Unobtrusive identification of cognitive states for improved knowledge acquisition

One of the major steps in the evolution of Human-computer interaction (HCI) is the introduction of physiological computing. Real-time measurements and analysis of physiological signals through sensors help opening up an implicit communication channel between computers and their users. Awareness of the user's emotional and cognitive states allow computers to react and adapt in real time, and therefore potentially maximize HCI performance efficiency and communication. Ubiquitous wearable computing enables us to make use of computational power in everyday situations. This liberates experiments and applications from necessary heavy stationary devices and controlled laboratory environments. Physiological computing has been used to increase the efficiency of performance, and improve the pleasure derived from interacting with computers. By analyzing physiological data from the user, cognitive states can be monitored and identified [6]. Thereby, the computer becomes aware of the physical, mental, and emotional context of a user. Consequently, the physical data describing negative or positive states can be used as an input modality to dynamically adjust the system, e.g. by altering certain contents, by providing assistance with additional information, turning on/off of certain functions, or triggering a reminder to take a break or walk when sleepiness or frustration result in decreasing attention. These context-aware systems have a proactive nature and therefore omit the necessity for explicit input devices, such as a mouse or a keyboard. They are able to create an interactive loop between a user and a computer. Since the user is constantly processing the information received (e.g. from a watched video), and the ubiquity of mobile devices allows for sensor data to be constantly monitored and processed, we can create biocybernetic loops that are able to respond to desirable and undesirable states [13].