Research Areas

This page briefly introduces some of the topics that Team PhyPA specialises in.

Passive Brain-Computer Interfaces

In the past three decades multiple types of BCI systems have been developed to support severely disabled persons in their daily life. Although they all aim at the same goal—allowing brain-based control over a computer system—there are significant differences between these systems. In particular, Team PhyPA emphasises that BCI technology need not be limited to detect correlates of cognition that is dealing with direct control. Based on these insights, a new categorisation of BCI systems was proposed, dividing applications based on BCI technology into active, reactive and passive BCI systems.

Active BCI. An active BCI is one that derives its outputs from brain activity that is directly and consciously controlled by the user, independent of external events, for controlling an application.

Reactive BCI. A reactive BCI is one that derives its outputs from brain activity arising in reaction to external stimulation, that is indirectly modulated by the user for controlling an application.

Passive BCI. A passive BCI is one that derives its outputs from arbitrary brain activity arising without the purpose of voluntary control, to enrich human–machine interaction with implicit information about the current user state.

One main goal of Team PhyPA is the exploration of the possibilities that arise from applying passive BCIs to human-machine systems in general, explicitly also for users without disabilities.

Team PhyPA has also launched the Community for Passive BCI Research.

Detection of Covert Aspects of User State

Covert aspects of user state (CAUS) are processes occurring within the user that overt measurement methods can only detect with weak reliability. Many cognitive and affective processes, therefore, are such covert aspects: Cognitive load, arousal, happiness... Information about such states can be relevant to the ongoing human-computer interaction, for example, enabling the system to adapt to the user's current workload.

One can divide aspects of cognitive and affective user state roughly into two distinct groups, both of which can carry relevant information. First, there are latent processes called cognitive conditions, such as arousal, fatigue, and more complex states like cognitive load. Second, there are cognitive processes bound in time that are called cognitive or affective events. Investigated examples include perception, error processing, bluffing in a game context, and surprise.

Team PhyPA investigates how such cognitive/affective events and conditions can be utilised to enable a more ‘intelligent’ interaction, by training the system to learn about the user state.

Implicit Interaction

Passive BCIs are usually used to support or augment a primary human-computer interaction. Typically, the machine gains information about the current state of its user and uses this information to automatically adapt to any change in user state. However, using passive BCIs to detect cognitive or affective responses to single events, enables a new type of human–computer interaction to be formed—implicit interaction. Here, a passive BCI is used as primary input modality: Brain activity that arises without the purpose of voluntary control, is nonetheless used to control a computer system.

Implicit interaction aims at controlling a computer system by assessing behavioural or psychophysiological aspects of the user state that are independent of any intentionally communicated command. This constitues a distinct form of human–computer interaction, which in contrast to the classic forms of interaction implemented today, does not require the user to explicitly communicate with the machine. Instead, users can focus on understanding the current state of the system and on developing strategies to optimally reach the goal of the given interaction; relevant control of the machine follows automatically.

Team PhyPA is developing new interfaces for human-computer interaction that allow for implicit control.

Brain-Computer Interfaces in the Real World

Team PhyPA is convinced that BCI can be meaningfully applied in the real world for all sorts of tasks. Moreover, we are interested in exploring the hypothesis that BCIs may in fact work better outside of the laboratory than inside of it, because users may be highly motivated and more involved in the task. The higher immersion may lead to stronger reactions, and therefore, stronger brain signals. Although the amount of noise and interference may also be higher outside of a well-controlled laboratory setting, such influences can be controlled.

Another aspect of applying BCI technology in the real word is that most current BCI equipment is impractical to be used in everyday scenarios. Most electrodes require a significant amount of time to be properly prepared, and because of instationarities in the signals, the BCI needs to be calibrated before each separate session. Team PhyPA is working on new algorithms and methods to improve or even remove calibration.