Feasibility study for an EEG-based semantic tagging game

Semantic tagging of images has a broad spectrum of applications, of both scientific and commercial interest. Experience has indicated that designing t he interaction with human taggers in the form of a web game can be a very effective way to collect val uable tagged data. So far, implementations of such games have depended on explicit, deliberate input b y the users. A novel alternative is presented that doesn’t only use signals from mouse and keyboard, b ut also utilizes brain activity data in the form of electroencephalography (EEG) scans. Brain response data from an earlier experiment [1] is analyzed in an attempt to find and evaluate a measure of sim ilarity between responses of two subjects, suitable for use in a game. From the perspective of the test subjects, depending on the characteristics of the proposed game and the target audience, such a game could spark their interest in a computational approach to cognitive science.