Bayesian Brain Hypothesis Research Articles

Dyspnea, dysfunctional breathing disorders, and the Bayesian brain hypothesis.

Clinicians frequently evaluate patients who present with dyspnea. This term describes uncomfortable breathing during physical activity, and the intensity or degree of dyspnea can vary in an individual depending on circumstances and between individuals. In some cases, the level of dyspnea appears out of proportion to other information relevant to the cardiorespiratory system, and this situation has been described as dysfunctional breathing. The Bayesian brain hypothesis helps clinicians understand this symptom in these patients. This hypothesis suggests that prior experiences with dyspnea during physical activity or a respiratory disorder provide the background that is used to interpret current symptoms. This review outlines problems associated with the use of the term "dyspnea" and briefly describes how the Bayesian brain hypothesis might help clinicians understand this symptom better.

Pain perception as hierarchical Bayesian inference: A test case for the theory of constructed emotion.

An intriguing perspective about human emotion, the theory of constructed emotion considers emotions as generative models according to the Bayesian brain hypothesis. This theory brings fresh insight to existing findings, but its complexity renders it challenging to test experimentally. We argue that laboratory studies of pain could support the theory because although some may not consider pain to be a genuine emotion, the theory must at minimum be able to explain pain perception and its dysfunction in pathology. We review emerging evidence that bear on this question. We cover behavioral and neural laboratory findings, computational models, placebo hyperalgesia, and chronic pain. We conclude that there is substantial evidence for a predictive processing account of painful experience, paving the way for a better understanding of neuronal and computational mechanisms of other emotions.

Bayesian brain computing and the free-energy principle: an interview with Karl Friston.

The free-energy principle entails the Bayesian brain hypothesis that can be implemented by many schemes considered in this field. The combination of multimodal brain imaging and free-energy minimization has shown promise in unraveling complex brain dynamics and understanding the interactions among distinct brain regions. The Bayesian mechanics of brain computing gives a unique route to understanding authentic (neuromimetic) intelligence and, more importantly, points towards the development of brain-inspired intelligence. NSR spoke to a leading theoretical neuroscientist and authority on brain imaging-Karl Friston, the inventor of statistical parametric mapping, voxel-based morphometry and dynamic causal modeling. Friston is also known for his contributions to theoretical biology in the form of the free-energy principle and applications such as active inference. Friston is currently the Scientific Director of the Wellcome Trust Centre for Neuroimaging, Professor of Neuroscience at Queen Square Institute of Neurology, University College London and Honorary Consultant at The National Hospital for Neurology and Neurosurgery, UK.

Animal affect, welfare and the Bayesian brain.

According to the Bayesian brain hypothesis, the brain can be viewed as a predictive machine, such that predictions (or expectations) affect how sensory inputs are integrated. This means that in many cases, affective responses may depend more on the subject's perception of the experience (driven by expectations built on past experiences) rather than on the situation itself. Little research to date has applied this concept to affective states in animals. The aim of this paper is to explore how the Bayesian brain hypothesis can be used to understand the affective experiences of animals and to develop a basis for novel predictions regarding animal welfare. Drawing from the literature illustrating how predictive processes are important to human well-being, and are often impaired in affective disorders, we explore whether the Bayesian brain theories may help understanding animals' affective responses and whether deficits in predictive processes may lead to previously unconsidered welfare consequences. We conclude that considering animals as predictive entities can improve our understanding of their affective responses, with implications for basic research and for how to provide animals a better life.

Bayesian encoding and decoding as distinct perspectives on neural coding.

The Bayesian brain hypothesis is one of the most influential ideas in neuroscience. However, unstated differences in how Bayesian ideas are operationalized make it difficult to draw general conclusions about how Bayesian computations map onto neural circuits. Here, we identify one such unstated difference: some theories ask how neural circuits could recover information about the world from sensory neural activity (Bayesian decoding), whereas others ask how neural circuits could implement inference in an internal model (Bayesian encoding). These two approaches require profoundly different assumptions and lead to different interpretations of empirical data. We contrast them in terms of motivations, empirical support and relationship to neural data. We also use a simple model to argue that encoding and decoding models are complementary rather than competing. Appreciating the distinction between Bayesian encoding and Bayesian decoding will help to organize future work and enable stronger empirical tests about the nature of inference in the brain.

Puzzled by dysfunctional breathing disorder(s)? Consider the Bayesian brain hypothesis!

There is currently growing clinical concern regarding dysfunctional breathing disorder(s) (DBD), an umbrella term for a set of multidimensional clinical conditions that are characterized by altered breathing pattern associated with a variety of intermittent or chronic symptoms, notably dyspnea, in the absence or in excess of, organic disease. However, several aspects of DBD remain poorly understood and/or open to debate, especially the inconsistent relationship between the array of experienced symptoms and their supposedly underlying mechanisms. This may be partly due to a more general problem, i.e., the prevailing way we conceptualize symptoms. In the present article, after a brief review of the different aspects of DBD from the current perspective, I submit a call for considering DBD under the innovating perspective of the Bayesian brain hypothesis, i.e., a potent and novel model that fundamentally changes our views on symptom perception.

Correlations between within-subject variability of pain intensity reports and rubber hand illusion proprioceptive drift

IntroductionConsistent with the Bayesian brain hypothesis, the within-subject variability of pain intensity reports as captured with the Focused Analgesia Selection Test (FAST) might be a surrogate measure of the certainty in ascending noxious signals. The outcomes of a non-pain-related task, the rubber hand illusion, were hypothesized to reflect the same construct. This study aimed to explore whether within-subject differences in variability of pain intensity reports and the outcomes of the rubber hand illusion might be related. MethodsNonclinical participants underwent the classic rubber hand illusion under synchronous (experimental) and asynchronous (control) conditions. Two outcomes were assessed: proprioceptive drift and feeling of ownership. Thereafter, participants underwent the FAST to assess the within-subject variability of pain reports in response to heat stimuli. Intraclass correlation (ICC) and the correlation coefficient (R2) were the main outcomes. Spearman’s correlations were used to assess associations between the outcomes of the 2 tasks. ResultsThirty-six volunteers completed the study. Both FAST outcomes—ICC (Spearman’s r = 0.355, p = 0.033) and R2 (Spearman’s r = 0.349, p = 0.037)—were positively correlated with proprioceptive drift in the synchronous but not asynchronous conditions (p > 0.05). The subjective feeling of ownership and FAST outcomes did not correlate (p > 0.05). ConclusionsThe associations between the 2 tasks’ outcomes imply that both tasks at least partly assess similar constructs. Current knowledge suggests that this construct represents the person’s certainty in perceiving ascending sensory signals, or, in Bayesian terminology, the certainty of the likelihood.

Longing to act: Bayesian inference as a framework for craving in behavioral addiction

Traditionally, craving is considered a defining feature of drug addiction. Accumulating evidence suggests that craving can also exist in behavioral addictions (e.g., gambling disorder) without drug-induced effects. However, the degree to which mechanisms of craving overlap between classic substance use disorders and behavioral addictions remains unclear. There is, therefore, an urgent need to develop an overarching theory of craving that conceptually integrates findings across behavioral and drug addictions. In this review, we will first synthesize existing theories and empirical findings related to craving in both drug-dependent and -independent addictive disorders. Building on the Bayesian brain hypothesis and previous work on interoceptive inference, we will then propose a computational theory for craving in behavioral addiction, where the target of craving is execution of an action (e.g., gambling) rather than a drug. Specifically, we conceptualize craving in behavioral addiction as a subjective belief about physiological states of the body associated with action completion and is updated based on both a prior belief (“I need to act to feel good”) and sensory evidence (“I cannot act”). We conclude by briefly discussing the therapeutic implications of this framework. In summary, this unified Bayesian computational framework for craving generalizes across addictive disorders, provides explanatory power for ostensibly conflicting empirical findings, and generates strong hypotheses for future empirical studies. The disambiguation of the computational components underlying domain-general craving using this framework will lead to a deeper understanding of, and effective treatment targets for, behavioral and drug addictions.

Can an algorithm become delusional? Evaluating ontological commitments and methodology of computational psychiatry

AbstractThe computational approach to psychiatric disorders, including delusions, promises explanation and treatment. Here, we argue that an information processing approach might be misleading to understand psychopathology and requires further refinement. We explore the claim of computational psychiatry being a bridge between phenomenology and physiology while focussing on the ontological commitments and corresponding methodology computational psychiatry is based on. Interconnecting ontological claims and methodological practices, the paper illustrates the structure of theory-building and testing in computational psychiatry.First, we will explain the ontological commitments computational psychiatry is grounded in, theBayesian Brain hypothesis(BBH) of unconscious inference, paired with normative deontic approaches applied to gauge psychopathology. We then turn to the steps taken in empirical paradigms, from definitions, which are used as starting points, to the operationalisation and isolation of cognitive processes and hypothesis testing based on algorithmic models, to consecutive interpretations regarding the aetiology of psychiatric disorders. We outline how experimental paradigms in computational psychiatry are specifically designed to confirm aberrations in assumed inferential processes, which are thought of as being the underlyingcore invariant features.We will illustrate a gap between the ontological commitments of computational psychiatry and the operationalisation and testing of the cognition assumed to be relevant for psychopathology. This conceptual gap is of utmost importance when designing computational paradigms and may impede a crisp understanding of the approach. Lastly, in evaluating the conceptual gap, it becomes apparent that the information processing formalism used in computational psychiatry is still grounded in rational cognitive psychology.

Separating Uncertainty from Surprise in Auditory Processing with Neurocomputational Models: Implications for Music Perception.

Advances in neuroscience have led to the emergence of two complementary theories of neural information processing. First, the “Bayesian brain hypothesis” proposes that the brain actively predicts and represents incoming sensory information as probabilities that are updated in a near-optimal

Could Computational Approaches Challenge How we Understand and Tackle Mental Disorders Stigma?

Most psychiatric disorders (between 50% and 90%) remain untreated although effective treatments exist (Kohn et al., 2004). Stigma is frequent toward patients and mental health professionals (Rüsch et al., 2005). It impairs access to care especially during the early signs, amplifying the treatment gap, mainly due to a shortage of resources. How the brain makes prior internal models of the world could explain why we all stigmatize. The Bayesian brain hypothesis describes the optimal combination of priors, coming from our evolution, our memory and our perceptions, to make decisions (Parr et al., 2018). This can be intuitively understood by visual illusions, but also medical categorical diagnostic reasoning (Medow, 2011). These priors allow to simplify the world and categorize people (Sherman et al., 1998). Beyond cognitive modelling, Computational approaches also led to the development of several electronic devices, from apps to social robots, (Gargot et al., 2021). Patients reported that it was easier to self-disclose toward an avatar, that cannot judge, than toward a therapist (Hang and Gratch, 2010). These technologies could improve early, scalable as well as efficient access to care (Andersson et al., 2019). Empathy is the best strategy to tackle stigma. What is it exactly? Good therapist-patient synchronization of brains (Czeszumski et al., 2020) and bodies (Ramseyer et Tschacher, 2011) could foster better communication and thus empathy. There is an urgent need to promote research in empathy (Belzung, 2017), non-specific factors in psychotherapy mechanisms (Miller et Moyers, 2021) and psychotherapy online training resources (Gargot et al., 2020).DisclosureThomas Gargot was paid by the French ministry of research (Doctoral school), French ministry of health (médaille d’argent, CCA-AHU) for a PhD and CCA position. During his PhD, he prepared several scientific presentations with Dr Asselborn. This work led

Bayesian Brains and the Rényi Divergence.

Under the Bayesian brain hypothesis, behavioral variations can be attributed to different priors over generative model parameters. This provides a formal explanation for why individuals exhibit inconsistent behavioral preferences when confronted with similar choices. For example, greedy preferences are a consequence of confident (or precise) beliefs over certain outcomes. Here, we offer an alternative account of behavioral variability using Rényi divergences and their associated variational bounds. Rényi bounds are analogous to the variational free energy (or evidence lower bound) and can be derived under the same assumptions. Importantly, these bounds provide a formal way to establish behavioral differences through an α parameter, given fixed priors. This rests on changes in α that alter the bound (on a continuous scale), inducing different posterior estimates and consequent variations in behavior. Thus, it looks as if individuals have different priors and have reached different conclusions. More specifically, α→0+ optimization constrains the variational posterior to be positive whenever the true posterior is positive. This leads to mass-covering variational estimates and increased variability in choice behavior. Furthermore, α→+∞ optimization constrains the variational posterior to be zero whenever the true posterior is zero. This leads to mass-seeking variational posteriors and greedy preferences. We exemplify this formulation through simulations of the multiarmed bandit task. We note that these α parameterizations may be especially relevant (i.e., shape preferences) when the true posterior is not in the same family of distributions as the assumed (simpler) approximate density, which may be the case in many real-world scenarios. The ensuing departure from vanilla variational inference provides a potentially useful explanation for differences in behavioral preferences of biological (or artificial) agents under the assumption that the brain performs variational Bayesian inference.

A Predictive Processing Account of Card Sorting: Fast Proactive and Reactive Frontoparietal Cortical Dynamics during Inference and Learning of Perceptual Categories.

For decades, a common assumption in cognitive neuroscience has been that prefrontal executive control is mainly engaged during target detection [Posner, M. I., & Petersen, S. E. The attention system of the human brain. Annual Review of Neuroscience, 13, 25-42, 1990]. More recently, predictive processing theories of frontal function under the Bayesian brain hypothesis emphasize a key role of proactive control for anticipatory action selection (i.e., planning as active inference). Here, we review evidence of fast and widespread EEG and magnetoencephalographic fronto-temporo-parietal cortical activations elicited by feedback cues and target cards in the Wisconsin Card Sorting Test. This evidence is best interpreted when considering negative and positive feedback as predictive cues (i.e., sensory outcomes) for proactively updating beliefs about unknown perceptual categories. Such predictive cues inform posterior beliefs about high-level hidden categories governing subsequent response selection at target onset. Quite remarkably, these new views concur with Don Stuss' early findings concerning two broad classes of P300 cortical responses evoked by feedback cues and target cards in a computerized Wisconsin Card Sorting Test analogue. Stuss' discussion of those P300 responses-in terms of the resolution of uncertainty about response (policy) selection as well as the participants' expectancies for future perceptual or motor activities and their timing-was prescient of current predictive processing and active (Bayesian) inference theories. From these new premises, a domain-general frontoparietal cortical network is rapidly engaged during two temporarily distinct stages of inference and learning of perceptual categories that underwrite goal-directed card sorting behavior, and they each engage prefrontal executive functions in fundamentally distinct ways.

Neuronal Sequence Models for Bayesian Online Inference.

Various imaging and electrophysiological studies in a number of different species and brain regions have revealed that neuronal dynamics associated with diverse behavioral patterns and cognitive tasks take on a sequence-like structure, even when encoding stationary concepts. These neuronal sequences are characterized by robust and reproducible spatiotemporal activation patterns. This suggests that the role of neuronal sequences may be much more fundamental for brain function than is commonly believed. Furthermore, the idea that the brain is not simply a passive observer but an active predictor of its sensory input, is supported by an enormous amount of evidence in fields as diverse as human ethology and physiology, besides neuroscience. Hence, a central aspect of this review is to illustrate how neuronal sequences can be understood as critical for probabilistic predictive information processing, and what dynamical principles can be used as generators of neuronal sequences. Moreover, since different lines of evidence from neuroscience and computational modeling suggest that the brain is organized in a functional hierarchy of time scales, we will also review how models based on sequence-generating principles can be embedded in such a hierarchy, to form a generative model for recognition and prediction of sensory input. We shortly introduce the Bayesian brain hypothesis as a prominent mathematical description of how online, i.e., fast, recognition, and predictions may be computed by the brain. Finally, we briefly discuss some recent advances in machine learning, where spatiotemporally structured methods (akin to neuronal sequences) and hierarchical networks have independently been developed for a wide range of tasks. We conclude that the investigation of specific dynamical and structural principles of sequential brain activity not only helps us understand how the brain processes information and generates predictions, but also informs us about neuroscientific principles potentially useful for designing more efficient artificial neuronal networks for machine learning tasks.

Certainty in ascending sensory signals – The unexplored driver of analgesic placebo response

Previous frameworks have failed to adequately explain the observed correlation between within-subject variability in pain reporting and analgesic placebo response. These relationships have been observed in both clinical and experimental setups. Within-subject variability of clinical pain scores is traditionally assessed based on daily pain diaries collected during the pre-intervention stage. Experimental variability can be assessed by the Focused Analgesia Selection Test (FAST), which calculates the relationship between noxious stimuli administrated at various intensities and pain reports. The variability, either clinical or experimental, has been shown to predict the placebo response. In explaining the placebo response, Bayesian Brain Hypothesis (BBH) posits that pain perception (posterior), is composed of certainty (precision) of expectations (priors due to belief or conditioning) and incoming sensory information (likelihood), with the bulk of research focused on the precision of priors. Virtually all placebo analgesia research has focused on the priors and their certainty, rather than on the certainty of the likelihood, mainly because it cannot be assessed directly. We propose that the within-subject variability, as encapsulated by the FAST, is a proxy for certainty in (or, precision of) ascending sensory signals, and our results suggest that it could not only be assessed, but also manipulated. If true, our hypothesis will facilitate new lines of research and could potentially promote precision analgesic medicine by use of variability of pain scores as a diagnostic method to identify pain patients who will benefit from specific treatments.

Thinking through prior bodies: autonomic uncertainty and interoceptive self-inference.

The Bayesian brain hypothesis, as formalized by the free-energy principle, is ascendant in cognitive science. But, how does the Bayesian brain obtain prior beliefs? Veissière and colleagues argue that sociocultural interaction is one important source. We offer a complementary model in which "interoceptive self-inference" guides the estimation of expected uncertainty both in ourselves and in our social conspecifics.

Generative models as parsimonious descriptions of sensorimotor loops.

The Bayesian brain hypothesis, predictive processing, and variational free energy minimisation are typically used to describe perceptual processes based on accurate generative models of the world. However, generative models need not be veridical representations of the environment. We suggest that they can (and should) be used to describe sensorimotor relationships relevant for behaviour rather than precise accounts of the world.

A Review of Auditory Prediction and Its Potential Role in Tinnitus Perception.

The precise mechanisms underlying tinnitus perception and distress are still not fully understood. A recent proposition is that auditory prediction errors and related memory representations may play a role in driving tinnitus perception. It is of interest to further explore this. To obtain a comprehensive narrative synthesis of current research in relation to auditory prediction and its potential role in tinnitus perception and severity. A narrative review methodological framework was followed. The key words Prediction Auditory, Memory Prediction Auditory, Tinnitus AND Memory, Tinnitus AND Prediction in Article Title, Abstract, and Keywords were extensively searched on four databases: PubMed, Scopus, SpringerLink, and PsychINFO. All study types were selected from 2000-2016 (end of 2016) and had the following exclusion criteria applied: minimum age of participants <18, nonhuman participants, and article not available in English. Reference lists of articles were reviewed to identify any further relevant studies. Articles were short listed based on title relevance. After reading the abstracts and with consensus made between coauthors, a total of 114 studies were selected for charting data. The hierarchical predictive coding model based on the Bayesian brain hypothesis, attentional modulation and top-down feedback serves as the fundamental framework in current literature for how auditory prediction may occur. Predictions are integral to speech and music processing, as well as in sequential processing and identification of auditory objects during auditory streaming. Although deviant responses are observable from middle latency time ranges, the mismatch negativity (MMN) waveform is the most commonly studied electrophysiological index of auditory irregularity detection. However, limitations may apply when interpreting findings because of the debatable origin of the MMN and its restricted ability to model real-life, more complex auditory phenomenon. Cortical oscillatory band activity may act as neurophysiological substrates for auditory prediction. Tinnitus has been modeled as an auditory object which may demonstrate incomplete processing during auditory scene analysis resulting in tinnitus salience and therefore difficulty in habituation. Within the electrophysiological domain, there is currently mixed evidence regarding oscillatory band changes in tinnitus. There are theoretical proposals for a relationship between prediction error and tinnitus but few published empirical studies.

Constructing Hierarchical Bayesian Networks With Pooling

Inspired by the Bayesian brain hypothesis and deep learning, we develop a Bayesian autoencoder, a method of constructing recognition systems using a Bayesian network. We construct hierarchical Bayesian networks based on feature extraction and implement pooling to achieve invariance within a Bayesian network framework. The constructed networks propagate information bidirectionally between layers. We expect they will be able to achieve brain-like recognition using local features and global information such as their environments.

A Goal-Directed Bayesian Framework for Categorization.

Categorization is a fundamental ability for efficient behavioral control. It allows organisms to remember the correct responses to categorical cues and not for every stimulus encountered (hence eluding computational cost or complexity), and to generalize appropriate responses to novel stimuli dependant on category assignment. Assuming the brain performs Bayesian inference, based on a generative model of the external world and future goals, we propose a computational model of categorization in which important properties emerge. These properties comprise the ability to infer latent causes of sensory experience, a hierarchical organization of latent causes, and an explicit inclusion of context and action representations. Crucially, these aspects derive from considering the environmental statistics that are relevant to achieve goals, and from the fundamental Bayesian principle that any generative model should be preferred over alternative models based on an accuracy-complexity trade-off. Our account is a step toward elucidating computational principles of categorization and its role within the Bayesian brain hypothesis.