Critical slowing down and noise-induced intermittency in bistable perception: bifurcation analysis

Hysteresis reveals a happiness bias effect in dynamic emotion recognition from ambiguous biological motion

Considering the nonlinear dynamic nature of emotion recognition, it is believed to be strongly dependent on temporal context. This can be investigated by resorting to the phenomenon of hysteresis, which features a form of serial dependence, entailed by continuous temporal stimulus trajectories. Under positive hysteresis, the percept remains stable in visual memory (persistence) while in negative hysteresis, it shifts earlier (adaptation) to the opposite interpretation. Here, we asked whether positive or negative hysteresis occurs in emotion recognition of inherently ambiguous biological motion, while testing for the controversial debate of a negative versus positive emotional bias. Participants (n = 22) performed a psychophysical experiment in which they were asked to judge stimulus transitions between two emotions, happiness and sadness, from an actor database, and report perceived emotion across time, from one emotion to the opposite as physical cues were continuously changing. Our results reveal perceptual hysteresis in ambiguous emotion recognition, with positive hysteresis (visual persistence) predominating. However, negative hysteresis (adaptation/fatigue) was also observed in particular in the direction from sadness to happiness. This demonstrates a positive (happiness) bias in emotion recognition in ambiguous biological motion recognition. Finally, the interplay between positive and negative hysteresis suggests an underlying competition between visual persistence and adaptation mechanisms during ambiguous emotion recognition.

Positive hysteresis in emotion recognition: Face processing visual regions are involved in perceptual persistence, which mediates interactions between anterior insula and medial prefrontal cortex

Facial emotion perception can be studied from the point of view of dynamic systems whose output may depend not only on current input but also on prior history - a phenomenon known as hysteresis. In cognitive neuroscience, hysteresis has been described as positive (perceptual persistence) or negative (fatigue of current percept) depending on whether perceptual switching occurs later or earlier than actual physical stimulus changes. However, its neural correlates remain elusive. We used dynamic transitions between emotional expressions and combined behavioral assessment with functional magnetic resonance imaging (fMRI) to investigate the underlying circuitry of perceptual hysteresis in facial emotion recognition. Our findings revealed the involvement of face-selective visual areas - fusiform face area (FFA) and superior temporal sulcus (STS) - in perceptual persistence as well as the right anterior insula. Moreover, functional connectivity analyses revealed an interplay between the right anterior insula and medial prefrontal cortex, which showed to be dependent on the presence of positive hysteresis. Our results support the hypothesis that high-order regions are involved in perceptual stabilization and decision during perceptual persistence (positive hysteresis) and add evidence to the role of the anterior insula as a hub of sensory information in perceptual decision-making.

Coherence resonance in neuronal populations: Mean-field versus network model

The counterintuitive phenomenon of coherence resonance describes a nonmonotonic behavior of the regularity of noise-induced oscillations in the excitable regime, leading to an optimal response in terms of regularity of the excited oscillations for an intermediate noise intensity. We study this phenomenon in populations of FitzHugh-Nagumo (FHN) neurons with different coupling architectures. For networks of FHN systems in an excitable regime, coherence resonance has been previously analyzed numerically. Here we focus on an analytical approach studying the mean-field limits of the globally and locally coupled populations. The mean-field limit refers to an averaged behavior of a complex network as the number of elements goes to infinity. We apply the mean-field approach to the globally coupled FHN network. Further, we derive a mean-field limit approximating the locally coupled FHN network with low noise intensities. We study the effects of the coupling strength and noise intensity on coherence resonance for both the network and the mean-field models. We compare the results of the mean-field and network frameworks and find good agreement in the globally coupled case, where the correspondence between the two approaches is sufficiently good to capture the emergence of coherence resonance, as well as of anticoherence resonance.

Dynamics of a Mutual Inhibition Circuit between Pyramidal Neurons Compared to Human Perceptual Competition

Neural competition plays an essential role in active selection processes of noisy and ambiguous input signals, and it is assumed to underlie emergent properties of brain functioning, such as perceptual organization and decision-making. Despite ample theoretical research on neural competition, experimental tools to allow neurophysiological investigation of competing neurons have not been available. We developed a "hybrid" system where real-life neurons and a computer-simulated neural circuit interacted. It enabled us to construct a mutual inhibition circuit between two real-life pyramidal neurons. We then asked what dynamics this minimal unit of neural competition exhibits and compared them with the known behavioral-level dynamics of neural competition. We found that the pair of neurons shows bistability when activated simultaneously by current injections. The addition of modeled synaptic noise and changes in the activation strength showed that the dynamics of the circuit are strikingly similar to the known properties of bistable visual perception: The distribution of dominance durations showed a right-skewed shape, and the changes of the activation strengths caused changes in dominance, dominance durations, and reversal rates as stated in the well-known empirical laws of bistable perception known as Levelt's propositions.SIGNIFICANCE STATEMENT Visual perception emerges as the result of neural systems actively organizing visual signals that involves selection processes of competing neurons. While the neural competition, realized by a "mutual inhibition" circuit has been examined in many theoretical studies, its properties have not been investigated in real neurons. We have developed a "hybrid" system where two real-life pyramidal neurons in a mouse brain slice interact through a computer-simulated mutual inhibition circuit. We found that simultaneous activation of the neurons leads to bistable activity. We investigated the effect of noise and the effect of changes in the activation strength on the dynamics. We observed that the pair of neurons exhibit dynamics strikingly similar to the known properties of bistable visual perception.

Effect of rate of change of parameter on early warning signals for critical transitions

Many dynamical systems exhibit abrupt transitions or tipping as the control parameter is varied. In scenarios where the parameter is varied continuously, the rate of change of the control parameter greatly affects the performance of early warning signals (EWS) for such critical transitions. We study the impact of variation of the control parameter with a finite rate on the performance of EWS for critical transitions in a thermoacoustic system (a horizontal Rijke tube) exhibiting subcritical Hopf bifurcation. There is a growing interest in developing early warning signals for tipping in real systems. First, we explore the efficacy of early warning signals based on critical slowing down and fractal characteristics. From this study, lag-1 autocorrelation (AC) and Hurst exponent (H) are found to be good measures to predict the transition well before the tipping point. The warning time, obtained using AC and H, reduces with an increase in the rate of change of the control parameter following an inverse power law relation. Hence, for very fast rates, the warning time may be too short to perform any control action. Furthermore, we report the observation of a hyperexponential scaling relation between the AC and the variance of fluctuations during such a dynamic Hopf bifurcation. We construct a theoretical model for noisy Hopf bifurcation wherein the control parameter is continuously varied at different rates to study the effect of rate of change of the parameter on EWS. Similar results, including the hyperexponential scaling, are observed in the model as well.

Voluntary and Involuntary Attention in Bistable Visual Perception: A MEG Study

In this study, voluntary and involuntary visual attention focused on different interpretations of a bistable image, were investigated using magnetoencephalography (MEG). A Necker cube with sinusoidally modulated pixels' intensity in the front and rear faces with frequencies 6.67 Hz (60/9) and 8.57 Hz (60/7), respectively, was presented to 12 healthy volunteers, who interpreted the cube as either left- or right-oriented. The tags of these frequencies and their second harmonics were identified in the average Fourier spectra of the MEG data recorded from the visual cortex. In the first part of the experiment, the subjects were asked to voluntarily control their attention by interpreting the cube orientation as either being on the left or right. Accordingly, we observed the dominance of the corresponding spectral component, and voluntary attention performance was measured. In the second part of the experiment, the subjects were asked to focus their gaze on a red marker at the center of the cube image without putting forth effort in its interpretation. The alternation of the dominant spectral energies at the second harmonics of the stimulation frequencies was treated as changes in the cube orientation. Based on the results of the first experimental stage and using a wavelet analysis, we developed a method which allowed us to identify the currently perceived cube orientation. Finally, we characterized involuntary attention using the distribution of dominance times when focusing attention on one of the cube orientations, which was related to voluntary attention performance and brain noise. In particular, we confirmed our hypothesis that higher attention performance is associated with stronger brain noise.

Identification of competing neural mechanisms underlying positive and negative perceptual hysteresis in the human visual system

Hysteresis is a well-known phenomenon in physics that relates changes in a system with its prior history. It is also part of human visual experience (perceptual hysteresis), and two different neural mechanisms might explain it: persistence (a cause of positive hysteresis), which forces to keep a current percept for longer, and adaptation (a cause of negative hysteresis), which in turn favors the switch to a competing percept early on. In this study, we explore the neural correlates underlying these mechanisms and the hypothesis of their competitive balance, by combining behavioral assessment with fMRI. We used machine learning on the behavioral data to distinguish between positive and negative hysteresis, and discovered a neural correlate of persistence at a core region of the ventral attention network, the anterior insula. Our results add to the understanding of perceptual multistability and reveal a possible mechanistic explanation for the regulation of different forms of perceptual hysteresis.

Optimal Self-Induced Stochastic Resonance in Multiplex Neural Networks: Electrical vs. Chemical Synapses

Electrical and chemical synapses shape the dynamics of neural networks, and their functional roles in information processing have been a longstanding question in neurobiology. In this paper, we investigate the role of synapses on the optimization of the phenomenon of self-induced stochastic resonance in a delayed multiplex neural network by using analytical and numerical methods. We consider a two-layer multiplex network in which, at the intra-layer level, neurons are coupled either by electrical synapses or by inhibitory chemical synapses. For each isolated layer, computations indicate that weaker electrical and chemical synaptic couplings are better optimizers of self-induced stochastic resonance. In addition, regardless of the synaptic strengths, shorter electrical synaptic delays are found to be better optimizers of the phenomenon than shorter chemical synaptic delays, while longer chemical synaptic delays are better optimizers than longer electrical synaptic delays; in both cases, the poorer optimizers are, in fact, worst. It is found that electrical, inhibitory, or excitatory chemical multiplexing of the two layers having only electrical synapses at the intra-layer levels can each optimize the phenomenon. Additionally, only excitatory chemical multiplexing of the two layers having only inhibitory chemical synapses at the intra-layer levels can optimize the phenomenon. These results may guide experiments aimed at establishing or confirming to the mechanism of self-induced stochastic resonance in networks of artificial neural circuits as well as in real biological neural networks.

How positive emotional content overrules perceptual history effects: Hysteresis in emotion recognition

The human visual system is constantly processing multiple and often conflicting sensory cues to make perceptual decisions. Given the nonlinear nature of emotion recognition, this often leads to different percepts of the same physical facial expression. Moreover, the state of the emotion recognition system might depend on the trajectory of temporal context, potentially leading to a phenomenon known as perceptual hysteresis. Here, we aimed to explore temporal context-related mechanisms underlying perceptual hysteresis during emotion recognition. We hypothesized that dependence on recent perceptual experience might reveal important clues about the role of short-term memory on the perception of emotional stimuli. Behavioral data were acquired using reality-based, changing emotion expressions morphed from a source to a target emotion with different valences, always passing through a neutral expression. Participants identified the onset and offset of what they perceived as the neutral expression interval. Our results showed that current perception of emotional expression is affected by recent temporal context, thus revealing perceptual hysteresis. We also found a relation between recent perceptual history effects and stimulus emotional Content: The positive valence of the stimulus emotional content appeared to abolish perceptual history effects, whereas negatively loaded stimuli induced clear short-term memory effects and positive hysteresis. Our findings show direct competition between recent perceptual experience and stimulus emotional content during decision making, which affects the formation of current percepts in emotion recognition.

A neural network model for exogenous perceptual alternations of the Necker cube

When a bistable visual image, such as the Necker cube, is continuously viewed, the percept of the image endogenously alternates between one possible percept and the other. However, perceptual alternation can also be induced by an exogenous perturbation. For example, a typical external perturbation is the flashlight, which is expected to pervasively activate many brain regions. Therefore, the neural mechanism related to exogenous perceptual alternation remains to be clarified. As a cue to solving this problem, our recent psychophysiological experiment reported a positive correlation between the enhancement of visual mismatch negativity evoked by breaks in the sequential regularity of the visual stimuli and the proportion of perceptual alternation. To elucidate the mechanism underlying exogenous perceptual alternation induced by visual mismatch negativity, the present study attempted to construct a neural network model for bistable perception of the Necker cube, whose perceptual alternation is facilitated by an increase in visual mismatch negativity. The model consists of both a prediction layer and a prediction error layer, following the predictive coding framework for biologically plausible relationships between the change detection process and the perceptual alternation mechanism. Computer simulations showed that the mean duration of perception decreased as the response increased, which is in concordance with the experimental data. This result suggested that the excitatory feedforward and inhibitory feedback connections play an important role. Additionally, the validity of this model suggests that the visual mismatch signal propagates in the neural systems and affects the visual perceptual mechanism as a prediction error signal.

Bistable perception of ambiguous images: simple Arrhenius model

Watching an ambiguous image leads to the bistability of its perception, that randomly oscillates between two possible interpretations. The relevant evolution of the neuron system is usually described with the equation of its "movement" over the nonuniform energy landscape under the action of the stochastic force, corresponding to noise perturbations. We utilize the alternative (and simpler) approach suggesting that the system is in the quasi-stationary state being described by the Arrhenius equation. The latter, in fact, determines the probability of the dynamical variation of the image being percepted (for example, the left Necker cube ↔ the right Necker cube) along one scenario or another. Probabilities of transitions from one perception to another are defined by barriers detaching corresponding wells of the energy landscape, and the relative value of the noise (analog of temperature) influencing this process. The mean noise value could be estimated from experimental data. The model predicts logarithmic dependence of the perception hysteresis width on the period of cyclic sweeping the parameter, controlling the perception (for instance, the contrast of the presented object). It agrees with the experiment and allows to estimate the time interval between two various perceptions.

Embodied gestalts: Unstable visual phenomena become stable when they are stimuli for competitive action selection

An animal's environment is rich with affordances. Different possible actions are specified by visual information while competing for dominance over neural dynamics. Affordance competition models account for this in terms of winner-takes-all cross-inhibition dynamics. Multistable phenomena also reveal how the visual system deals with ambiguity. Their key property is spontaneous instability, in forms such as alternating dominance in binocular rivalry. Theoretical models of self-inhibition or self-organized instability posit that the instability is tied to some kind of neural adaptation and that its functional significance is to enable flexible perceptual transitions. We hypothesized that the two perspectives are interlinked. Spontaneous instability is an intrinsic property of perceptual systems, but it is revealed when they are stripped from the constraints of possibilities for action. To test this, we compared a multistable gestalt phenomenon against its embodied version and estimated the neural adaptation and competition parameters of an affordance transition dynamic model. Wertheimer's (Zeitschrift fur Psychologie 61, 161-265, 1912) optimal (β) and pure (φ) forms of apparent motion from a stroboscopic point-light display were endowed with action relevance by embedding the display in a visual object-tracking task. Thus, each mode was complemented by its action, because each perceptual mode uniquely enabled different ways of tracking the target. Perceptual judgment of the traditional apparent motion exhibited spontaneous instabilities, in the form of earlier switching when the frame rate was changed stepwise. In contrast, the embodied version exhibited hysteresis, consistent with affordance transition studies. Consistent with our predictions, the parameter for competition between modes in the affordance transition model increased, and the parameter for self-inhibition vanished.

Percept-related EEG classification using machine learning approach and features of functional brain connectivity

Machine learning is a promising approach for electroencephalographic (EEG) trials classification. Its efficiency is largely determined by the feature extraction and selection techniques reducing the dimensionality of input data. Dimensionality reduction is usually implemented via the mathematical approaches (e.g., principal component analysis, linear discriminant analysis, etc.) regardless of the origin of analyzed data. We hypothesize that since EEG features are determined by certain neurophysiological processes, they should have distinctive characteristics in spatiotemporal domain. If so, it is possible to specify the set of EEG principal features based on the prior knowledge about underlying neurophysiological processes. To test this hypothesis, we consider the classification of EEG trials related to the perception of ambiguous visual stimuli. We observe that EEG features, underlying the different ambiguous stimuli interpretations, are defined by the network properties of neuronal activity. Having analyzed functional neural interactions, we specify the brain area in which neural network architecture exhibits differences for different classes of EEG trials. We optimize the feedforward multilayer perceptron and develop a strategy for the training set selection to maximize the classification accuracy, being 85% when all channels are used. The revealed localization of the percept-related features allows about 95% accuracy, when the number of channels is reduced up to 90%. Obtained results can be used for classification of EEG trials associated with more complex cognitive tasks. Taking into account that cognitive activity is subserved by a distributed functional cortical network, its topological properties have to be considered when selecting optimal features for EEG trial classification.

Control of coherence resonance by self-induced stochastic resonance in a multiplex neural network

We consider a two-layer multiplex network of diffusively coupled FitzHugh-Nagumo (FHN) neurons in the excitable regime. We show that the phenomenon of coherence resonance (CR) in one layer can not only be controlled by the network topology, the intra- and interlayer time-delayed couplings, but also by another phenomenon, namely, self-induced stochastic resonance (SISR) in the other layer. Numerical computations show that when the layers are isolated, each of these noise-induced phenomena is weakened (strengthened) by a sparser (denser) ring network topology, stronger (weaker) intralayer coupling forces, and longer (shorter) intralayer time delays. However, CR shows a much higher sensitivity than SISR to changes in these control parameters. It is also shown, in contrast to SISR in a single isolated FHN neuron, that the maximum noise amplitude at which SISR occurs in the network of coupled FHN neurons is controllable, especially in the regime of strong coupling forces and long time delays. In order to use SISR in the first layer of the multiplex network to control CR in the second layer, we first choose the control parameters of the second layer in isolation such that in one case CR is poor and in another case, nonexistent. It is then shown that a pronounced SISR can not only significantly improve a poor CR, but can also induce a pronounced CR, which was nonexistent in the isolated second layer. In contrast to strong intralayer coupling forces, strong interlayer coupling forces are found to enhance CR, while long interlayer time delays, just as long intralayer time delays, deteriorate CR. Most importantly, we find that in a strong interlayer coupling regime, SISR in the first layer performs better than CR in enhancing CR in the second layer. But in a weak interlayer coupling regime, CR in the first layer performs better than SISR in enhancing CR in the second layer. Our results could find novel applications in noisy neural network dynamics and engineering.

Chimera state in complex networks of bistable Hodgkin-Huxley neurons

In this paper we study a chimera state in complex networks of bistable Hodgkin-Huxley neurons with excitatory coupling, which manifests as a termination of spiking activity of a part of interacting neurons. We provide a detailed investigation of this phenomenon in scale-free, small-world, and random networks and show that the chimera state is robust to the network topology. Nevertheless, network topological properties determine the stability of spatiotemporal states and therefore affect the excitability of the chimera state in the whole network. In particular, the scale-free network whose higher degree nodes are more stable to small perturbations is least exposed to chimera formation and exhibits an abrupt transition from a spiking to a silent regime. On the other hand, small-world and random networks are more likely to provide transitions to the chimera state.

Interplay between random fluctuations and rate dependent phenomena at slow passage to limit-cycle oscillations in a bistable thermoacoustic system

We study the impact of noise on the rate dependent transitions in a noisy bistable oscillator using a thermoacoustic system as an example. As the parameter-the heater power-is increased in a quasi-steady manner, beyond a critical value, the thermoacoustic system undergoes a subcritical Hopf bifurcation and exhibits periodic oscillations. We observe that the transition to this oscillatory state is often delayed when the control parameter is varied as a function of time. However, the presence of inherent noise in the system introduces high variability in the characteristics of this critical transition. As a result, if the value of the system variable-the acoustic pressure-approaches the noise floor before the system crosses the unstable manifold, the effect of rate on the critical transition becomes irrelevant in determining the transition characteristics, and the system undergoes a noise-induced tipping to limit-cycle oscillations. The presence of noise-induced tipping makes it difficult to identify the stability regimes in such systems by using stability maps for the corresponding deterministic system.

Increasing Human Performance by Sharing Cognitive Load Using Brain-to-Brain Interface

Brain-computer interfaces (BCIs) attract a lot of attention because of their ability to improve the brain's efficiency in performing complex tasks using a computer. Furthermore, BCIs can increase human's performance not only due to human-machine interactions, but also thanks to an optimal distribution of cognitive load among all members of a group working on a common task, i.e., due to human-human interaction. The latter is of particular importance when sustained attention and alertness are required. In every day practice, this is a common occurrence, for example, among office workers, pilots of a military or a civil aircraft, power plant operators, etc. Their routinely work includes continuous monitoring of instrument readings and implies a heavy cognitive load due to processing large amounts of visual information. In this paper, we propose a brain-to-brain interface (BBI) which estimates brain states of every participant and distributes a cognitive load among all members of the group accomplishing together a common task. The BBI allows sharing the whole workload between all participants depending on their current cognitive performance estimated from their electrical brain activity. We show that the team efficiency can be increased due to redistribution of the work between participants so that the most difficult workload falls on the operator who exhibits maximum performance. Finally, we demonstrate that the human-to-human interaction is more efficient in the presence of a certain delay determined by brain rhythms. The obtained results are promising for the development of a new generation of communication systems based on neurophysiological brain activity of interacting people. Such BBIs will distribute a common task between all group members according to their individual physical conditions.

Multiscale neural connectivity during human sensory processing in the brain

Stimulus-related brain activity is considered using wavelet-based analysis of neural interactions between occipital and parietal brain areas in alpha (8-12 Hz) and beta (15-30 Hz) frequency bands. We show that human sensory processing related to the visual stimuli perception induces brain response resulted in different ways of parieto-occipital interactions in these bands. In the alpha frequency band the parieto-occipital neuronal network is characterized by homogeneous increase of the interaction between all interconnected areas both within occipital and parietal lobes and between them. In the beta frequency band the occipital lobe starts to play a leading role in the dynamics of the occipital-parietal network: The perception of visual stimuli excites the visual center in the occipital area and then, due to the increase of parieto-occipital interactions, such excitation is transferred to the parietal area, where the attentional center takes place. In the case when stimuli are characterized by a high degree of ambiguity, we find greater increase of the interaction between interconnected areas in the parietal lobe due to the increase of human attention. Based on revealed mechanisms, we describe the complex response of the parieto-occipital brain neuronal network during the perception and primary processing of the visual stimuli. The results can serve as an essential complement to the existing theory of neural aspects of visual stimuli processing.

Positive and negative hysteresis effects for the perception of geometric and emotional ambiguities

AIM

The present study utilizes perceptual hysteresis effects to compare the ambiguity of Mona Lisa's emotional face expression (high-level ambiguity) and of geometric cube stimuli (low-level ambiguity).

METHODS

In two experiments we presented series of nine Mona Lisa variants and nine cube variants. Stimulus ambiguity was manipulated by changing Mona Lisa's mouth curvature (Exp. 1) and the cubes' back-layer luminance (Exp. 2). Each experiment consisted of three conditions, two with opposite stimulus presentation sequences with increasing and decreasing degrees of ambiguity, respectively, and a third condition with a random presentation sequence. Participants indicated happy or sad face percepts (Exp. 1) and alternative 3D cube percepts (Exp. 2) by key presses. We studied the influences of a priori perceptual biases (long-term memory) and presentation order (short-term memory) on perception.

RESULTS

Perception followed sigmoidal functions of the stimulus ambiguity morphing parameters. The morphing parameter for the functions' inflection points depended strongly on stimulus presentation order with similar effect sizes but different signs for the two stimulus types (positive hysteresis / priming for the cubes; negative hysteresis / adaptation for Mona Lisa). In the random conditions, the inflection points were located in the middle between those from the two directional conditions for the Mona Lisa stimuli. For the cube stimuli, they were superimposed on one sigmoidal function for the ordered condition.

DISCUSSION

The hysteresis effects reflect the influence of short-term memory during the perceptual disambiguation of ambiguous sensory information. The effects for the two stimulus types are of similar size, explaining up to 34% of the perceptual variance introduced by the paradigm. We explain the qualitative difference between positive and negative hysteresis with adaptation for Mona Lisa and with priming for the cubes. In addition, the hysteresis paradigm allows a quantitative determination of the impact of adaptation and priming during the resolution of perceptual ambiguities. The asymmetric shifts of inflection points in the case of the cube stimuli is likely due to an a priori perceptual bias, reflecting an influence of long-term memory. Whether corresponding influences also exist for the Mona Lisa variants is so far unclear.

Weak multiplexing induces coherence resonance

Using the model of a FitzHugh-Nagumo system in the excitable regime, we study the impact of multiplexing on coherence resonance in a two-layer network. We show that multiplexing allows for the control of the noise-induced dynamics. In particular, we find that multiplexing induces coherence resonance in networks that do not demonstrate this phenomenon in isolation. Examples are provided by deterministic networks and networks where the strength of interaction between the elements is not optimal for coherence resonance. In both cases, we show that the control strategy based on multiplexing can be successfully applied even for weak coupling between the layers. Moreover, for the case of deterministic networks, we obtain a counter-intuitive result: the multiplex-induced coherence resonance in the layer which is deterministic in isolation manifests itself even more strongly than that in the noisy layer.

Artificial neural network detects human uncertainty

Artificial neural networks (ANNs) are known to be a powerful tool for data analysis. They are used in social science, robotics, and neurophysiology for solving tasks of classification, forecasting, pattern recognition, etc. In neuroscience, ANNs allow the recognition of specific forms of brain activity from multichannel EEG or MEG data. This makes the ANN an efficient computational core for brain-machine systems. However, despite significant achievements of artificial intelligence in recognition and classification of well-reproducible patterns of neural activity, the use of ANNs for recognition and classification of patterns in neural networks still requires additional attention, especially in ambiguous situations. According to this, in this research, we demonstrate the efficiency of application of the ANN for classification of human MEG trials corresponding to the perception of bistable visual stimuli with different degrees of ambiguity. We show that along with classification of brain states associated with multistable image interpretations, in the case of significant ambiguity, the ANN can detect an uncertain state when the observer doubts about the image interpretation. With the obtained results, we describe the possible application of ANNs for detection of bistable brain activity associated with difficulties in the decision-making process.

Visual perception affected by motivation and alertness controlled by a noninvasive brain-computer interface

The influence of motivation and alertness on brain activity associated with visual perception was studied experimentally using the Necker cube, which ambiguity was controlled by the contrast of its ribs. The wavelet analysis of recorded multichannel electroencephalograms (EEG) allowed us to distinguish two different scenarios while the brain processed the ambiguous stimulus. The first scenario is characterized by a particular destruction of alpha rhythm (8–12 Hz) with a simultaneous increase in beta-wave activity (20–30 Hz), whereas in the second scenario, the beta rhythm is not well pronounced while the alpha-wave energy remains unchanged. The experiments were carried out with a group of financially motivated subjects and another group of unpaid volunteers. It was found that the first scenario occurred mainly in the motivated group. This can be explained by the increased alertness of the motivated subjects. The prevalence of the first scenario was also observed in a group of subjects to whom images with higher ambiguity were presented. We believe that the revealed scenarios can occur not only during the perception of bistable images, but also in other perceptual tasks requiring decision making. The obtained results may have important applications for monitoring and controlling human alertness in situations which need substantial attention. On the base of the obtained results we built a brain-computer interface to estimate and control the degree of alertness in real time.

Classifying the Perceptual Interpretations of a Bistable Image Using EEG and Artificial Neural Networks

In order to classify different human brain states related to visual perception of ambiguous images, we use an artificial neural network (ANN) to analyze multichannel EEG. The classifier built on the basis of a multilayer perceptron achieves up to 95% accuracy in classifying EEG patterns corresponding to two different interpretations of the Necker cube. The important feature of our classifier is that trained on one subject it can be used for the classification of EEG traces of other subjects. This result suggests the existence of common features in the EEG structure associated with distinct interpretations of bistable objects. We firmly believe that the significance of our results is not limited to visual perception of the Necker cube images; the proposed experimental approach and developed computational technique based on ANN can also be applied to study and classify different brain states using neurophysiological data recordings. This may give new directions for future research in the field of cognitive and pathological brain activity, and for the development of brain-computer interfaces.

Critical slowing down in networks generating temporal complexity

We study a nonlinear Langevin equation describing the dynamic variable X(t), the mean field (order parameter) of a finite size complex network at criticality. The conditions under which the autocorrelation function of X shows any direct connection with criticality are discussed. We find that if the network is prepared in a state far from equilibrium, X(0)=1, the autocorrelation function is characterized by evident signs of critical slowing down as well as by significant aging effects, while the preparation X(0)=0 does not generate evident signs of criticality on X(t), in spite of the fact that the same initial state makes the fluctuating variable η(t)≡sgn(X(t)) yield significant aging effects. These latter effects arise because the dynamics of η(t) are directly dependent on crucial events, namely the re-crossings of the origin, which undergo a significant aging process with the preparation X(0)=0. The time scale dominated by temporal complexity, aging, and ergodicity breakdown of η(t) is properly evaluated by adopting the method of stochastic linearization which is used to explain the exponential-like behavior of the equilibrium autocorrelation function of X(t).