Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jul 2:11:1301.
doi: 10.3389/fpsyg.2020.01301. eCollection 2020.

On the Assessment of Functional Connectivity in an Immersive Brain-Computer Interface During Motor Imagery

Myriam Alanis-Espinosa  1 , David Gutiérrez  1
Affiliations

On the Assessment of Functional Connectivity in an Immersive Brain-Computer Interface During Motor Imagery

Myriam Alanis-Espinosa et al. Front Psychol. .

Abstract

New trends on brain-computer interface (BCI) design are aiming to combine this technology with immersive virtual reality in order to provide a sense of realism to its users. In this study, we propose an experimental BCI to control an immersive telepresence system using motor imagery (MI). The system is immersive in the sense that the users can control the movement of a NAO humanoid robot in a first person perspective (1PP), i.e., as if the movement of the robot was his/her own. We analyze functional brain connectivity between 1PP and 3PP during the control of our BCI using graph theory properties such as degree, betweenness centrality, and efficiency. Changes in these metrics are obtained for the case of the 1PP, as well as for the traditional third person perspective (3PP) in which the user can see the movement of the robot as feedback. As proof-of-concept, electroencephalography (EEG) signals were recorded from two subjects while they performed MI to control the movement of the robot. The graph theoretical analysis was applied to the binary directed networks obtained through the partial directed coherence (PDC). In our preliminary assessment we found that the efficiency in the α brain rhythm is greater in 1PP condition in comparison to the 3PP at the prefrontal cortex. Also, a stronger influence of signals measured at EEG channel C3 (primary motor cortex) to other regions was found in 1PP condition. Furthermore, our preliminary results seem to indicate that α and β brain rhythms have a high indegree at prefrontal cortex in 1PP condition, and this could be possibly related to the experience of sense of agency. Therefore, using the PDC combined with graph theory while controlling a telepresence robot in an immersive system may contribute to understand the organization and behavior of brain networks in these environments.

Keywords: brain-computer interface; functional brain connectivity; graph theory; partial directed coherence; sense of immersion.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Training sequence as shown to the subject in the computer monitor.
Figure 2
Figure 2
Power spectra and r2 values of the channels used to obtain the features to train the classifier for the BCI control task in one subject.
Figure 3
Figure 3
Control sequence as shown in the computer monitor and HMD.
Figure 4
Figure 4
Different arrangements for our experiments. (A) Third-person perspective, (B) First-person perspective, and (C) Robot in remote location.
Figure 5
Figure 5
General framework for the implementation of the immersive telepresence system.
Figure 6
Figure 6
Software architecture.
Figure 7
Figure 7
Stereoscopic image and BCI stimulus. The blue arrow is the stimulus to perform the motor imagery movement.
Figure 8
Figure 8
p-value as a function of τ in α band. Dash-dotted line indicates p = 0.05.
Figure 9
Figure 9
Significant connectivity (left) and its corresponding degree distributions (right) for the 3PP (A,C) and 1PP (B,D) conditions in α band for P1 and P2, respectively. Indegree distribution is represented by dark-color bars and the outdegree is represented by light-color bars.
Figure 10
Figure 10
Comparison of the indegree (vertical axes) in β for both conditions and participants. The indegree values for P1 and P2 are indicated with ○ and □, respectively.
Figure 11
Figure 11
Mean indegree for both participants shown in color only at the sensors with an incremental trend in comparison to the other condition. For each frequency band, the heads to the left (A,C,E,G) show the cases when the indegree in 1PP increased in comparison to 3PP, while the heads to the right (B,D,F,H) show those for which 3PP increased in comparison to 1PP.
Figure 12
Figure 12
Mean outdegree for both participants shown in color only at the sensors with an incremental trend in comparison to the other condition. For each frequency band, the heads to the left (A,C,E,G) show the cases when the outdegree in 1PP increased in comparison to 3PP, while the heads to the right (B,D,F,H) show those for which 3PP increased in comparison to 1PP.
Figure 13
Figure 13
Mean node betweenness centrality for both participants shown in color only at the sensors with an incremental trend in comparison to the other condition. For each frequency band, the heads to the left (A,C,E) show the cases when the node betweenness centrality in 1PP increased in comparison to 3PP, while the heads to the right (B,D,F) show those for which 3PP increased in comparison to 1PP. No increase in node betweenness was found in α.
Figure 14
Figure 14
Mean efficiency for both participants shown in color only at the sensors with an incremental trend in comparison to the other condition. The first two heads (A,B) show the cases when the efficiency in 1PP increased in comparison to 3PP, while the third head (C) shows those for which 3PP increased in comparison to 1PP. No increase in node efficiency was found in α for 3PP, nor in θ and γ in any modality.
See this image and copyright information in PMC

References

    1. Alanís-Espinosa M., Gutiérrez D. (2018). “Using the partial directed coherence to understand brain functional connectivity during movement imagery tasks,” in Proceedings of the 11th International Conference on Brain Informatics (Arlington, TX: ), 119–128.
    1. Athanasiou A., Klados M. A., Styliadis C., Foroglou N., Polyzoidis K., Bamidis P. D. (2018). Investigating the role of alpha and beta rhythms in functional motor networks. Neuroscience 378, 54–70. 10.1016/j.neuroscience.2016.05.044 - DOI - PubMed
    1. Athanasiou A., Lithari C., Kalogianni K., Klados M. A., Bamidis P. D. (2012). Source detection and functional connectivity of the sensorimotor cortex during actual and imaginary limb movement: a preliminary study on the implementation of econnectome in motor imagery protocols. Adv. Hum. Comput. Interact. 2012, 1–10. 10.1155/2012/127627 - DOI
    1. Baccala L., Sameshima K. (2001). Partial directed coherence: a new concept in neural structure determination. Biol. Cybern. 84, 463–474. 10.1007/PL00007990 - DOI - PubMed
    1. Baka E., Stavroulia K. E., Magnenat-Thalmann N., Lanitis A. (2015). “An EEG-based evaluation for comparing the sense of presence between virtual and physical environments,” in CGI 2018, ICPS, eds N. Magnenat-Thalmann, J. Kim, H. E. Rushmeier, B. Levy, R. Zhang, and D. Thalmann (New York, NY: Association for Computing Machinery; ), 107–116.