Event Abstract

The use of Passive BCIs for Neuroadaptive Technology

  • 1 Technische Universität Berlin, Biopsychology and Neuroergonomics, Germany

Today's interaction with technology is asymmetrical in the sense that * the operator has access to any and all details concerning the machine’s internal state, while the machine only has access to the few commands explicitly communicated to it by the human, and * while the human user is capable of dealing with and working around errors and inconsistencies in the communication, the machine is not. With increasingly powerful machines this asymmetry has grown, but our interaction techniques have remained the same, resulting in a communication bottleneck. Passive Brain-Computer Interfaces (pBCIs, Zander & Kothe, 2011) can assess brain activity resulting from cognitive processes related to the interaction and interpret them in real time. This information can then be used by the machine to adapt to its user’s intentions, ideas and reactions to support their goals. This process does not require any action of the user, but transfers relevant information to the machine, widening the above-mentioned communication bottleneck. More general, this defines Neuroadaptive Technology – technology that automatically adapts to the current cognitive and affective state of its user (Zander et al., 2016). To ensure a proper interpretation of relevant brain activity it is of crucial importance to understand on which aspects of the recorded brain activity the technology adapts to. In the context of brain-computer interfacing, it thus is important to investigate what regions of the brain the chosen approach focuses on. For one, this will clarify to what extent the approach relies on brain activity, as opposed to undesirable non-cortical signals. More generally, the practice is informative as it allows conclusions to be drawn about the cortical regions—and thus, cortical functions—that contribute to the effect under investigation. I will discuss a method to visualise the regions of interest of standard BCI approaches. The method takes individually reconstructed source spaces and transforms the BCIs classifier filter weights into relevance weights indicating the relative contribution of each source to the approach. This is visualised across participants in an average brain. By decomposing the approaches weights into separate sources and localising these in the brain, the method provides a tool to evaluate BCI approaches and test hypotheses. The method was tested on simulated data as well as on real data. We simulated two data sets of 10 simulated participants each. We simulated the data using the toolbox SEREEGA, short for Simulating Event-Related EEG Activity (Krol et al., 2018). SEREEGA is a MATLAB-based open-source toolbox dedicated to the generation of simulated epochs of EEG data. It is modular and extensible, at initial release supporting five different publicly available head models and capable of simulating multiple different types of signals mimicking brain activity. Each participant’s data was simulated with unique locations of that data’s generator sources, with some consistency maintained for the sources that generated the class differences. In one data set, these class differences were caused by the presence of event-related potentials from two sources. In the other, differences were caused by the presence of alpha-band activity in different sources. The use of simulated data enabled us to compare the method’s results to a known ground truth. We see that under these circumstances, the method accurately recovers the correct sources in the brain. The method was also applied to data investigating responses to perceived errors of different degrees (Zander et al., 2016). It revealed that the used BCI approach prominently used data from the medial prefrontal cortex (mPFC). The mPFC is associated with correlates of predictive coding which is a cognitive process that is expected to be related to the task at hand. Our hypothesis that the BCI approach was assessing information of predictive coding was hence supported by these results. It is important to be able to perform an inspection of a BCIs regions of interest and compare the results to hypotheses taken and other perspectives on the data. When a BCI-based application is designed, we hypothesize what functions (and thus, what regions) of the brain will be targeted. Visualization methods such as this one enables us to compare a classifier’s actual regions of interest to these hypotheses and validate our assumptions—and to gather new insights about the cortical processes underlying the observed effects. This allows to associate the BCI to specific cognitive tasks and design Neuroadaptive Technology purposefully.

References

Zander, T. O., & Kothe, C. (2011). Towards passive brain–computer interfaces: applying brain–computer interface technology to human–machine systems in general. Journal of neural engineering, 8(2), 025005. Zander, T. O., Krol, L. R., Birbaumer, N. P., & Gramann, K. (2016). Neuroadaptive technology enables implicit cursor control based on medial prefrontal cortex activity. Proceedings of the National Academy of Sciences, 113(52), 14898-14903. Krol, L. R., Pawlitzki, J., Lotte, F., Gramann, K., & Zander, T. O. (2018). SEREEGA: Simulating event-related EEG activity, Journal of Neuroscience Methods, ISSN 0165-0270, https://doi.org/10.1016/j.jneumeth.2018.08.001.

Keywords: passive BCI, neuroadaptive technology, EEG, simulation, implicit control

Conference: 2nd International Neuroergonomics Conference, Philadelphia, PA, United States, 27 Jun - 29 Jun, 2018.

Presentation Type: Oral Presentation

Topic: Neuroergonomics

Citation: Zander TO and Krol LR (2019). The use of Passive BCIs for Neuroadaptive Technology. Conference Abstract: 2nd International Neuroergonomics Conference. doi: 10.3389/conf.fnhum.2018.227.00005

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 25 Aug 2018; Published Online: 27 Sep 2019.

* Correspondence: Dr. Thorsten O Zander, Technische Universität Berlin, Biopsychology and Neuroergonomics, Berlin, Deutschland, 10587, Germany, thorsten.zander@b-tu.de