— EEG has been studied in the nonstationary mode of cognitive tasks. In this mode, cognitive task performance is reduced to seconds, cognitive activity is repeated many times, the stages of short-term dominance of externally oriented (sensory) attention are followed by stages of longer dominance of either sensory or internally oriented (mental) attention. Previously, the stationary mode was used, in which the subjects were for a long time either in the state of dominance of sensory attention or in the state of dominance of mental attention. The non-stationary mode is close to real learning modes, including the use of electronic education techniques. The non-stationary mode also allows the comparison of EEG correlates of mental and sensory attention dominance at different stages of cognitive task performance. An analysis of preliminary data shows that, at the stage of preparatory attention, the EEG differences between tasks with a dominance of mental attention and with a dominance of sensory attention are much smaller than the differences at the stage of executive attention, which suggests non-specificity of the nature of preparatory attention.

BACKGROUND: Significant widespread differences in hemodynamics and electrodynamics between states of rest with open and closed eyes (default mode brain states) probably are consequences of involuntary shifts between interoceptive and exteroceptive modes of attention. These results make grounds for searches of similar correlates in active mental states. Results of such an attempt applied to states of verbal memorization and recollection using quantitative state-related EEG are presented. METHODS: Electroencephalograms were recorded in 88 subjects in a resting state with the eyes opened (state EO), during memorization (state M) of verbal bilingual semantic pairs (Latin and Russian), presented on a screen, and during recollection (state R) of the learned information Statistical comparison of the EEG spectral power (local synchronization) and coherence (spatial synchronization) in the frequency bands theta, alpha1, apha2, beta1, beta2 and gamma showed that induction of the states M and R led to multiple significant changes in the EEG absolute power and coherence as compared to the state EO and as compared between active states R and M. RESULTS: The results demonstrate that mnestic states of memorization and recollection are catered with rather different brain functional states reflected in system reorganizations of brain electrodynamics on the levels of local and spatial synchronizations both. CONCLUSION: The observed differences can be related to changes of the exteroceptive– introceptive attention balance.

Group 1 (N = 30) and Group 2 (N = 22) of healthy volunteers participated in the experiment. In both groups, the EEG was recorded in the state of rest with the eyes closed (REC); at rest with eyes open (REO); and during passively watching TV channel noises (white noise). EEG was also recorded in the state of forced visual attention: when counting colored symbols appearing on the screen (group 1) and when searching for an image of a real object in the noise (group 2). The averaged values of EEG power in each derivation were calculated for every subject and for each state in the Δ, θ, α1, α2, β1, β2, and γ frequency bands. The results demonstrate that the exposure to unstructured noninformative video noise may lead to significant changes in the EEG power in various frequency bands, with the most prominent changes taking place in the α2 band. These changes are topically wide, reflecting systemic changes in the corresponding brain mechanisms, but are much less intense compared to the difference between the states of rest with the eyes open and closed.

Electroencephalograms (EEGs) were recorded in 19 standard derivations in 88 subjects (students) in states of: rest with the eyes open; memorization (learning) of verbal bilingual semantic pairs (Latin and Russian); and recollection (control) of the memorized information. Estimates of EEG coherence in these states were compared using statistical methods for the frequency bands θ, α1, α2, β1, β2, and γ. The results of this comparison showed that transition from the state of rest to those of memorization and recollection was accompanied by numerous changes in coherence in all the frequency bands. These changes varied in intensity and embrace practically the entire convexital cerebral cortex. Decrease in EEG coherence was predominant during the transition to the memorization state in various frequency bands, whereas during the transition from the rest state to that of recollection, EEG coherence increased in most of the frequency bands except for the band α2. The reproducibility of this pattern of changes in EEG coherence is confirmed by the results in the subgroups formed by randomly subdividing the subjects into two groups. We think that the observed intense rearrangement of the spatial synchronization of the cortex electrical activity reflects the reorganization of the functional systems of neuronal ensembles to provide efficient memorization and recollection, respectively.

The tip-of-the-tongue state, or memory blocking, is considered with regard to the feasibility of its neural network modeling. The results of psycholinguistic and neurobiological studies on memory blocking are reviewed, and basic problems that need be solved to comprehend this phenomenon are formulated. One such point is the dramatic discrepancy between the subjective assurance that an image is familiar and the inability to recollect it fully. To explain this discrepancy, we propose a biologically plausible neural network model of recognition, demonstrating cardinal superiority in the capacity of image recognition over its recollection.

The capacities of a specially designed neural network for familiarity recognition and recollection have been compared. Recognition is based on calculating “image familiarity” as a modified Hopfield energy function in which the value of the inner sum is replaced by the sign of this value. This replacement makes the calculation of familiarity compatible with the basic dynamic equations of the Hopfield network and is in fact reduced to calculating the scalar product of the neuronet state vectors at two successive time steps.

We consider a neural network model of age-related cognitive changes in aging brain based on Hopfield network with a sigmoid function of neuron activation. Age is included in the activation function as a parameter in the form of exponential rate denominator, which makes it possible to take into account the weakening of interneuronal links really observed in the aging brain. Analysis of properties of the Lyapunov function associated with the network shows that, with increasing parameter of age, its relief becomes smoother and the number of local minima (network attractors) decreases. As a result, the network gets less frequently stuck in the nearest local minima of the Lyapunov function and reaches a global minimum corresponding to the most effective solution of the cognitive task. It is reasonable to assume that similar changes really occur in the aging brain. Phenomenologically, these changes can be manifested as emergence in aged people of a cognitive quality such as wisdom i.e. ability to find optimal decisions in difficult controversial situations, to distract from secondary aspects and to see the problem as a whole.

Cognitive and regulatory processes in organism are ensured by the functioning of several different network systems - neural, endocrine, immune, and gene ones. These systems are, however, closely related and form a single integrated neurogenohumoral cognitive-regulatory dynamic system of organism. A review of publications is given which shows that it is possible to associate with this dynamic system a corresponding Lyapunov function (energy function, potential function) and that analyzing this function allows, due to its geometrical insight, to easily discover a set of general properties of cognitive and regulatory functioning of organism.

Elaborated data-mining techniques are widely available today. Nevertheless, many non-linear relations among variables remain undiscovered in multi-dimensional datasets. To address this issue we propose a method based on the concept of fractal dimension that explores the structure of multivariate data and apply the method to simulated data, as well as to local field potentials recorded from cat visual cortex. We find that with changes in the analysis scale, the dimensionality of the data often changes, indicating first that the data are not simple fractals with one unique dimension and second, that, at a certain scale, important changes in the geometric structure of the data may occur. The method can be used as a data-mining tool but also as a method for testing a model's fit to the data. We achieve the latter by comparing the dimensionality of the original data to the dimensionality of the data reconstructed from a model's description of the data (here using the general linear model). The method provides indispensable help in estimating the complexity of non-linear relationships within multivariate datasets.

Spatial EEG synchronization was studied using mean EEG coherences in 57 subjects in a resting state with the eyes open, during memorization of verbal bilingual semantic pairs (Latin and Russian), and during retrieval (monitoring) of the learned information. Statistical comparison of the EEG spectral power in the frequency bands θ, α 1, α2, β1, β2, and γ showed that induction of the states of memorization and retrieval of the verbal information resulted in multiple significant increases in the mean coherence (spatial synchronization) as compared to the state of rest. These increases were significantly higher in the state of retrieval than in the state of memorization. Such changes simultaneously occurred in different frequency bands and over the entire cortex. The highest relative augmentations of coherence with the highest significance estimates were observed in long-distance derivation pairs involving the temporal areas or combining the anterofrontal and frontal derivations with the parietal and occipital ones. The results are discussed in combination with data obtained in the same study for changes in EEG local synchronization and reported earlier. The intense rearrangements of the bioelectrical activity of the cortex may be determined both by the effect of cognitively specific mechanisms of encoding-decoding of information in the memory system and by a cognitively nonspecific modulating system of the brain.

The article is devoted to commemoration of full member of Russian Academy of Medical Sciences, Oleg Andreevich Adrianov, who would have celebrated his 75-th anniversary in 1998. O. S. Adrianov, author of numerous works on physiology and morphology of central nervous system, in the recent years of his was studying the problem of the processes relationship at macro and micro levels of brain organization. Further to the concept created by O.S. Adrianov, data on action of two peptides: delta-sleep and tafcine, on behavior, neurophysiological and neurochemical processes have been consolidated. Experimental data were obtained for rabbits, cats, and dogs, both intact and in the state of pathology (psychomotoric excitement, bradykinesia, penicillin epilepsy). Impact of peptides on convergation processes is discussed: peptide of delta-sleep depresses reactions of brain structures to photo- and phono-stimulation, and activates the serotoninergic system in general; tafcine enforces the convergation processes and activates the dopaminergic system.

Chronic injections of haloperidol and sulpiride to dogs synchronized processes in the basal ganglia and visual analyzer, which may be explained by therapeutic effect of the neuroleptics. Haloperidol destroyed interaction between basal ganglia and motor cortex. Under the effect of sulpiride, the sensory tuning was observed not only in the basal ganglia, but also in the hippocampus.