Frontiers in Systems Neuroscience最新文献

This article describes a biological neural network model that explains how humans learn to understand large language models and their meanings. This kind of learning typically occurs when a student learns from a teacher about events that they experience together. Multiple types of self-organizing brain processes are involved, including content-addressable memory; conscious visual perception; joint attention; object learning, categorization, and cognition; conscious recognition; cognitive working memory; cognitive planning; neural-symbolic computing; emotion; cognitive-emotional interactions and reinforcement learning; volition; and goal-oriented actions. The article advances earlier results showing how small language models are learned that have perceptual and affective meanings. The current article explains how humans, and neural network models thereof, learn to consciously see and recognize an unlimited number of visual scenes. Then, bi-directional associative links can be learned and stably remembered between these scenes, the emotions that they evoke, and the descriptive language utterances associated with them. Adaptive resonance theory circuits control model learning and self-stabilizing memory. These human capabilities are not found in AI models such as ChatGPT. The current model is called ChatSOME, where SOME abbreviates Self-Organizing MEaning. The article summarizes neural network highlights since the 1950s and leading models, including adaptive resonance, deep learning, LLMs, and transformers.

Techniques used over decades in brain-based neural network modeling are applied to understanding processes involved in psychoanalysis. Behavioral change is interpreted as a transition, using simulated annealing, from a less to a more optimal attractor in a competitive-cooperative dynamical system that includes analogs of the amygdala, prefrontal cortex, and hypothalamus, and the neurotransmitter norepinephrine. The article explores how psychoanalysis can facilitate the quest for the life that is as meaningful as possible. The resulting network theory allows for new understanding of several traditional Freudian concepts. The theory provides insights about the life and death drives. It also helps us understand object and narcissistic libido, and the contrast of healthy forms of libido based on autonomy vs. unhealthy forms based on dependence. This inquiry relates to the balance between self-interest and empathy, mediated by various areas of the limbic system. It illuminates transference, which involves both an emotional and intellectual relationship between the analyst and analysand, mediated by cognitive-emotional interactions in amygdala and orbitofrontal cortex. Sublimation, or redirection of socially inappropriate urges toward more appropriate behaviors, is interpreted via lateral inhibition between representations of similar complex behaviors.

Background: Shift work sleep disorder (SWSD) in nurses is highly prevalent and is increasingly recognized for its profound impact on human health. However, the brain functional network topology, which provides a comprehensive map of the brain's information processing architecture, remains partially understood in nurses with SWSD.

Methods: 45 nurses with SWSD and 45 healthy controls (HCs) underwent a resting-state functional magnetic resonance imaging (rs-fMRI) scan. Graph theoretical analysis was used to investigate alterations in brain functional network topology. Functional network connectivity was further examined in nurses with SWSD relative to HCs. Correlations between network metrics and clinical sleep scores were also examined.

Results: Compared to HCs, the SWSD group exhibited significantly lower global network metrics. Additionally, at the regional level, the SWSD group showed reduced nodal efficiency in specific regions, particularly within the visual processing areas and the caudate nucleus. Functional network connectivity analysis revealed a predominant pattern of weakened connectivity within the limbic network (LN), visual network (VN), default mode network (DMN), subcortical network (SN) and between the LN and SN in the SWSD group, although some inter-network connections were strengthened, predominantly the VN-ventral attention network (VAN), frontoparietal network (FPN)-VN, somatomotor network-VAN, and VN-DMN. Furthermore, poorer sleep quality correlated with reduced local efficiency in the visual cortex and insomnia severity was associated with weakened frontal connectivity.

Conclusions: This study reveals significant alterations in brain functional network topology and predominantly weakened functional connectivity across multiple brain networks, despite some strengthened inter-network links. These neuroimaging changes correlated with clinical measures of sleep disturbance. Our findings highlight compromised brain network organization in SWSD, offering insights into its neural mechanisms and potential biomarkers.

Background: Brain functional reorganization in primary blepharospasm (BSP) remains incompletely understood. This study aimed to add to the increasing knowledge by examining abnormalities in local spontaneous neural activity in this disorder.

Methods: Resting-state functional magnetic resonance imaging data were acquired from 32 medication-naïve patients with BSP and 32 age- and sex-matched healthy controls in this study. The imaging data were analyzed using the amplitude of low frequency fluctuation (ALFF) to measure spontaneous neural activity. Partial correlation analyses between the altered ALFF values and clinical variables (illness duration and Jankovic Rating Scale score) in patients with BSP were further conducted.

Results: Compared to healthy controls, medication-naïve patients with BSP exhibited significantly increased ALFF in the bilateral putamen and left premotor cortex and decreased ALFF in the bilateral thalamus (p < 0.05, threshold-free cluster enhancement with family-wise error correction for multiple comparisons). Furthermore, ALFF values in the left putamen in the patient group were positively correlated with illness duration (r = 0.53, p = 0.002).

Conclusion: Our findings reveal aberrant spontaneous neural activity within key regions of the motor control network in medication-naïve BSP patients. These ALFF alterations, especially the progressive changes observed in the putamen, provide novel insights into BSP neuropathophysiology and highlight the value of studying untreated cohorts to understand the disorder's intrinsic characteristics.

Spinal cord injury (SCI) often results in irreversible sensorimotor and autonomic dysfunction, severely limiting independence and quality of life. Individuals with clinically complete SCI (AIS A) have traditionally been considered beyond functional recovery. This pilot study aimed to evaluate the therapeutic potential and broader impact of epidural spinal cord stimulation (eSCS) on this population. Three participants with chronic, sensorimotor complete thoracic SCI received implantation of a 32-contact epidural stimulator targeting the lumbosacral spinal cord. Personalized stimulation protocols were developed to support lower limb activation and upright posture. Motor, autonomic, and quality of life outcomes were assessed over a 12-month follow-up. All participants achieved independent standing through eSCS-enabled muscle activation, despite the absence of voluntary movement. The intervention led to improvements in postural control and trunk stability, as well as clinically meaningful gains in autonomic functions, including bladder, bowel, and sexual health, alongside reductions in spasticity and neuropathic pain. Importantly, participants reported sustained and substantial improvements in their perceived quality of life. This study provides evidence that eSCS can restore key functional capacities and significantly enhance the quality of life for individuals with sensorimotor complete SCI.

McFadden's conscious electromagnetic information (CEMI) field theory proposes that the human brain functions as a hybrid digital-EM field computer. The digital computations are implemented by the matter-based neuronal-synaptic network analogous to conventional digital computers operating Boolean-like logic gates nonconsciously and in parallel. Yet neuronal electrical firing and synaptic transmission generate the brain's immaterial but equally physical endogenous electromagnetic (EM) input into the brain's CEMI field. The CEMI field is proposed to implement analogue information processing through constructive and destructive wave mechanical interference. The output of this field-based processing is uploaded by EM field-sensitive neurons via voltage-gated ion channels to generate conscious actions. According to the theory, non-conscious brain processing occurs solely within the EM field-insensitive digital neuronal network, enabling fast, parallel computations, but cannot form complex, integrated concepts, so it is limited to specialised functions necessary for tasks like motor coordination. In contrast, conscious thought arises from EM field interactions, where integrated information is encoded and processed holistically to deliver general intelligence and creativity as its output. Because the brain's EM field is singular, conscious processing occurs serially, allowing our mind to hold only one thought at a time. This paper proposes a route towards developing novel hybrid computers that, like the human brain, similarly operate both modes of computation to deliver general intelligent and potentially conscious AI.

This perspective piece presents the concept of the role and mechanisms of cells' electromagnetic communication. These data deepen the scientific understanding of the fundamental aspects of the phenomenon of human life. A promising model of biophoton signaling as a scientific tool for further developing of biophotonics of the human body is substantiated.

Multisensory information is combined across the cortex and assimilated into the continuous production of ongoing behavior. In the hippocampus, theta oscillations (4-12 Hz) radiate as large-scale traveling waves, and serve as a scaffold for neuronal ensembles of multisensory information involved in memory and movement-related processing. An extension of such an encoding framework across the neocortex could similarly serve to bind disparate multisensory signals into ongoing, coherent, phase-coded processes. Whether the neocortex exhibits unique large-scale traveling waves distinct from that of the hippocampus, however, remains unknown. Here, using cortex-wide electrocorticography in freely moving mice, we find that theta oscillations are organized into bilaterally-symmetric spatiotemporal "modes" that span virtually the entire neocortex. The dominant mode (Mode 1) is a divergent traveling wave that originates from retrosplenial cortex and whose amplitude correlates with mouse speed. Secondary modes are asynchronous spiral waves centered over primary somatosensory cortex (Modes 2 and 3), which become prominent during rapid drops in amplitude and synchrony (null spikes) and which underlie a phase reset of Mode 1. These structured cortex-wide traveling waves may provide a scaffold for large-scale phase-coding of information across the cortex.

The brain criticality hypothesis has been a central research topic in theoretical neuroscience for two decades. This hypothesis suggests that the brain operates near the critical point at the boundary between order and disorder, where it acquires its information-processing capabilities. The mechanism that maintains this critical state has been proposed as a feedback system known as self-organized criticality (SOC); brain parameters, such as synaptic plasticity, are regulated internally without external adjustment. Therefore, clarifying how SOC occurs may provide insights into the mechanisms that maintain brain function and cause brain disorders. From the standpoint of neural network structures, the topology of neural circuits also plays a crucial role in information processing, with healthy neural networks exhibiting small world, scale-free, and modular characteristics. However, how these network structures affect SOC remains poorly understood. In this study, we conducted numerical simulations using a simplified neural network model to investigate how network structure may influence SOC. Our results reveal that the time scales at which synaptic plasticity operates to achieve a critical state differ depending on the network structure. Additionally, we observed Dragon king phenomena associated with abnormal neural activity, depending on the network structure and synaptic plasticity time scales. Notably, Dragon king was observed over a wide range of synaptic plasticity time scales in scale-free networks with high-degree hub nodes. These findings highlight the potential importance of neural network topology in shaping SOC dynamics in simplified models of neural systems.

Nanoscale motility of cells is a fundamental phenomenon, closely associated with biological status and response to environmental solicitations, whose investigation has disclosed new perspectives for the comprehension of cell behavior and fate. To investigate intracellular interactions, we designed an experiment to monitor movements of clusters of neuroblastoma cells (SH-SY5Y) growing on a nanomechanical oscillator (nanomotion sensor) suspended few hundreds of microns over the surface of a Petri dish where other neuroblastoma cells are freely moving. We observed that the free-to-move cells feel the presence of cells on the nearby nanosensor (at a distance of up to 300 microns) and migrate toward them, even in presence of environmental hampering factors, such as medium microflows. The interaction is bidirectional since, as evidenced by nanomotion sensing, the cells on the sensor enhance their motion when clusters of freely moving cells approach. Considering the geometry and environmental context, our observations extend beyond what can be explained by sensing of chemical trackers, suggesting the presence of other physical mechanisms. We hypothesize that the acoustic field generated by cell vibrations can have a role in the initial recognition between distant clusters. Integrating our findings with a suitable wave propagation model, we show that mechanical waves produced by cellular activity have sufficient energy to trigger mechanotransduction in target cells hundreds of microns away. This interaction can explain the observed distance-dependent patterns of cellular migration and motion alteration. Our results suggest that acoustic fields generated by cells can mediate cell-cell interaction and contribute to signaling and communication.