Predictive coding

Normality as a concept in phenomenology refers neither to a statistical or objectively measure average nor to what counts as normal within a specific time, culture, or society. Nor is it related to notions that equate normality with naturalness or natality. Normality, approached from a first-person perspective is characterized as a mode of experience, i.e., it is about what it is like to experience normally. Normality is comprised of two aspects, concordance and optimality. A normal course of experience must be concordant regarding prior experiences and optimal in relation to its contents, on the levels of individual and intersubjective experience. While normality as concordance is the condition of coherent experience, normality as optimality refers to the intentional aims of perception or action. Phenomenologically, normality is thus not an empirical, but a transcendental-philosophical, concept as it refers to the (general or genetic) conditions of experience. A minimal normality is regarded as necessary for every stable object-perception, smooth practical engagement with the environment, and the givenness of an intersubjectively shared world. The concept of normality in phenomenology is thus directly related to notions of objectivity (reality), and normativity. Concordance thereby represents the descriptive aspect of normality, while optimality introduces a normative dimension to normality.

Digital video technology has witnessed many important improvements in the last decade. Increasing video quality to be high resolutions is one of the most important developments in the field of video processing. Modern video sequences with higher resolutions led to larger consuming of storage capacity and transmission capability. Video compression became essential concern to avoid the impact of dealing with this huge amount of data. Devising efficient coding techniques came with high complexity and computation time. The High Efficiency Video Coding (HEVC) is one of the most popular encoding standards. It was designed for high-resolution video sequences. However, HEVC suffers from high encoding complexity. This paper proposes a fast intra prediction technique by reducing the number of comparisons in RDO process. In HEVC intra prediction, there are many candidates that may contain similar contents to the neighboring blocks. The content information and the direction of the most similar ...

A typical expository strategy for the predictive processing account begins with perception and then extends to other cognitive domains, such as action or non-human animal cognition. Because this standard, perception-first expository strategy begins at the end of an evolutionary process, it may introduce both diachronic and synchronic distortions into the overall account. As far as the diachronic distortion is concerned, because the perception-first strategy presupposes a highly decoupled cognitive architecture, it invites us to project this architecture onto the coupled cognitive systems of our evolutionary ancestors and other non-human organisms. Turning to the synchronic distortion, as the perception-first strategy is extended to conative states, it invites a cognitivism that seeks to reconstrue motivational states as a species of belief. In this paper, I explore what a predictive processing account would look like if it began at the beginning, with the highly coupled representations carried by our evolutionary ancestors and other non-human organisms. I conclude by exploring the proposal’s implications for a non-cognitivist, predictive processing account of action.

The amplitude of the auditory N1 component of the event‐related potential (ERP) is typically suppressed when a sound is accompanied by visual anticipatory information that reliably predicts the timing and identity of the sound. While this visually induced suppression of the auditory N1 is considered an early electrophysiological marker of fulfilled prediction, it is not yet fully understood whether this internal predictive coding mechanism is primarily driven by the temporal characteristics, or by the identity features of the anticipated sound. The current study examined the impact of temporal and identity predictability on suppression of the auditory N1 by visual anticipatory motion with an ecologically valid audiovisual event (a video of a handclap). Predictability of auditory timing and identity was manipulated in three different conditions in which sounds were either played in isolation, or in conjunction with a video that either reliably predicted the timing of the sound, the i...

Autism spectrum disorder is a pervasive neurodevelopmental disorder that has been linked to a range of perceptual processing alterations, including hypo- and hyperresponsiveness to sensory stimulation. A recently proposed theory that attempts to account for these symptoms, states that autistic individuals have a decreased ability to anticipate upcoming sensory stimulation due to overly precise internal prediction models. Here, we tested this hypothesis by comparing the electrophysiological markers of prediction errors in auditory prediction by vision between a group of autistic individuals and a group of age-matched individuals with typical development. Between-group differences in prediction error signaling were assessed by comparing event-related potentials evoked by unexpected auditory omissions in a sequence of audiovisual recordings of a handclap in which the visual motion reliably predicted the onset and content of the sound. Unexpected auditory omissions induced an increased ...

The amplitude of the auditory N1 component of the event‐related potential (ERP) is typically attenuated for self‐initiated sounds, compared to sounds with identical acoustic and temporal features that are triggered externally. This effect has been ascribed to internal forward models predicting the sensory consequences of one's own motor actions. The predictive coding account of autistic symptomatology states that individuals with autism spectrum disorder (ASD) have difficulties anticipating upcoming sensory stimulation due to a decreased ability to infer the probabilistic structure of their environment. Without precise internal forward prediction models to rely on, perception in ASD could be less affected by prior expectations and more driven by sensory input. Following this reasoning, one would expect diminished attenuation of the auditory N1 due to self‐initiation in individuals with ASD. Here, we tested this hypothesis by comparing the neural response to self‐ versus external...

The mind, within the framework of the Dynamic Model of the Mind, is viewed as a holistic and unified representation of reality. This dynamic model continuously updates in response to new inputs, including emotions, which play a central role in guiding decisions, enhancing social interaction, and reshaping an individual's perceptions of their environment. This paper highlights emotions as a pivotal dynamic element ensuring the model's alignment with reality, with a focus on the influence of modern technology in reshaping this dynamic.

The concept of the brain as a prediction machine has enjoyed a resurgence in the context of the Bayesian brain and predictive coding approaches within cognitive science. To date, this perspective has been applied primarily to exteroceptive perception (e.g., vision, audition), and action. Here, I describe a predictive, inferential perspective on interoception: 'interoceptive inference' conceives of subjective feeling states (emotions) as arising from actively-inferred generative (predictive) models of the causes of interoceptive afferents. The model generalizes 'appraisal' theories that view emotions as emerging from cognitive evaluations of physiological changes, and it sheds new light on the neurocognitive mechanisms that underlie the experience of body ownership and conscious selfhood in health and in neuropsychiatric illness.

Science and art have long recognized that perceptual experience depends on the involvement of the experiencer. In art history, this idea is captured by Ernst Gombrich’s ‘beholder’s share’. In neuroscience, it traces to Helmholtz’s concept of ‘perception as inference’, which is enjoying renewed prominence in the guise of ‘prediction error minimization’ (PEM) or the ‘Bayesian brain’. The shared idea is that our perceptual experience – whether of the world, of ourselves, or of an artwork – depends on the active ‘top-down’ interpretation of sensory input. Perception becomes a generative act, in which perceptual, cognitive, affective, and sociocultural expectations conspire to shape the brain’s ‘best guess’ of the causes of sensory signals. In this article, I explore the parallels between the Bayesian brain and the beholders’ share, illustrated, somewhat informally, with examples from Impressionist, Expressionist, and Cubist art. By connecting phenomenological insights from these traditi...

The concept of the brain as a prediction machine has enjoyed a resurgence in the context of the Bayesian brain and predictive coding approaches within cognitive science. To date, this perspective has been applied primarily to exteroceptive perception (e.g., vision, audition), and action. Here, I describe a predictive, inferential perspective on interoception: 'interoceptive inference' conceives of subjective feeling states (emotions) as arising from actively-inferred generative (predictive) models of the causes of interoceptive afferents. The model generalizes 'appraisal' theories that view emotions as emerging from cognitive evaluations of physiological changes, and it sheds new light on the neurocognitive mechanisms that underlie the experience of body ownership and conscious selfhood in health and in neuropsychiatric illness.

Throughout his career Andy Clark has shaped how scientists and philosophers think about the role of representation in action, perception, and cognition. In the latest iteration of this debate he has foregrounded the influential perspective of ‘predictive processing’, which sees perception as a process of action-oriented ‘best guessing’ (inference) about the causes of noisy and ambiguous sensory signals, and which involves the brain inducing ‘generative’ models of how hidden causes mediate the effects of actions on sensory signals. Here, I will develop this position in the context of interoception (the sense of the body from within) and physiological regulation. A key idea here, which recalls 20th century cybernetic theory, is that interoceptive inference is targeted towards maintaining physiological homeostasis rather than inducing complete and accurate internal models of an external state-of-affairs. I explore how this perspective helps connect control-oriented interoceptive infere...

The human mind is an intricate system, constantly evolving as it interacts with the external world and internal states. This paper presents the 'Dynamic Model of the Mind,' a probabilistic framework for understanding perception as a dynamic, ever-changing process. The model builds upon the concept that the mind continuously updates its internal representation of the world based on sensory inputs, emotions, memories, and cultural contexts. This approach challenges traditional, static views of cognition and introduces a more fluid understanding of mental processes.

The development of an advanced computational analysis of unsteady aerodynamic loads on isolated helicopter rotors in forward flight is described. The primary technical focus of the development was the implementation of a freely distorting filamentary wake model composed of curved vortex elements laid out along contours of constant vortex sheet strength in the wake. This model captures the wake generated by the full span of each rotor blade and makes possible a unified treatment of the shed and trailed vorticity in the wake. This wake model was coupled to a modal analysis of the rotor blade dynamics and a vortex lattice treatment of the aerodynamic loads to produce a comprehensive model for rotor performance and air loads in forward flight dubbed RotorCRAFT (Computation of Rotor Aerodynamics in Forward Flight). The technical background on the major components of this analysis are discussed and the correlation of predictions of performance, trim, and unsteady air loads with experiment...

A conventional stereoscopic system with a single left and right image pair that should be transmitted simultaneously is twice of the raw data in a monoscopic image system, so it needs an effective coding algorithm. Disparity compensation is a good methodology for stereo image compression that uses one view as a reference to predict another, and the difference data is coded. Due to the application of reference image, it also needs the effective coding. Therefore, we have proposed a new efficient coding technique, called Quadrant Vector Quantization (QVQ) based on wavelet transform.

A exceptionally large number of excellent commentary proposals inspired a special research topic for further discussion of this target article's subject matter, edited by Axel Cleeremans and Shimon Edelman in Frontiers in Theoretical and Philosophical Psychology. This discussion has a preface by Cleeremans and Edelman and 25 commentaries and includes a separate rejoinder from Andy Clark. See:

This paper uses several famous problems from philosophy of science and philosophical psychology—underdetermination of theory by evidence, Nelson Goodman’s new problem of induction, theory-ladenness of observation, and “Kripkenstein’s” rule-following paradox—to show that it is empirically impossible to reliably interpret which functions, concepts, or rules a large-language model (LLM) AI has learned, and thus, that reliably aligning AI behavior with human values is provably impossible. Sections 1 and 2 show that because of how complex LLMs are, researchers must interpret the functions they have learned largely in terms of empirical observations of their behavior. Sections 3-6 then show that for every “aligned” function that might appear to be confirmed by observations of the behavior of an LLM or LLM subcomponent, there is always an infinitely larger number of “misaligned”, arbitrarily time-limited functions equally confirmed by the same data. Section 7 shows that, from an empirical perspective, we can thus never reliably infer that an LLM or subcomponent of one has learned any particular function at all—either ‘aligned’ or ‘misaligned’ ones—before any of an uncountably large number of unpredictable future conditions obtain. Finally, Section 8 concludes that the probability of LLM “misalignment” is—at every point in time, given any arbitrarily large body of empirical evidence—always vastly larger than the probability of “alignment.”

Editor's Note: These short, critical reviews of recent papers in the Journal, written exclusively by graduate students or postdoctoral fellows, are intended to summarize the important findings of the paper and provide additional insight and commentary. For more information on the format and purpose of the Journal Club, please see http://www.jneurosci.org/misc/ifa_features.shtml.

This paper challenges long-held beliefs about consciousness and selfhood, proposing that these phenomena-far from being intrinsic or metaphysical-are evolved survival tools. We argue that emotions and the sense of self function as adaptive mechanisms shaped to reduce uncertainty and optimize behavior in complex environments. Leveraging Karl Friston's free-energy principle and Mark Solms' affective neuroscience, we present a new framework: selfhood, much like emotions, is a construct of the brain's predictive coding system. It emerges not as an independent entity but as a dynamic narrative created by the brain to maintain coherence and minimize prediction errors. The self, traditionally seen as a discrete and central entity, may instead be an illusion generated by the brain's ego-driven processes. We also explore the implications of this functionalist view for artificial intelligence, suggesting that intelligent systems can operate effectively without subjective experiences like qualia. By reframing emotions and selfhood as adaptive artifacts, this paper invites empirical and philosophical inquiry into the true role of consciousness and self-perception, challenging foundational ideas of individual discreteness and exploring the mind's role in shaping reality. The Pragmatic Role of Consciousness and Emotion

This paper delves into one of the most fundamental questions in cognitive science and philosophy of mind: How does language, the very tool we use to understand and describe our consciousness, distort our perception of reality? Drawing on insights from neuroscience, artificial intelligence, and philosophy, this work explores how symbolic representation—particularly through language—creates cognitive artifacts like selfhood, emotions, and qualia. Building on the predictive coding framework, the paper argues that what we commonly experience as stable realities—such as the self and emotional experiences—are not metaphysical truths, but functional constructs developed by the brain to minimize uncertainty and optimize survival. By compressing the complexity of individual emotions and experiences into simplified symbols, language creates an illusion of shared reality, inflating the ego and misguiding our understanding of consciousness. The paper appeals to scholars across disciplines—from neuroscience and AI to psychology and philosophy—by challenging the widely accepted view of selfhood as an intrinsic part of human identity. Instead, it presents a compelling argument that emotions and consciousness are cognitive tools, shaped pragmatically to serve adaptive purposes. The text invites a deeper exploration into how these cognitive artifacts function both in biological systems and in artificial intelligence, encouraging interdisciplinary collaboration to further investigate the illusionary nature of subjective experiences. Ultimately, this paper is a thought-provoking contribution to ongoing debates about the nature of consciousness and invites readers from diverse fields to rethink long-held assumptions about the role of language and symbols in shaping our cognitive realities.

This paper presents a result on information preserving compression of NMR images for the archiving purpose. Roth Lynch-Davisson coding and linear predictive coding have been studied. For NMR images of 256 x 256 X 12 resolution, the Lynch-Davisson coding with a block size of 64 as applied to prediction error sequences in the Gray code bit planes of each image gave an average compression ratio of 2.3:1 for 14 testing images. The predictive coding with a third order linear predictor and the HuEman encoding of the prediction error gave an average compression ratio of 3.1:1 for 54 images under test, while the maximum compression ratio achieved was 3.&l. This result is one step further toward the improvement, albeit small, of the information preserving image compression for medical applications.

The N400 event-related component has been widely used to investigate the neural mechanisms underlying real-time language comprehension. However, despite decades of research, there is still no unifying theory that can explain both its temporal dynamics and functional properties. In this work, we show that predictive coding – a biologically plausible algorithm for approximating Bayesian inference – offers a promising framework for characterizing the N400. Using an implemented predictive coding computational model, we demonstrate how the N400 can be formalized as the lexico-semantic prediction error produced as the brain infers meaning from linguistic form of incoming words. We show that the magnitude of lexico-semantic prediction error mirrors the functional sensitivity of the N400 to various lexical variables, priming, contextual effects, as well as their higher-order interactions. We further show that the dynamics of the predictive coding algorithm provide a natural explanation for ...

In this article, we review a recent paper by Stevenson et al. (J Neurosci 34: 691-697, 2014). This paper illustrates the need to present different forms of stimuli in order to characterize the perceptual abilities of people with autism spectrum disorder (ASD). Furthermore, we will discuss their behavioral results and offer an opposing viewpoint to the suggested neuronal drivers of ASD.

Aging is accompanied by unisensory decline; but to compensate for this, two complementary strategies are potentially relied upon increasingly: first, older adults integrate more information from different sensory organs. Second, according to predictive coding (PC) we form ‘templates’ (internal models or ‘priors’) of the environment through our experiences. It is through increased life experience that older adults may rely more on these templates compared to younger adults. Multisensory integration and predictive coding would be effective strategies for the perception of near-threshold stimuli, but they come at the cost of integrating irrelevant information. Their role can be studied in multisensory illusions because these require the integration of different sensory information, as well as an internal model of the world that can take precedence over sensory input. Here, we elicited a classic multisensory illusion, the sound-induced flash illusion, in younger (mean: 27 yrs) and older...

Recommender systems (RSs) often focus on learning users' long-term preferences, while the sequential pattern of behavior is ignored. On the other hand, sequential RSs try to predict the next action by exploring relations between items in a user's last interactions but do not consider the general preference. Recently, the performance of RSs has increased by unifying these two types of paradigms. However, existing methods still have two limitations. First, the user's behavior uncertainty impedes precise learning of preferences. Second, being unable to understand the semantics of items makes the effect of the same item considered in the same way. These limitations jointly prevent RS from learning multifaceted preferences to capture the actual intentions of users. Existing methods have not properly addressed these problems since they ignore context-aware interactions between the user and item in terms of the links between the user and item attributes and sequential user actions over time. To address these challenges, this paper proposes a novel model, called the Dynamic Intention-Aware Recommendation with attention-based context-aware item attributes modeling (DIARec), which is capable of determining users' preferences based on their goal intention, taking into account the influence of various item features on user decision-making in their current context. Specifically, to model users' dynamic intentions, we introduce a dynamic intent-aware module to represent the hierarchical relations between items and their attributes in a given session. Experiments on benchmark datasets indicate that the proposed model DIARec outperforms other stateof-the-art methods.

The amplitude of the auditory N1 component of the event‐related potential (ERP) is typically suppressed when a sound is accompanied by visual anticipatory information that reliably predicts the timing and identity of the sound. While this visually induced suppression of the auditory N1 is considered an early electrophysiological marker of fulfilled prediction, it is not yet fully understood whether this internal predictive coding mechanism is primarily driven by the temporal characteristics, or by the identity features of the anticipated sound. The current study examined the impact of temporal and identity predictability on suppression of the auditory N1 by visual anticipatory motion with an ecologically valid audiovisual event (a video of a handclap). Predictability of auditory timing and identity was manipulated in three different conditions in which sounds were either played in isolation, or in conjunction with a video that either reliably predicted the timing of the sound, the i...

This paper describes a new method of segmentation of time-varying image sequences whose goal is object-oriented image coding. The segmentation represents a partition of each frame of the sequence into a set of regions which are homogeneous with regard to motion criterion. The region borders correspond to spatial contours of objects in the frame. Each spatio-temporal region is characterized by its temporal component, which is a model-dependent vector of motion parameters, and a structural component representing the polygonal approximation of the spatial contour of the region. The construction of spatio-temporal segmentation includes two phases: the initialization step and temporal tracking. The initialization step is based on the spatial segmentation of the first frame of the sequence. Then homogeneous spatial regions are merged through motion estimation in accordance with a motion-based criterion. The temporal tracking consists of the projection of the segmentation along the time axis, and its adjustment. Special attention is paid to the processing of occlusions. A predictive coding scheme is proposed which is based on the temporal coherence of the segmentation. This scheme is promising for a low bit-rate image compression. The results for teleconference and TV sequences show the high visual quality of reconstructed only by prediction images. Moreover, the bit-rates for motion coding are very low: from 0.002 to 0.007 bit/pixel for teleconference sequence and from 0.004 to 0.021 bit/pixel for complex TV sequence. A scheme for encoding of the structural information is proposed which requires 0.083-0.17 bit per pixel depending on the content of the sequence.

Tracking the positions of objects in local space is a core function of animal brains. We do not yet fully understand how it is done with limited neural resources. The challenges of spatial cognition are discussed under the criteria: (a) scaling of computational costs; (b) feature binding; (c) precise calculation of spatial displacements; (d) fast learning of invariant patterns; and (e) exploiting the strong Bayesian prior of object constancy. The predominant current models of spatial cognition are Hierarchical Bayesian models of vision, and Deep Neural Nets. These are typically fully distributed models, which compute using direct communication links between a set of modular knowledge sources, and no other essential components. This distributed nature leads to difficulties with the criteria (a)-(e). I discuss an alternative model of spatial cognition, which uses a single central position aggregator to store estimated locations of each object or feature and applies constraints on locations in an iterative cycle between the aggregator and the knowledge sources. This model has advantages in addressing the criteria (a)-(e). If there is an aggregator in mammalian brains, there are reasons to believe that it is in the thalamus. I outline a possible neural realisation of the ensuing aggregator function in the thalamus.

More than 60 zoonoses are linked to small mammals, including some of the most devastating pathogens in human history. Millions of museum-archived tissues are available to understand natural history of those pathogens. Our goal was to maximize the value of museum collections for pathogen-based research by using targeted sequence capture. We generated a probe panel that includes 39,916 80-bp RNA probes targeting 32 pathogen groups, including bacteria, helminths, fungi, and protozoans. Laboratory-generated, mock-control samples showed that we are capable of enriching targeted loci from pathogen DNA 2,882-6,746-fold. We identified bacterial species in museum-archived samples, including Bartonella, a known human zoonosis. These results showed that probe-based enrichment of pathogens is a highly customizable and efficient method for identifying pathogens from museum-archived tissues. GenBank accession no. Anaplasma Bacteria 368 49 57 Anaplasma phagocytophilum GCF000013125 Apicomplexa Eukaryote 3,219 49 64 Plasmodium falciparum GCA000002765 Bacillus cereus group* Bacteria 833 98 134 Bacillus anthracis GCF000008165 Bartonella Bacteria 1,812 49 31 Bartonella bacilliformis GCF000015445 Borrelia Bacteria 688 49 16 Borreliella burgdorferi GCF000502155 Burkholderia Bacteria 683 49 39 Burkholderia mallei GCF000011705 Campylobacter Bacteria 2,194 49 33 Campylobacter jejuni GCF000009085 Cestoda Eukaryote 907 49 18 Taenia multiceps GCA001923025 Chlamydia Bacteria 830 49 15 Chlamydia trachomatis GCF000008725 Coxiella Bacteria 144 49 70 Coxiella burnetii GCF000007765 Ehrlichia Bacteria 235 49 7 Ehrlichia canis GCF000012565 Eurotiales Eukaryote 4,097 49 158 Talaromyces marneffei GCF000001985 Francisella Bacteria 470 49 14 Francisella tularensis GCF000008985 Hexamitidae Eukaryote 782 49 19 Giardia intestinalis GCA000002435 Kinetoplastea Eukaryote 2,917 49 49 Leishmania major GCF000002725 Leptospira Bacteria 2,517 49 69 Leptospira interrogans GCF000092565 Listeria Bacteria 765 49 23 Listeria monocytogenes GCF000196035 Mycobacterium Bacteria 2,463 49 86 Mycobacterium tuberculosis GCF000195955 Nematodes, clade I Eukaryote 357 49 13 Trichinella spiralis GCA000181795 Nematodes, clade III Eukaryote 1,494 49 25 Brugia malayi GCA000002995 Nematodes, clade IVa Eukaryote 252 49 7 Strongyloides stercoralis GCA000947215 Nematodes, clade IVb Eukaryote 1,487 43 34 Steinernema carpocapsae GCA000757645 Nematodes, clade V Eukaryote 3,242 48 47 Haemonchus contortus GCA007637855 Onygenales Eukaryote 1,973 49 38 Histoplasma capsulatum GCF000149585 Pasteurella Bacteria 615 49 11 Pasteurella multocida GCF000754275 Rickettsia Bacteria 394 49 37 Rickettsia rickettsii GCF001951015 Salmonella Bacteria 145 49 35 Salmonella enterica GCF001159405 Streptobacillus Bacteria 245 49 7 Streptobacillus moniliformis GCF000024565 Trematoda Eukaryote 924 49 18 Schistosoma mansoni GCA000237925 Tremellales Eukaryote 1,999 49 26 Cryptococcus neoformans GCF000091045 Trypanosoma* Eukaryote 617 97 10 Trypanosoma cruzi GCF000209065 Yersinia Bacteria 225 49 22 Yersinia pestis GCF000009065 *Supplemented.

Dans cette étude, nous explorons les dynamiques probabilistes du jeu Monopoly à travers l'utilisation des chaînes de Markov, en nous concentrant sur l'impact du dé rapide introduit dans les versions récentes du jeu. L'objectif principal de cette analyse est de rendre ces concepts mathématiques accessibles à un public non-initié, en proposant une tentative de vulgarisation scientifique. À travers une approche pédagogique, nous montrons comment la théorie des chaînes de Markov permet de modéliser les déplacements sur le plateau, en tenant compte des règles du jeu et des différents types de dés. Ce travail s'efforce de simplifier des notions complexes tout en offrant une vision claire des probabilités à long terme dans différentes versions du jeu. En plus de fournir une analyse quantitative, cette étude vise à encourager une compréhension plus large de l'application des mathématiques à des phénomènes ludiques et à ouvrir la voie à d'autres réflexions sur le sujet.

According to the ideomotor theory, action may serve to produce desired sensory outcomes. Perception has been widely described in terms of sensory predictions arising due to top-down input from higher order cortical areas. Here, we demonstrate that the action intention results in reliable top-down predictions that modulate the auditory brain responses. We bring together several lines of research, including sensory attenuation, active oddball, and action-related omission studies: Together, the results suggest that the intention-based predictions modulate several steps in the sound processing hierarchy, from preattentive to evaluationrelated processes, also when controlling for additional prediction sources (i.e., sound regularity). We propose an integrative theoretical framework-the extended auditory event representation system (AERS), a model compatible with the ideomotor theory, theory of event coding, and predictive coding. Initially introduced to describe regularity-based auditory predictions, we argue that the extended AERS explains the effects of action intention on auditory processing while additionally allowing studying the differences and commonalities between intention-and regularity-based predictions-we thus believe that this framework could guide future research on action and perception.

O objetivo central deste artigo é demonstrar que existe uma forma de autoconsciência onde a quantidade de informação gerada por (q) (um conjunto de previsões de resultados sensoriais, sejam eles interoceptivos ou exteroceptivos, de ações hipotéticas) é uma função direta do grau de certeza que o indivíduo atribui a (q). Nesse

It has been suggested that the brain employs probabilistic generative models to optimally interpret sensory information. This hypothesis has been formalised in distinct frameworks, focusing on explaining separate phenomena. On one hand, predictive coding theory proposed how the probabilistic models can be learned by networks of neurons employing local synaptic plasticity. On the other hand, neural sampling theories have demonstrated how stochastic dynamics enable neural circuits to represent the posterior distributions of latent states of the environment. Here, we bring together these two lines of theoretic work by introducing Monte Carlo predictive coding (MCPC). We demonstrate that the integration of predictive coding with neural sampling results in a neural network that learns precise generative models using local computation and plasticity. The neural dynamics of MCPC infer the posterior distributions of the latent states in the presence of sensory inputs, and can generate likely inputs in their absence. Furthermore, MCPC captures the experimental observations on the variability of neural activity during perceptual tasks. By combining predictive coding and neural sampling, MCPC offers a unifying theory of cortical computation which can account for both sets of neural data that previously had been explained by these individual frameworks. .

This article explores the dynamic interplay between human and non-human entities, focusing on how embodied data representation is distributed. It examines how predictive coding, which utilizes preconceived knowledge, interacts with tangible experiences to shape our understanding of the world. Emphasizing this, I propose the concept of co-Ability as a deep underlying explanatory framework for understanding adaptive behaviors within a networked world. A nonverbal dialog between humans and a data-saturated environment is analyzed through an action-oriented perspective and the predictive coding framework in cognition, utilizing digital craft and rapid prototyping. This transformative approach augments human interaction with digital landscapes through tangible prototypes, bridging physical experience with abstract information, and identifying potential ways to conceptualize data materially. The article discusses the various aspects of connectivity among network agents and the evolving nature of these connections as they adapt to real-world conditions and dynamic shifts in data, highlighting that information exchange in an interconnected network is more than bilateral; it generates ripple effects that extend beyond immediate connections. These reciprocal exchanges simultaneously alter both the digital and analog domains, with data constantly bifurcating into multiple pathways and outcomes. A significant challenge addressed in this article is the question of how to frame information materially, inviting further exploration.

The paper attempts to show that Predictive Processing (PP), despite recent attempts by its proponents to ward off accusations that lead to skepticism (Clark 2016, 2019), is susceptible to undesirable skeptical consequences of a Kantian (rather than Cartesian) character. Specifically, I shall argue that Clark's version of PP is susceptible to a particularly Kantian version of skepticism in which the external world directly revealed by PP generative models is a phenomenal one in the Kantian sense: A world perceived and conceived as external, but at the same time essentially 'internal' in its categorial form, where this 'internality' only diverges from Kant in that it is a consequence of evolution. It will be suggested that these skeptical consequences can be avoided by articulating a more nuanced notion of the boundary between mind and world in PP, namely, one that differentiates an ontological from an epistemological understanding of the boundary between mind (generative model) and world. Moreover, it will be argued that in order to avoid Kantian skepticism, we must construe the very distinction between the phenomenal world and the world as it is in itself in non-metaphysical, pragmatic terms, as a framework condition for epistemically coordinating empirical inquiry within an ever-changing and unpredictable world. As a bonus, this view seems capable of accommodating the insights of autopoietic enactivism without buying into the latter's controversial 'transcendental idealist' organism-relative ontology.