New Grant: AutoMentor: Virtual Mentoring and Assessment in Computer Games for STEM Learning

Working in conjunction with the Massachusetts Audubon Society, we will develop an AutoMentor module based on a program called AutoTutor which helps teach students about science and technology topics. Using a Wizard of Oz methodology (in which data is collected on student/mentor interactions over multiple instances of game play), we hope to determine a method for producing effective professional feedback from non-player characters. Professional feedback helps students develop professional identities for themselves: engineers think like engineers, are interested in engineering, and know about physics, mathematics, chemistry, and other technical fields. In short, to be an engineer, you have to think like an engineer and see yourself as such.

AutoMentor represents an approach to the problem of providing the opportunity to learn STEM subjects. By creating a virtual professional mentor for students, we hope to make STEM education more motivating and inclusive. AutoMentor will initially be attached to Urban Science; eventually the software will be widely available and represent an important component of STEM games in the future. The development of AutoMentor will also represent an important contribution to our knowledge about game-based learning.

The Principal Investigator and Co-PIs on this project are Dr. David Shaffer, Dr. Michael Gleicher, Dr. Arthur Graesser, Dr. Robert Mislevy, and Kristen Scopinich.  The grant, for $3,500,000, was awarded on September 1, 2009 through the National Science Foundation’s Division on Research in Learning – Discovery Research K-12.  The summary of the grant proposal can be viewed here.