bioRxiv Science⌕ Search

Biology subjects

Friston, K. J.

Publications and source records attributed to Friston, K. J..

At least 19 recordsLinked to original sources
bioRxiv2026

Heterogeneity in neural network engagement supports individual differences in top-down attention

Top-down attention enables selective prioritisation of goal-relevant information, yet how attentional control varies across individuals remains unclear, particularly in relation to sensory processing. Here, we combined magnetoencephalography (MEG) and pupillometry--in older adults performing a spatial cueing task--to examine both shared and subject-specific aspects of top-down attention. Informative spatial cues elicited sustained activity in fronto-parietal and sensory regions, alpha lateralisation, and increased pupil dilation, replicating established neural and physiological markers. Crucially, these group-level patterns were accompanied by systematic intersubject variability: individuals with greater hearing loss showed increased preparatory engagement of precentral and postcentral gyri, and variation in pupil dilation predicted activity in frontal regions. These results reveal that top-down attention arises from heterogeneous network configurations shaped by sensory processing and strategic resource allocation. Overall, our findings underscore the need to move beyond group averages to capture the neural architecture underlying individual differences in top-down attention.

neuroscience↗
bioRxiv2025

A Canonical Microcircuit for Estimating Excitation/Inhibition (E/I) Balance

Excitation/inhibition (E/I) balance is crucial for maintaining healthy brain function and can be disrupted in various neurological and psychiatric disorders. Despite its importance, there are few tools to study E/I balance non-invasively in humans. Here, we propose a canonical microcircuit model to estimate E/I balance from non-invasive magneto- and electroencephalography (M/EEG) recordings by parameterising global pyramidal and inhibitory cell excitability. We first establish that E/I parameters are identifiable and recoverable. We then explore the effects of these new parameters and their interaction with other parameters in a series of simulations. To highlight the clinical relevance of this new model, we simulate changes in E/I balance and their impact on event-related potentials (ERPs) derived from paired-click, passive and active oddball paradigms, which are among the most robust clinical biomarkers of schizophrenia. Our simulations show that a loss of pyramidal cell excitability can explain reduced ERP amplitudes across all three paradigms, mirroring empirical findings in schizophrenia. This method may serve as a computational assay for estimating synaptopathy and E/I balance from non-invasive M/EEG recordings across various clinical conditions thereby advancing efforts to develop personalised interventions to restore E/I balance.

neuroscience↗
bioRxiv2024

Stimulus-repetition effects on macaque V1 and V4 microcircuits explain gamma-synchronization increase

Under natural conditions, animals repeatedly encounter the same visual scenes, objects or patterns repeatedly. These repetitions constitute statistical regularities, which the brain captures in an internal model through learning. A signature of such learning in primate visual areas V1 and V4 is the gradual strengthening of gamma synchronization. We used a V1-V4 Dynamic Causal Model (DCM) to explain visually induced responses in early and late epochs from a sequence of several hundred grating presentations. The DCM reproduced the empirical increase in local and inter-areal gamma synchronization, revealing specific intrinsic connectivity effects that could explain the phenomenon. In a sensitivity analysis, the isolated modulation of several connection strengths induced increased gamma. Comparison of alternative models showed that empirical gamma increases are better explained by (1) repetition effects in both V1 and V4 intrinsic connectivity (alone or together with extrinsic) than in extrinsic connectivity alone, and (2) repetition effects on V1 and V4 population input rather than output gain. The best input gain model included effects in V1 granular and superficial excitatory populations and in V4 granular and deep excitatory populations. Our findings are consistent with gamma reflecting bottom-up signal precision, which increases with repetition and, therefore, with predictability and learning. HighlightsO_LIWe model learning effects in macaque visual cortex using Dynamic Causal Modeling. C_LIO_LIMicrocircuit-level changes explain the repetition-induced gamma increases. C_LIO_LIThe best models include changes 1) within V1 and V4 and 2) in neuronal input gain. C_LIO_LIGamma may reflect bottom-up signal precision. C_LI

neuroscience↗
bioRxiv2023

Human hippocampal ripples tune cortical responses in uncertain visual contexts

To be able to encode information efficiently, our perceptual system should detect when situations are unpredictable (i.e., informative), and modulate brain dynamics to prepare for encoding. Here we show, with direct recordings from the human hippocampus and visual cortex, that after exposure to unpredictable visual stimulus streams, hippocampal ripple activity increases in frequency and duration prior to stimulus presentation, indicating context and experience-dependent prediction of predictability. Pre-stimulus hippocampal ripples suppress changes in visual (occipital) cortex gamma activity associated with uncertainty, and modulate post-stimulus prediction error gamma responses in higher-level visual (fusiform) cortex to surprising (i.e., unpredicted) stimuli. These results link hippocampal ripples with predictive coding accounts of neuronal message passing--and precision-weighted prediction errors--revealing a mechanism relevant for perceptual synthesis and subsequent memory encoding.

neuroscience↗
bioRxiv2022

Degeneracy in the neurological model of auditory speech repetition

In the neurological model of language, repeating heard speech involves four left hemisphere regions: primary auditory cortex for processing sounds; Wernickes area for processing auditory images of speech; Brocas area for processing motor images of speech; and primary motor cortex for overt speech articulation. Previous functional-MRI (fMRI) studies confirm that auditory repetition activates these regions. Here, we used dynamic causal modelling (DCM) to test how the four regions interact with each other during single word and pseudoword auditory repetition. Contrary to expectation, we found that, for both word and pseudoword repetition, the effective connectivity between Wernickes and Brocas areas was predominantly bidirectional and inhibitory; activity in the motor cortex could be driven by either Wernickes area or Brocas area; and the latter effect varied both within and between individuals. Such variability speaks to degenerate functional architectures that support auditory repetition and may explain resilience to functional loss after brain damage.

neuroscience↗
bioRxiv2022

Selective V1-to-V4 communication of attended stimuli mediated by attentional effects in V1

Selective attention implements preferential routing of attended stimuli, likely through increasing the influence of the respective synaptic inputs on higher-area neurons. As the inputs of competing stimuli converge onto postsynaptic neurons, presynaptic circuits might offer the best target for attentional top-down influences. If those influences enabled presynaptic circuits to selectively entrain postsynaptic neurons, this might explain selective routing. Indeed, when two visual stimuli induce two gamma rhythms in V1, only the gamma induced by the attended stimulus entrains gamma in V4. Here, we modeled induced responses with a Dynamic Causal Model for Cross-Spectral Densities and found that selective entrainment can be explained by attentional modulation of intrinsic V1 connections. Specifically, local inhibition was decreased in the granular input layer and increased in the supragranular output layer of the V1 circuit that processed the attended stimulus. Thus, presynaptic attentional influences and ensuing entrainment were sufficient to mediate selective routing. HIGHLIGHTSO_LIWe model selective visual attention in macaques using Dynamic Causal Modeling. C_LIO_LIIntrinsic V1 modulation can explain attention effects in V1-V4 communication. C_LIO_LIModulation of superficial and granular inhibition is key to induce the effects. C_LIO_LIThose modulations increase V1-V4 communication in a feedforward manner. C_LI GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=97 SRC="FIGDIR/small/484223v2_ufig1.gif" ALT="Figure 1"> View larger version (32K): org.highwire.dtl.DTLVardef@1cc79a7org.highwire.dtl.DTLVardef@146e6bdorg.highwire.dtl.DTLVardef@1d6bd6borg.highwire.dtl.DTLVardef@12dd663_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗
bioRxiv2021

Estimating anisotropy directly via neural timeseries

An isotropic dynamical system is one that looks the same in every direction, i.e., if we imagine - standing somewhere within an isotropic system, we would not be able to differentiate between different lines of sight. Conversely, anisotropy is a measure of the extent to which a system deviates from perfect isotropy, with larger values indicating greater discrepancies between the structure of the system along its axes. Here, we derive the form of a generalised scalable (mechanically similar) discretized field theoretic Lagrangian that allows for levels of anisotropy to be directly estimated via timeseries of arbitrary dimensionality. We generate synthetic data for both isotropic and anisotropic systems and, by using Bayesian model inversion and reduction, show that we can discriminate between the two datasets - thereby demonstrating proof of principle. We then apply this methodology to murine calcium imaging data collected in rest and task states, showing that anisotropy can be estimated directly from different brain states and cortical regions in an empirical in vivo biological setting. We hope that this theoretical foundation, together with the methodology and publicly available MATLAB code, will provide an accessible way for researchers to obtain new insight into the structural organization of neural systems in terms of how scalable neural regions grow - both ontogenetically during the development of an individual organism, as well as phylogenetically across species.

neuroscience↗
bioRxiv2021

Altered Effective Connectivity in Sensorimotor Cortices: a Novel Signature of Severity and Clinical Course in Depression

Functional neuroimaging research on depression has traditionally targeted neural networks associated with the psychological aspects of depression. In this study, instead, we focus on alterations of sensorimotor function in depression. We used resting-state functional MRI data and Dynamic Causal Modeling (DCM) to assess the hypothesis that depression is associated with aberrant effective connectivity within and between key regions in the sensorimotor hierarchy. Using hierarchical modeling of between-subject effects in DCM with Parametric Empirical Bayes we first established the architecture of effective connectivity in sensorimotor cortices. We found that in (interoceptive and exteroceptive) sensory cortices across participants, the backward connections are predominantly inhibitory whereas the forward connections are mainly excitatory in nature. In motor cortices these parities were reversed. With increasing depression severity, these patterns are depreciated in exteroceptive and motor cortices and augmented in the interoceptive cortex: an observation that speaks to depressive symptomatology. We established the robustness of these results in a leave-one-out cross validation analysis and by reproducing the main results in a follow-up dataset. Interestingly, with (non-pharmacological) treatment, depression associated changes in backward and forward effective connectivity partially reverted to group mean levels. Overall, altered effective connectivity in sensorimotor cortices emerges as a promising and quantifiable candidate marker of depression severity and treatment response. Significance StatementResearch into neurobiology of depression primarily focuses on its complex psychological aspects. Here, we propose an alternative approach and target sensorimotor alterations - a prominent but often neglected feature of depression. We demonstrated using resting-state fMRI data and computational modelling that top-down and bottom-up information flow in sensory and motor cortices is altered with increasing depression severity in a way that is consistent with depression symptoms. Depression associated changes were found to be consistent across sessions, amenable to treatment and of effect size sufficiently large to predict whether somebody has mild or severe depression. These results pave the way for a new avenue of research into the neural underpinnings of mental health conditions.

neuroscience↗
bioRxiv2020

Model-based prediction of muscarinic receptor function from auditory mismatch negativity responses

AO_SCPLOWBSTRACTC_SCPLOWDrugs affecting neuromodulation, for example by dopamine or acetylcholine, take centre stage among therapeutic strategies in psychiatry. These neuromodulators can change both neuronal gain and synaptic plasticity and therefore affect electrophysiological measures. An important goal for clinical diagnostics is to exploit this effect in the reverse direction, i.e., to infer the status of specific neuromodulatory systems from electrophysiological measures. In this study, we provide proof-of-concept that the functional status of cholinergic (specifically muscarinic) receptors can be inferred from electrophysiological data using generative (dynamic causal) models. To this end, we used epidural EEG recordings over two auditory cortical regions during a mismatch negativity (MMN) paradigm in rats. All animals were treated, across sessions, with muscarinic receptor agonists and antagonists at different doses. Together with a placebo condition, this resulted in five levels of muscarinic receptor status. Using a dynamic causal model - embodying a small network of coupled cortical microcircuits - we estimated synaptic parameters and their change across pharmacological conditions. The ensuing parameter estimates associated with (the neuromodulation of) synaptic efficacy showed both graded muscarinic effects and predictive validity between agonistic and antagonistic pharmacological conditions. This finding illustrates the potential utility of generative models of electrophysiological data as computational assays of muscarinic function. In application to EEG data of patients from heterogeneous spectrum diseases, e.g. schizophrenia, such models might help identify subgroups of patients that respond differentially to cholinergic treatments. SO_SCPLOWIGNIFICANCEC_SCPLOW SO_SCPLOWTATEMENTC_SCPLOWIn psychiatry, the vast majority of pharmacological treatments affect actions of neuromodulatory transmitters, e.g. dopamine or acetylcholine. As treatment is largely trial-and-error based, one of the goals for computational psychiatry is to construct mathematical models that can serve as "computational assays" and infer the status of specific neuromodulatory systems in individual patients. This neuro-modelling strategy has great promise for electrophysiological data in particular but requires careful validation. The present study demonstrates that the functional status of cholinergic (muscarinic) receptors can be inferred from electrophysiological data using dynamic causal models of neural circuits. While accuracy needs to be enhanced and our results must be replicated in larger samples, our current results provide proof-of-concept for computational assays of muscarinic function using EEG.

neuroscience↗
bioRxiv2019

Difficulties with speech-in-noise perception related to fundamental grouping processes in auditory cortex

In our everyday lives, we are often required to follow a conversation when background noise is present (\"speech-in-noise\" perception). Speech-in-noise perception varies widely--and people who are worse at speech-in-noise perception are also worse at fundamental auditory grouping, as assessed by figure-ground tasks. Here, we examined the cortical processes that link difficulties with speech-in-noise perception to difficulties with figure-ground perception using functional magnetic resonance imaging (fMRI). We found strong evidence that the earliest stages of the auditory cortical hierarchy (left core and belt areas) are similarly disinhibited when speech-in-noise and figure-ground tasks are more difficult (i.e., at target-to-masker ratios corresponding to 60% than 90% thresholds)--consistent with increased cortical gain at lower levels of the auditory hierarchy. Overall, our results reveal a common neural substrate for these basic (figure-ground) and naturally relevant (speech-in-noise) tasks--which provides a common computational basis for the link between speech-in-noise perception and fundamental auditory grouping.

neuroscience↗
bioRxiv2019

Interictal discharges spread along local recurrent networks between tubers and surrounding cortex

The presence of interictal epileptiform discharges on electroencephalography (EEG) may indicate increased epileptic seizure risk and on invasive EEG are the signature of the irritative zone. In highly epileptogenic lesions - such as cortical tubers in tuberous sclerosis - these discharges can be recorded with intracranial stereotactic EEG as part of the evaluation for epilepsy surgery. Yet the network mechanisms that underwrite the generation and spread of these discharges remain poorly understood, limiting their current diagnostic use.\n\nHere, we investigate the dynamics of interictal epileptiform discharges using a combination of quantitative analysis of invasive EEG recordings and mesoscale neural mass modelling of cortical dynamics. We first characterise spatially organised local dynamics of discharges recorded from 36 separate tubers in 8 patients with tuberous sclerosis. We characterise these dynamics with a set of competing explanatory network models using dynamic causal modelling. Bayesian model comparison of plausible network architectures suggests that the recurrent coupling between neuronal populations within - and adjacent to - the tuber core explains the travelling wave dynamics observed in these patient recordings.\n\nOur results - based on interictal activity - unify competing theories about the pathological organisation of epileptic foci and surrounding cortex in patients with tuberous sclerosis. Coupled oscillator dynamics have previously been used to describe ictal activity, where fast travelling ictal discharges are commonly observed within the recruited seizure network. The interictal data analysed here add the insight that this functional architecture is already established in the interictal state. This links observations of interictal EEG abnormalities directly to pathological network coupling in epilepsy, with possible implications for epilepsy surgery approaches in tuberous sclerosis.\n\nSignificance StatementInterictal epileptiform discharges (IEDs) are clinically important markers of an epileptic brain. Here we link local IED spread to network coupling through a combination of clinical recordings in paediatric patients with tuberous sclerosis complex, quantitative EEG analysis of interictal discharges spread, and Bayesian inference on coupled neural mass model parameters. We show that the kinds of interictal discharges seen in our patients require recurrent local network coupling extending beyond the putative seizure focus and that in fact only those recurrent coupled networks can support seizure-like and interictal dynamics when run in simulation. Our findings provide a novel integrated perspective on emergent epileptic dynamics in human patients.

neuroscience↗
bioRxiv2019

Reverse engineering neural networks to characterise their cost functions

This work considers a class of biologically plausible cost functions for neural networks, where the same cost function is minimised by both neural activity and plasticity. We show that such cost functions can be cast as a variational bound on model evidence under an implicit generative model. Using generative models based on Markov decision processes (MDP), we show, analytically, that neural activity and plasticity perform Bayesian inference and learning, respectively, by maximising model evidence. Using mathematical and numerical analyses, we then confirm that biologically plausible cost functions--used in neural networks--correspond to variational free energy under some prior beliefs about the prevalence of latent states that generate inputs. These prior beliefs are determined by particular constants (i.e., thresholds) that define the cost function. This means that the Bayes optimal encoding of latent or hidden states is achieved when, and only when, the networks implicit priors match the process that generates the inputs. Our results suggest that when a neural network minimises its cost function, it is implicitly minimising variational free energy under optimal or sub-optimal prior beliefs. This insight is potentially important because it suggests that any free parameter of a neural networks cost function can itself be optimised--by minimisation with respect to variational free energy.

neuroscience↗
bioRxiv2019

A Bayesian account of generalist and specialist formation under the Active Inference framework

This paper offers a formal account of policy learning, or habitual behavioural optimisation, under the framework of Active Inference. In this setting, habit formation becomes an autodidactic, experience-dependent process, based upon what the agent sees itself doing. We focus on the effect of environmental volatility on habit formation by simulating artificial agents operating in a partially observable Markov decision process. Specifically, we used a two-step maze paradigm, in which the agent has to decide whether to go left or right to secure a reward. We observe that in volatile environments with numerous reward locations, the agents learn to adopt a generalist strategy, never forming a strong habitual behaviour for any preferred maze direction. Conversely, in conservative or static environments, agents adopt a specialist strategy; forming strong preferences for policies that result in approach to a small number of previously-observed reward locations. The pros and cons of the two strategies are tested and discussed. In general, specialization offers greater benefits, but only when contingencies are conserved over time. We consider the implications of this formal (Active Inference) account of policy learning for understanding the relationship between specialisation and habit formation.\n\nAuthor SummaryActive inference is a theoretical framework that formalizes the behaviour of any organism in terms of a single imperative - to minimize surprise. Starting from this principle, we can construct simulations of simple \"agents\" (artificial organisms) that show the ability to infer causal relationships and learn. Here, we expand upon currently-existing implementations of Active Inference by enabling synthetic agents to optimise the space of behavioural policies that they can pursue. Our results show that by adapting the probabilities of certain action sequences (which may correspond biologically to the phenomenon of synaptic plasticity), and by rejecting improbable sequences (synaptic pruning), the agents can begin to form habits. Furthermore, we have shown our agents habit formation to be environment-dependent. Some agents become specialised to a constant environment, while other adopt a more general strategy, each with sensible pros and cons. This work has potential applications in computational psychiatry, including in behavioural phenotyping to better understand disorders.

neuroscience↗
bioRxiv2019

Simulating emotions: An active inference model of emotional state inference and emotion concept learning

The ability to conceptualize and understand ones own affective states and responses - or "emotional awareness" (EA) - is reduced in multiple psychiatric populations; it is also positively correlated with a range of adaptive cognitive and emotional traits. While a growing body of work has investigated the neurocognitive basis of EA, the neurocomputational processes underlying this ability have received limited attention. Here, we present a formal Active Inference (AI) model of emotion conceptualization that can simulate the neurocomputational (Bayesian) processes associated with learning about emotion concepts and inferring the emotions one is feeling in a given moment. We validate the model and inherent constructs by showing (i) it can successfully acquire a repertoire of emotion concepts in its "childhood", as well as (ii) acquire new emotion concepts in synthetic "adulthood," and (iii) that these learning processes depend on early experiences, environmental stability, and habitual patterns of selective attention. These results offer a proof of principle that cognitive-emotional processes can be modeled formally, and highlight the potential for both theoretical and empirical extensions of this line of research on emotion and emotional disorders.

neuroscience↗
bioRxiv2019

An active inference model of concept learning

Within computational neuroscience, the algorithmic and neural basis of structure learning remains poorly understood. Concept learning is one primary example, which requires both a type of internal model expansion process (adding novel hidden states that explain new observations), and a model reduction process (merging different states into one underlying cause and thus reducing model complexity via meta-learning). Although various algorithmic models of concept learning have been proposed within machine learning and cognitive science, many are limited to various degrees by an inability to generalize, the need for very large amounts of training data, and/or insufficiently established biological plausibility. Using concept learning as an example case, we introduce a novel approach for modeling structure learning - and specifically state-space expansion and reduction - within the active inference framework and its accompanying neural process theory. Our aim is to demonstrate its potential to facilitate a novel line of active inference research in this area. The approach we lay out is based on the idea that a generative model can be equipped with extra (hidden state or cause) slots that can be engaged when an agent learns about novel concepts. This can be combined with a Bayesian model reduction process, in which any concept learning - associated with these slots - can be reset in favor of a simpler model with higher model evidence. We use simulations to illustrate this models ability to add new concepts to its state space (with relatively few observations) and increase the granularity of the concepts it currently possesses. We also simulate the predicted neural basis of these processes. We further show that it can accomplish a simple form of one-shot generalization to new stimuli. Although deliberately simple, these simulation results highlight ways in which active inference could offer useful resources in developing neurocomputational models of structure learning. They provide a template for how future active inference research could apply this approach to real-world structure learning problems and assess the added utility it may offer.

neuroscience↗
bioRxiv2019

In the Body’s Eye: The Computational Anatomy of Interoceptive Inference

A growing body of evidence highlights the intricate linkage of exteroceptive perception to the rhythmic activity of the visceral body. In parallel, interoceptive inference theories of emotion and self-consciousness are on the rise in cognitive science. However, thus far no formal theory has emerged to integrate these twin domains; instead most extant work is conceptual in nature. Here, we introduce a formal model of cardiac active inference, which explains how ascending cardiac signals entrain exteroceptive sensory perception and confidence. Through simulated psychophysics, we reproduce the defensive startle reflex and commonly reported effects linking the cardiac cycle to fear perception. We further show that simulated interoceptive lesions blunt fear expectations, induce psychosomatic hallucinations, and exacerbate metacognitive biases. Through synthetic heart-rate variability analyses, we illustrate how the balance of arousal-priors and visceral prediction errors produces idiosyncratic patterns of physiological reactivity. Our model thus offers the possibility to computationally phenotype disordered brain-body interaction.

neuroscience↗
bioRxiv2018

Hippocampal-entorhinal transformations in abstract frames of reference

Knowing how anothers preferences relate to our own is a central aspect of everyday decision-making, yet how the brain performs this transformation is unclear. Here, we ask whether the putative role of the hippocampal-entorhinal system in transforming relative and absolute spatial coordinates during navigation extends to transformations in abstract decision spaces. During fMRI scanning, subjects learned a strangers preference for an everyday activity - relative to one of three personally known individuals - and subsequently decided how the strangers preference relates to the other two individuals preferences. We found that entorhinal cortex/subiculum signals exhibited reference frame-sensitive responses to the absolute distance between the ratings of the stranger and the familiar choice options. In contrast, striatal signals increased when accurately determining the ordinal position of choice options in relation to the stranger. Paralleling its role in navigation, these data implicate the entorhinal/subicular region in assimilating relatively coded knowledge within abstract metric spaces.

neuroscience↗
bioRxiv2018

Computational mechanisms of curiosity and goal-directed exploration

Successful behaviour depends on the right balance between maximising reward and soliciting information about the world. Here, we show how different types of information-gain emerge when casting behaviour as surprise minimisation. We present two distinct mechanisms for goal-directed exploration that express separable profiles of active sampling to reduce uncertainty. Hidden state exploration motivates agents to sample unambiguous observations to accurately infer the (hidden) state of the world. Conversely, model parameter exploration, compels agents to sample outcomes associated with high uncertainty, if they are informative for their representation of the task structure. We illustrate the emergence of these types of information-gain, termed active inference and active learning, and show how these forms of exploration induce distinct patterns of Bayes-optimal behaviour. Our findings provide a computational framework to understand how distinct levels of uncertainty induce different modes of information-gain in decision-making.

neuroscience↗