Here I will discuss some of the projects that the Brain Game Center is working on and areas where there are significant advantages to moving beyond traditional approaches to data analytics. Issues that we are trying to solve are how can one classify people into subgroups based upon a collection of tests? What rehabilitation approaches produce the most desirable benefits for a given subgroup? How can one personalize a training approach so that it is best tailored for a given participant? In regard to the first two issues, the traditional approach in the field is to conduct simple hypothesis testing, for example what training approach leads to better improvements, can a hearing test distinguish between a group of poor vs good hearers, however, these fail to address individual differences. In regard to the last issue it is common to have training programs adapt to the appropriate difficulty for a given individual, but these typically target a particular accuracy level and don’t take into account learning and motivational differences across participants, how can one find the goldilocks zone where training challenges are just right for a given individual? While these questions are well-posed, existing approaches in the field are still very limited and there is significant opportunity for advancement of the state of the art.

Prof. Aaron Seitz