INFORMATION ABOUT PROJECT,
SUPPORTED BY RUSSIAN SCIENCE FOUNDATION

The information is prepared on the basis of data from the information-analytical system RSF, informative part is represented in the author's edition. All rights belong to the authors, the use or reprinting of materials is permitted only with the prior consent of the authors.

COMMON PART

Project Number24-75-00105

Project titleQuasi-movements in brain-computer interfaces: effects of movement type and feedback

Project LeadSvirin Evgeniy

AffiliationMoscow State University of Psychology and Education,

Research area 05 - FUNDAMENTAL RESEARCH IN MEDICINE, 05-106 - Neurobiology

Keywordsbrain-computer interface, EEG, electroencephalogram, human-machine interfaces, neurointerfaces, sensomotor, quasi-movements, motor imagery, intention, kinesthetic imagery, goal-directed movements, neurorehabilitation, attempted movements

PROJECT CONTENT

Annotation
Non-invasive brain-computer interfaces (BCIs) are increasingly used in neurorehabilitation. Based on them, assistive technologies are being developed for paralyzed patients and even for healthy people. One of the most common types of BCIs is a BCI based on motor imagery: their users, to give a command, imagine how they make a certain movement, without making it in reality. However, because motor imagery focuses the user's attention on the mental task at hand, cognitive resources in such BCIs are diverted from the perception of external stimuli, which makes it difficult to monitor feedback when the BCI is triggered. In addition, motor imagery tasks often poorly correspond to the movements that would be necessary to perform in reality to perform the desired action. The combination of these factors leads to a decrease in the user's sense of control over the operation of the interface and a sense of agency, which leads to a decrease in the effectiveness of using the interface. With this in mind, it is preferable to use tasks that do not require an active internal focus of attention to issue commands to the BCI. In addition, it is important to ensure that the actions the user performs and the feedback received are as close to “natural” as possible for the purpose for which the user is issuing the command. In high-performance invasive BCIs, this is increasingly achieved by using, instead of imagining, attempted movements that do not result in actual movement due to paralysis but are accompanied by clear patterns of brain activity. However, due to the difficulty of simulating such movement attempts without actual movements on healthy subjects, this technology is not widely used, including in neurorehabilitation using non-invasive BCIs. The project will evaluate the possibility of significantly improving the usability and naturalness of user interaction with a BCI by replacing motor imagination with the execution of specially selected quasi-movements. Quasi-movements are a relatively recently discovered and very little researched phenomenon observed when subjects are asked to decrease the amplitude of a movement until both the movement and the electromyographic (EMG) signal from the corresponding muscles disappear, with the EEG largely retaining the pattern of changes characteristic of real movements. Quasi-movements performed by healthy subjects can be considered as a model of attempts to perform movements by paralyzed people (Nikulin et al., 2008, Vasilyev et al., 2023). The possibility of using quasi-movements in BCIs has already been studied and these studies yielded positive results. It has also been shown that, unlike imaginary movements, quasi-movements are more often experienced by the user as an attempt to perform a real movement (Yashin et al., 2023). However, so far only a single movement has been used, namely, thumb abduction, which is not natural to the user and is not aimed at the target. In the present project, we will for the first time analyze the possibility of using ergonomic goal-directed quasi-movements in a BCI to give a command, as well as feedback that ensures a congruent user experience. In experiments on healthy subjects, such quasi-movements will be selected and the influence of the type of movement and its purposefulness on the severity of changes in the EEG will be assessed. In offline modelling, the accuracy and speed characteristics of the BCI on the selected quasi-movements will be assessed in comparison with the BCI on imaginary movements. Based on the simulation data, quasi-movements will be selected for subsequent use in an online BCI. In an experimental session with an online BCI (experiments with the real-time activation of a BCI), the influence of feedback on changes in EEG and accuracy-speed characteristics will be assessed, and, as in offline modelling, a comparison will be made between quasi-movements and similar imaginary movements as ways of giving a command.

Expected results
1. Will be developed the methodology of experimental EEG study of new types of quasi-movements, different from the only previously used one, namely, thumb abduction (used in Nikulin et al., 2008), including purposeful ones. These movements will later be used as a basis for developing a more ergonomic and "natural" brain-computer interface (BCI), compared to analogues that use motor imagery. 2. For the first time EEG patterns associated with the execution of such quasi-movements will be described, including the features of patterns accompanying purposeful quasi-movements (e.g., "pressing a button with the index finger", "pressing a pedal with the foot", as well as movements important for neurorehabilitation, such as grasping movement, flexion-extension of the hand) These patterns will be compared to those associated with real and imaginary movements. The expression and nature of changes in EEG patterns will allow us to justify the choice of quasi-movements for BCI, as well as to evaluate the possibility of using the technique as a model for studies of attempted movements in neurorehabilitation and fundamental studies of the motor system of the human brain. 3. For the first time accuracy and velocity estimates of BCI based on new types of quasi-movements will be obtained (in offline modelling) and compared with the characteristics of BCI based on the corresponding imaginary movements. 4. Based on the results of 2, and 3, it is supposed to investigate the influence of feedback in online BCI on EEG patterns and accuracy-velocity characteristics of IMC on the most "effective" variants of quasi-movements. These results would suggest perspectives for the development of BCI technology on the new types of quasi-movements, including those clinically significant for neurorehabilitation, which will make it possible to move from quasi-movements as used only in research to modelling of attempted movements in paralyzed patients on healthy subjects. This technology has the potential to significantly impact strategies for using BCI in neurorehabilitation, as it will allow the rapid development of techniques based on patients' attempted movements instead of motor imagination.

REPORTS