Article
Neurosciences
Qunjun Liang, Jinhui Li, Senning Zheng, Jiajun Liao, Ruiwang Huang
Summary: Hierarchical planning (HP) is a strategy that optimizes planning by storing lower-level steps as subgoals. Previous studies have identified the involvement of dmPFC, PMC, and SPL in the computation process of HP, but their interaction and contribution to HP computation remains unclear. Through an fMRI experiment, we confirmed the activity of dmPFC, PMC, and SPL, and used DCM and PEB models to quantify their connectivity and influence on response time.
Article
Neurosciences
Francisca Ferreira, Harith Akram, John Ashburner, Ludvic Zrinzo, Hui Zhang, Christian Lambert
Summary: The study found that the location of cd-Vim was more variable on the right during the reconstruction of DTCp using probabilistic tractography, and no significant effect of any confounds was observed. This work demonstrates significant individual anatomical variability of the cd-Vim that atlas-based targeting fails to capture.
Article
Neurosciences
Ya-Ning Chang, Ajay D. D. Halai, Matthew A. A. Lambon Ralph
Summary: Tractography is widely used in studying human brain connectivity, but the issue of systematically thresholding and comparing connectivity values for different track lengths across studies remains unsolved. This study utilized Monte Carlo derived distance-dependent distributions (DDDs) to generate thresholds for connections of varying lengths. The approach was applied to generate a language connectome, which showed expected structural connectivity. The findings demonstrate the feasibility and applicability of the DDD approach for thresholding probabilistic tracking datasets.
HUMAN BRAIN MAPPING
(2023)
Review
Neurosciences
Ines Pereira, Stefan Frassle, Jakob Heinzle, Dario Schobi, Cao Tri Do, Moritz Gruber, Klaas E. Stephan
Summary: Dynamic Causal Modeling (DCM) is a Bayesian framework used to infer hidden neuronal states based on brain activity measurements. While promising for understanding human brain dynamics, DCM variants can be challenging to fully understand due to their complexity and reliance on concepts from multiple fields. Solid theoretical knowledge of the models is crucial to avoid pitfalls in their application and interpretation of results.
Article
Mathematics
Liangjun Yu, Shengfeng Gan, Yu Chen, Dechun Luo
Summary: This paper introduces Naive Bayes and its improved model IWHNB, which combines the improved HNB model with instance weights to achieve significant improvements in classification performance.
Article
Neurosciences
Amirhossein Jafarian, Laura Hughes, Natalie Adams, Juliette Lanskey, Michelle Naessens, Matthew Rouse, Alexander G. Murley, Karl J. Friston, James B. Rowe
Summary: We propose a hierarchical empirical Bayesian framework for testing hypotheses on neurotransmitter concentration using ultra-high field magnetic resonance spectroscopy (7T-MRS) and magnetoencephalography data (MEG). A two-level model is developed to estimate the synaptic connectivity parameters and infer the influence of neurotransmitter levels on synaptic connections. The method is validated using resting-state MEG and 7T-MRS data from healthy adults, and it shows reliable results for hypothesis testing. This approach has great potential for studying neurological and psychiatric disorders and their response to pharmacological interventions.
Article
Engineering, Civil
Teng Wang, Zhila Bahrami, Guillaume Renaud, Chunsheng Yang, Min Liao, Zheng Liu
Summary: This paper proposes a probabilistic model for predicting fatigue crack growth and derives the posterior distribution of material parameters through Bayesian analysis. Based on this, a modified Paris-Erdogan model is adopted for probabilistic prediction.
Article
Neurosciences
Alessandro Crimi, Luca Dodero, Fabio Sambataro, Vittorio Murino, Diego Sona
Summary: This paper proposes a constrained autoregressive model to understand how brain structure modulates function and discover novel biomarkers. The model includes indirect connections and can be used with raw and deconvoluted BOLD signal, showing results closer to reality.
Article
Neurosciences
Alireza Borghei, Irem Kapucu, Robert Dawe, Mehmet Kocak, Sepehr Sani
Summary: The role of massa intermedia (MI) in normal neurocognitive function is unclear, but its absence has been linked to psychiatric disorders. Recent studies suggest MI may act as a midline white matter conduit with strong connectivity to limbic and cognitive regions of the brain. Women exhibit stronger connectivity through their MI compared to men.
HUMAN BRAIN MAPPING
(2021)
Article
Neurosciences
Amirhossein Jafarian, Peter Zeidman, Rob C. Wykes, Matthew Walker, Karl J. Friston
Summary: Adiabatic dynamic causal modelling is a method for inferring slow changes in biophysical parameters controlling fast neuronal fluctuations. It relies on established neural mass models and an adiabatic approximation to summarize fast neuronal states using second order statistics. The method introduces a circular causality involving synaptic parameters and neuronal activity, and is validated through simulations and an illustrative application to seizure activity in an animal model.
Article
Psychology, Clinical
Yinhuan Xu, Shaoqiang Han, Yarui Wei, Ruiping Zheng, Jingliang Cheng, Yan Zhang
Summary: This study reveals the characteristics of abnormal brain connectivity in patients with obsessive-compulsive disorder (OCD) and identifies different connectivity patterns within multiple brain networks through a large sample comparison. The results show a dysregulation among the default mode, salience, frontoparietal, and cerebellum networks, highlighting the importance of these four networks in top-down control for goal-directed behavior.
PSYCHOLOGICAL MEDICINE
(2023)
Review
Neurosciences
Eduardo A. Aponte, Yu Yao, Sudhir Raman, Stefan Frassle, Jakob Heinzle, Will D. Penny, Klaas E. Stephan
Summary: This paper introduces the use of thermodynamic integration (TI) for Bayesian model selection and averaging in the context of generative modeling of neuroimaging data. TI is based on Markov chain Monte Carlo sampling, which is accurate but slower than variational Bayes.
COGNITIVE NEURODYNAMICS
(2022)
Article
Computer Science, Artificial Intelligence
Bo Wang, Xiaodong Liu, Ming Chi, Yao Li
Summary: This article proposes a probabilistic weighted high-order fuzzy time series forecasting model using Bayesian network to address complex relationships and uncertainty in time series. The combination of fuzzy relationships and dependence relationships provides a comprehensive representation of the complex relationships. The proposed method calculates fuzzy-probabilistic weights to model uncertainty and has been validated to outperform existing models.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Neurosciences
Diego Szczupak, Pamela Meneses Iack, Danielle Rayee, Cirong Liu, Roberto Lent, Fernanda Tovar-Moll, Afonso C. Silva
Summary: The corpus callosum is the primary pathway for interhemispheric communication in the brain, and investigating its connectivity is crucial for understanding the brain's organization. Previous studies have underestimated the presence of heterotopic connections, but using advanced imaging techniques, it was found that around 75% of callosal connections are heterotopic. These heterotopic connections play an important role in determining the global properties of brain networks.
Article
Biochemical Research Methods
Frederik Van de Steen, Dimitris Pinotsis, Wouter Devos, Nigel Colenbier, Iege Bassez, Karl Friston, Daniele Marinazzo
Summary: This study used dynamic causal modeling to investigate the differences in alpha activity between eyes-open and eyes-closed conditions during resting-state EEG recordings. The findings suggest that there is increased alpha power over the posterior cortex in eyes-closed conditions and that this difference is mainly driven by intrinsic and extrinsic connectivity modulations within the visual cortex. The results highlight the importance of inhibitory intrinsic connections in generating alpha rhythms.
PLOS COMPUTATIONAL BIOLOGY
(2022)
Review
Biochemistry & Molecular Biology
Karl Friston
Summary: This review explores computational psychiatry from the perspective of pathophysiology, using generative models to explain psychopathology. It discusses the brain from cognitive and computational neuroscience viewpoints, providing a formal description of neuronal message passing, distributed processing, and belief propagation in neuronal networks. It also examines how dysconnections in the brain can lead to abnormal belief updating and false inference, and explores the use of computational models in various psychiatric research areas, including computational neuropsychology, computational phenotyping, and computational nosology.
MOLECULAR PSYCHIATRY
(2023)
Article
Psychiatry
Yukiko Matsumoto, Satoshi Nishida, Ryusuke Hayashi, Shuraku Son, Akio Murakami, Naganobu Yoshikawa, Hiroyoshi Ito, Naoya Oishi, Naoki Masuda, Toshiya Murai, Karl Friston, Shinji Nishimoto, Hidehiko Takahashi
Summary: This study used functional magnetic resonance imaging (fMRI) to evaluate the large-scale network structures of concept representations in patients with schizophrenia and found that their semantic networks exhibited differences and were associated with thought disorders. This provides pathophysiological evidence for the loosening of associations in schizophrenia.
SCHIZOPHRENIA BULLETIN
(2023)
Article
Computer Science, Artificial Intelligence
Natalie Kastel, Casper Hesp, K. Richard Ridderinkhof, Karl J. Friston
Summary: This paper proposes a testable deep active inference formulation of social behavior and conducts simulations of cumulative culture. By considering cultural transmission as a bi-directional process of communication and social exchange as a process of active inference, the study discovers that cumulative culture emerges from belief updating through a joint minimization of uncertainty.
FRONTIERS IN NEUROROBOTICS
(2023)
Article
Public, Environmental & Occupational Health
Cam Bowie, Karl Friston
Summary: This study analyzed the COVID-19 epidemic in the past 12 months and made predictions for the next year based on this analysis. It found that changes in transmissibility and public behavior led to an underestimation of the severity of the epidemic in previous predictions. The projections indicate that the number of infections in the coming year will be three times larger than last year, leading to more deaths and economic consequences.
FRONTIERS IN PUBLIC HEALTH
(2023)
Article
Physics, Multidisciplinary
Karl Friston, Daniel A. Friedman, Axel Constant, V. Bleu Knight, Chris Fields, Thomas Parr, John O. Campbell
Summary: This paper presents a variational formulation of natural selection, focusing on the nature of 'things' and how different 'kinds' of 'things' are individuated from each other and influence each other. Bayesian mechanics is used to understand the relationship between slow phylogenetic processes and fast phenotypic processes. The main result is the formulation of adaptive fitness as a phenotypic fitness path integral. Paths of least action at both phenotypic and phylogenetic scales can be seen as inference and learning processes respectively.
Article
Neurosciences
Marion Rouault, Ines Pereira, Herman Galioulline, Stephen M. Fleming, Klaas Enno Stephan, Zina-Mary Manjaly
Summary: Numerous disorders are characterized by fatigue, particularly in multiple sclerosis (MS), where fatigue significantly impacts quality of life. However, empirical data on interoception and metacognition in relation to fatigue in MS are scarce. This study examined these factors in a sample of 71 individuals with MS and found associations between interoceptive awareness and fatigue, as well as autonomic function and metacognition. Additionally, machine learning showed that fatigue levels could be predicted from questionnaire-based measures of interoception and sleep quality.
EUROPEAN JOURNAL OF NEUROSCIENCE
(2023)
Article
Neurosciences
R. L. Carhart-Harris, S. Chandaria, D. E. Erritzoe, A. Gazzaley, M. Girn, H. Kettner, P. A. M. Mediano, D. J. Nutt, F. E. Rosa, L. Roseman, C. Timmermann, B. Weiss, R. J. Zeifman, K. J. Friston
Summary: This theoretical article proposes a new model of a general factor of psychopathology, using the concept of 'canalization'. It distinguishes between two types of plasticity: 'TEMP' and 'canalization', which can be differentiated by their relationship to precision or inverse variance. The authors argue that 'pathological' phenotypes develop through mechanisms of canalization and increased model precision, as a response to adversity and distress. They suggest that TEMP, along with psychological support, can counter canalization and offer experiments and measures to test their hypotheses.
Correction
Computer Science, Artificial Intelligence
Natalie Kastel, Casper Hesp, K. Richard Ridderinkhof, Karl J. Friston
FRONTIERS IN NEUROROBOTICS
(2023)
Article
Pediatrics
Zoe McParlin, Francesco Cerritelli, Andrea Manzotti, Karl J. Friston, Jorge E. Esteves
Summary: Therapeutic affective touch is crucial for survival, nurturing supportive interactions, and promoting overall health. This paper presents an integrative model that combines therapeutic touch and communication to achieve biobehavioural synchrony. It explains the neurophysiological and behavioural mechanisms of developing synchronous relationships through touch and emphasizes the importance of therapeutic touch in building a solid therapeutic alliance.
FRONTIERS IN PEDIATRICS
(2023)
Review
Clinical Neurology
Achim Schilling, William Sedley, Richard Gerum, Claus Metzner, Konstantin Tziridis, Andreas Maier, Holger Schulze, Fan-Gang Zeng, Karl J. Friston, Patrick Krauss
Summary: This article reviews recent work at the intersection of artificial intelligence, psychology, and neuroscience, using tinnitus as an example of auditory phantom perception. The authors discuss the reasons behind the emergence of auditory phantom perceptions and their crucial role in healthy auditory perception. They propose that neural noise along the auditory pathway is generated as a compensatory mechanism and can be misinterpreted as auditory input, leading to tinnitus. The principles of predictive coding and adaptive stochastic resonance are identified as the most explanatory factors for phantom perceptions and may also improve machine learning techniques.
Review
Clinical Neurology
Kausar Raheel, Gemma Deegan, Irene Di Giulio, Diana Cash, Katarina Ilic, Valentina Gnoni, K. Ray Chaudhuri, Panagis Drakatos, Rosalyn Moran, Ivana Rosenzweig
Summary: Past research suggests that there are more cases and severe clinical manifestations of alpha-synucleinopathies in men, indicating potential neuroprotective properties of female sex hormones, especially estrogen. However, the underlying mechanisms of this effect are not well understood. This study aimed to systematically review and critically assess the current evidence on sex and gender differences in alpha-synucleinopathies.
FRONTIERS IN NEUROLOGY
(2023)
Article
Psychology, Developmental
Thomas Gauduel, Camille Blondet, Sibylle Gonzalez-Monge, James Bonaiuto, Alice Gomez
Summary: Developmental coordination disorder (DCD) affects the quality of life and coordinated actions in 5% of school-aged children. This study found that children with DCD have lower accuracy in sensory tasks compared to motor tasks. Additionally, children with DCD exhibited larger errors or synkinesis in both tasks. The results support the hypothesis of imprecise body representations in DCD.
DEVELOPMENTAL SCIENCE
(2023)
Article
Psychology, Experimental
Denis Brouillet, Karl Friston
Summary: The brain is known to be a predictive organ that predicts sensory content and the accuracy of its predictions. It must infer the reliability of its own beliefs in order to predict the precision of its predictions. This recognition process leads to the concept of "fluency", which is the perception of having a precise understanding of sensory processes. Changes in fluency, from unfelt to felt, are recognized and realized when updating predictions about accuracy.
CONSCIOUSNESS AND COGNITION
(2023)
Review
Biology
Karl Friston, Lancelot Da Costa, Dalton A. R. Sakthivadivel, Conor Heins, Grigorios A. Pavliotis, Maxwell Ramstead, Thomas Parr
Summary: This paper introduces a path integral formulation of the free energy principle to describe the trajectories of particles over time. By employing the principle of least action, it is possible to simulate the behavior of particles in exchange with their external environment. The paper discusses various types of particles and their different levels of inference or sentience.
PHYSICS OF LIFE REVIEWS
(2023)
Meeting Abstract
Mathematical & Computational Biology
Noor Sajid, Laura Convertino, Victorita Neacsu, Thomas Parr, Karl Friston
JOURNAL OF COMPUTATIONAL NEUROSCIENCE
(2023)
Article
Neurosciences
Jose Sanchez-Bornot, Roberto C. Sotero, J. A. Scott Kelso, Ozguer Simsek, Damien Coyle
Summary: This study proposes a multi-penalized state-space model for analyzing unobserved dynamics, using a data-driven regularization method. Novel algorithms are developed to solve the model, and a cross-validation method is introduced to evaluate regularization parameters. The effectiveness of this method is validated through simulations and real data analysis, enabling a more accurate exploration of cognitive brain functions.