Frontiers in Computational Neuroscience最新文献

{"title":"Design principles for cost functions in optimal sensory coding.","authors":"Veronika Koren, Stefano Panzeri","doi":"10.3389/fncom.2026.1924396","DOIUrl":"https://doi.org/10.3389/fncom.2026.1924396","url":null,"abstract":"<p><p>Studies of neural codes for sensory variables are rooted in the long-established principle that neural codes are organized to encode as much information as possible about sensory stimuli. This principle has been successful in explaining many features of neural activity, suggesting that a key component of a cost function determining sensory neural coding from first principles is the stimulus encoding error. However, recent work has shown that neural cost functions may have other terms reflecting costs or limitations of the biological hardware. One key piece is the metabolic cost of neural activity. Including metabolic cost in the cost function leads to optimized spiking models that exhibit spike-triggered adaptation. Another possible component of the neural cost function is reliable downstream transmission of information. Neurons have short integration time constants, which make it difficult for information to be transmitted reliably unless inputs to neurons are correlated. This suggests that cost functions accounting for constraints on downstream readout promote codes with redundancy-inducing correlations. We review and discuss evidence that neural codes in the brain have features consistent with the predictions expected if cost functions have additional terms beyond the encoding error. Adaptation currents are indeed present in biological neurons with the time scales predicted by metabolically efficient codes, and neural population codes have information-reducing correlations that support downstream information transmission. Finally, we discuss future directions important for fully establishing the different contributions to neural cost functions.</p>","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1924396"},"PeriodicalIF":3.3,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13631812/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149106310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A subject-specific mechanistic model of affect links electroencephalography, cardiorespiratory feedback, and monoaminergic dynamics.","authors":"Masami Iwamoto","doi":"10.3389/fncom.2026.1875657","DOIUrl":"https://doi.org/10.3389/fncom.2026.1875657","url":null,"abstract":"<p><strong>Introduction: </strong>Affective experience emerges from interactions among cortical activity, bodily physiology, and contextual interpretation, yet the mechanisms linking these processes remain incompletely understood. Although EEG-based studies have advanced affect decoding, many data-driven approaches provide limited mechanistic insight into inter-individual variability. We developed a subject-specific, hypothesis-generating computational framework linking EEG-derived cortical activity and cardiorespiratory dynamics to subjective affect through putative neuromodulatory processes.</p><p><strong>Methods: </strong>The framework integrates a respiratory-circulatory subsystem with spiking-neuron models representing the locus coeruleus, ventral tegmental area, raphe nucleus, and basolateral amygdala. EEG band-power features were used as hypothesis-driven, indirect inputs to these modeled processes rather than as direct measures of deep-brain activity, while cardiorespiratory variables provided physiological feedback to the neural circuit. Model-derived activity was converted into putative norepinephrine-, dopamine-, and serotonin-related signals. Together with two fitted serotonin-related modulation coefficients, these signals constituted five latent parameters used to represent valence-arousal states. The modulation coefficients were estimated for each participant-stimulus observation by nonlinear optimization so that model-derived arousal and valence matched the corresponding subjective ratings. Model calibration and predictive validation were evaluated separately. Predictive validity was assessed using leave-one-stimulus-out cross-validation, in which each stimulus was held out in turn and predicted from regression equations fitted only to the remaining 15 stimuli.</p><p><strong>Results: </strong>The framework was evaluated using the AMIGOS dataset (nine participants, 16 video stimuli; 144 participant-stimulus observations). Under the calibration procedure, the five latent parameters reproduced participant-specific valence-arousal ratings. In cross-validation across all participants, the mean root mean square error was 0.380 ± 0.288 for arousal and 1.283 ± 1.512 for valence, indicating more consistent prediction of unseen stimuli for arousal and substantial inter-individual variability in valence prediction. Cross-participant regression showed weaker generalization, further supporting the importance of subject-specific modeling.</p><p><strong>Discussion: </strong>The proposed framework does not directly measure deep-brain or monoaminergic activity. Rather, it provides a physiologically informed and interpretable framework for generating hypotheses about how cortical activity, cardiorespiratory feedback, and putative neuromodulatory processes may jointly contribute to individual affective experience. The findings also highlight the need for validation in larger independent datasets using direct neurophysiological measurements.</p>","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1875657"},"PeriodicalIF":3.3,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13631044/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149106298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neural state reduction: a quantum-inspired framework for how latent conscious thought becomes reportable.","authors":"A K Pradeep","doi":"10.3389/fncom.2026.1865388","DOIUrl":"https://doi.org/10.3389/fncom.2026.1865388","url":null,"abstract":"<p><p>Human conscious thought often appears as a diffuse, unstable, and only partly verbalizable field until attention, questioning, decision, or speech stabilizes it into a reportable form. This manuscript develops a quantum-inspired neural framework for that transition. The proposal does not claim that consciousness literally depends on microscopic quantum computation, that the brain implements a physical quantum computer, or that neuronal tissue maintains macroscopic quantum coherence. Instead, it uses the mathematical architecture of quantum theory-state vectors, density matrices, projection operators, non-commuting measurements, interference, decoherence-like contextual stabilization, and state reduction-as a formal language for modeling the transformation from latent conscious potential to reportable thought. In the proposed framework, pre-report consciousness is represented as a high-dimensional latent neural-cognitive state over candidate thoughts. This state evolves under a neural-cognitive Hamiltonian shaped by memory, salience, affect, attention, language, task demands, and spontaneous associative dynamics. A question such as \"What are you thinking about?\" is modeled as a measurement context that selects a basis and projects the latent state into a reportable cognitive eigenstate. The framework, therefore, treats an introspective report not as a passive readout of a pre-existing mental object but as an active state-reducing operation. Illustrative simulations demonstrate three formal consequences: (1) Non-commuting introspective questions generate order effects; (2) selective reporting reduces candidate-state entropy and suppresses off-diagonal latent coherence; and (3) overlapping candidate thoughts can produce interference-like probability structures. The model generates testable predictions for experience sampling, electroencephalography (EEG)/magnetoencephalography (MEG)/functional magnetic resonance imaging (fMRI) representational decoding, intracranial recordings, and perturbation studies: Introspective reports should depend on question order; spoken reports should reduce latent-state entropy more strongly than silent attention; ambiguous thought states should exhibit interference-like probability signatures; and neural activity should transition from high-dimensional latent dynamics toward lower-dimensional, report-stabilized states. Neural state reduction provides a computational bridge among quantum cognition, consciousness science, metacognition, global access, and the neuroscience of reportability.</p>","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1865388"},"PeriodicalIF":3.3,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13627331/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149106495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial: Computational neuroethology: simulating natural behaviors.","authors":"Arij Daou, Angel A Caputi","doi":"10.3389/fncom.2026.1971192","DOIUrl":"https://doi.org/10.3389/fncom.2026.1971192","url":null,"abstract":"","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1971192"},"PeriodicalIF":3.3,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13617012/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149073830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial: Complex network dynamics of cognitive processing in health and disease: current knowledge and future research.","authors":"Simone Cauzzo, Camillo Porcaro, Chandan Kumar Behera, Suriya Prakash Muthukrishnan","doi":"10.3389/fncom.2026.1976480","DOIUrl":"10.3389/fncom.2026.1976480","url":null,"abstract":"","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1976480"},"PeriodicalIF":3.3,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13612520/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149058814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Going deeper with morphologically detailed neural networks by simulation-based gradient propagation.","authors":"Gan He, Kai Du, Tiejun Huang","doi":"10.3389/fncom.2026.1904220","DOIUrl":"10.3389/fncom.2026.1904220","url":null,"abstract":"<p><p>Morphologically detailed dendrites possess powerful computational capabilities but are computationally expensive. Therefore, exploring how large-scale, detailed multi-compartment neural networks achieve learning proves challenging, presenting significant obstacles both in learning algorithms and simulation infrastructures. Here, we provide an extension to the DeepDendrite framework to enable the construction and data-driven training of multi-layer, detailed multi-compartment neural networks with highly modularized layer components. The gradient at each layer is computed by simulating gradient-mirror neurons in the feedback pathway simultaneously with the detailed neurons in the feedforward pathway, providing a simulation-based implementation of backpropagation in detailed neural networks. We demonstrate comparable results on classic image classification datasets with fully-connected and convolutional architectures. Furthermore, we analyze transfer attack robustness between artificial neural networks (ANNs), detailed neural networks and single-compartment networks. In conclusion, we provide a useful framework to investigate learning in multi-layer detailed neural networks, potentially offering further insights into exploiting the computational potential of dendrites at a large scale.</p>","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1904220"},"PeriodicalIF":3.3,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13601319/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149037464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial: Modern applications of EEG in neurological and cognitive research.","authors":"Kuldeep Singh, Rajamanickam Yuvaraj, Geetanjali Sharma, Amit Joshi","doi":"10.3389/fncom.2026.1969406","DOIUrl":"https://doi.org/10.3389/fncom.2026.1969406","url":null,"abstract":"","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1969406"},"PeriodicalIF":3.3,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13601333/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149037498","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial: Machine learning algorithms and software tools for early detection and prognosis of schizophrenia.","authors":"Amit Singhal, Megha Agarwal, Animesh Kumar Paul, Bishal Lamichhane","doi":"10.3389/fncom.2026.1966648","DOIUrl":"https://doi.org/10.3389/fncom.2026.1966648","url":null,"abstract":"","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1966648"},"PeriodicalIF":3.3,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13597728/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149037448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reduced dynamical variability in depression: evidence from EEG and QIF-E network simulation.","authors":"Gabriel Moreno Cunha, Gisele Santana, Jose Garcia Vivas Miranda, Gilberto Corso, Thaise Graziele L de O Toutain, Marcelo M S Lima, Matheus Phellipe Brasil de Sousa, Gustavo Zampier Dos Santos Lima","doi":"10.3389/fncom.2026.1899952","DOIUrl":"10.3389/fncom.2026.1899952","url":null,"abstract":"<p><p>Alterations in neural signal complexity have been consistently reported in Major Depressive Disorder (MDD), suggesting changes in the underlying dynamics of brain activity. In this study, we investigate whether changes in neural signal complexity observed in MDD can be characterized using multiscale entropy (MSE) analysis, a method that quantifies temporal complexity across multiple scales. We analyzed electroencephalographic (EEG) recordings from individuals diagnosed with MDD and healthy controls, and compared these results with simulations of neural network models incorporating different forms of local electrical coupling. The EEG analysis revealed significant alterations in signal complexity in MDD, characterized by higher mean entropy at broader time scales and reduced inter-individual variability when compared to healthy controls. To explore potential dynamical mechanisms underlying these observations, we employed a Quadratic Integrate-and-Fire (QIF) neuronal network model with tunable local electrical coupling and small-world synaptic topology. Network topology and coupling parameters were systematically varied to assess their impact on signal complexity, without assuming a direct physiological correspondence between model components and specific biological mechanisms. MSE was used to quantify the irregularity and predictability of both empirical EEG signals and simulated network activity across multiple temporal scales. We found that network configurations lacking local electrical coupling reproduced key entropy features observed in the EEG signals of individuals with MDD, including increased mean entropy and reduced dispersion across realizations. In contrast, simulations with local electrical coupling exhibited lower average entropy and greater variability, resembling the entropy patterns observed in healthy control EEG data. These results suggest that differences in local coupling structure can modulate the balance between complexity and variability in network dynamics, potentially influencing the range of accessible dynamical states. Rather than establishing causality, this comparative analysis highlights how simplified models of local electrical coupling can phenomenologically account for entropy alterations observed in MDD, providing a computational framework for exploring links between network dynamics and large-scale brain signal complexity.</p>","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1899952"},"PeriodicalIF":3.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13558199/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148929490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial: Computational approaches to neurodegenerative diseases.","authors":"Milos Antonijevic","doi":"10.3389/fncom.2026.1952655","DOIUrl":"https://doi.org/10.3389/fncom.2026.1952655","url":null,"abstract":"","PeriodicalId":12363,"journal":{"name":"Frontiers in Computational Neuroscience","volume":"20 ","pages":"1952655"},"PeriodicalIF":3.3,"publicationDate":"2026-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13553467/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148918888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}