The brain is likely the most convoluted and enigmatic object for comprehensive studies attracting the burning interest of the broad scientific community. The understanding of brain functionality requires a multidisciplinary approach involving different areas of science, engineering and technology. Traditional brain – computer interfaces (BCIs) imply the interaction between brains and machines with the aim to repair or increase human performance in solving different tasks or to help paralyzed people to interact with an environment. Unlike a BCI, a brain – to – brain interface (BBI) allows a direct information exchange between brains. The BBI development is one of the most progressing research directions at the intersection of physics, mathematics, informatics, psychology and neuroscience. The main trend in the BBI research is aimed at providing people with a new way for communication directly from one brain to another, to monitor and control mental states and increase working performance by using cognitive recourses of multiple brains. This particularly important for a group of people working together on a common task which requires sustained attention and alertness.


In addition, understanding the neurophysiological mechanisms responsible for motor imagery (MI) is essential for the development of brain-computer interfaces (BCI) and bioprostheses. MEG experiments with volunteers confirm the existence of two types of motor imagery, kinesthetic (KI) and visual (VI), which can be distinguished by activation or inhibition of different brain areas in α and β frequency bands.


The first successful attempt to implement our solution for Brain to Brain Interfaces has been made in our laboratory. We developed novel methods of nonlinear and stochastic analyses of neurophysiological data, as well as artificial intelligence for pattern recognition and classification in electro – and magneto – encephalograms.


Although KI brain activity is usually observed in specially trained subjects or athletes, we show that it is also possible to identify characteristics of MI in untrained subjects. Similar to the actual movement, KI implies muscular sensation when performing MI that leads to event related desynchronization (ERD) of cerebral rhythms associated with MI. On the contrary, VI refers to the visualization of the corresponding action that results in even related synchronization (ERS) of the brain activity at α and β waves. A notable difference between the KI and VI groups occurs in the frontal area of the brain. In particular, the analysis of the evoked responses associated with MI shows that in all KI subjects the frontal cortex activity is suppressed during MI, while in VI subjects the frontal cortex is always active. The accuracy in the classification of the left arm and the MI of the right arm using artificial intelligence is similar for KI and VI. Since untrained subjects generally demonstrate the VI mode, the possibility of increasing accuracy for VI is in demand for BCI. The application of artificial neural networks allows us to classify the MI by raising the right and left arms with an average accuracy of 70% for both KI and VI using adequate filtration of the input signals. The same average accuracy is achieved by optimizing the MEG channels and reducing their number to only 13.



Development of novel methods for experimental study and control of stochastic processes in the human brain during visual perception (2016-2019)
Ministerio de Economía y Competitividad

PI: Alexander Pisarchik


Characterization and prediction of extreme events in neurophysiological brain activity (2018). UPM PI: Alexander Pisarchik


Predictability of catastrophic transitions in climate (2017). UPM PI: Alexander Pisarchik