. Happy Match is a points-based tool-like game, and Forgotten Island is a story-based adventure game that uses the classification task only as a way to advance in the game play. Happy Match allows science enthusiasts to focus on the task of image classification and thus generate much more data, while Forgotten Island can bring in more potential users such as gamers through the public attention it receives (for example, it was featured in The Observer newspaper about gamification of scientific research projects). Citizen Sort has attracted 4,500+ registered users, including many school and college students, who have contributed over 80 scientific image classifications on average. To learn more, visit the site at www.citizensort.org and the Forgotten Island video trailer at http://youtu.be/pM57DY-evik
