Search Results

This chapter examines the history of the core ideas concerning neural prediction, presenting them in both a general and a more specific form. The general form is a vision of the brain (and especially the cortex) as fundamentally an engine of probabilistic prediction. The more specific form is a proposal (hierarchical predictive coding, or predictive processing) describing the possible shape and nature of that core process of multilevel probabilistic prediction. The chapter includes a historical overview of the development of these ideas. It also displays some of the evidence in favour of such approaches, and addresses some worries.

This chapter presents the regularity account of phenomenal unity (RPU). The basic idea of RPU is that when the brain tracks a regularity that is predictive of different features (or of different objects or events), there will be an experienced connection between those features (or the respective objects or events). We can then say that the regularity connects those features (or objects or events). According to RPU, unity comes in degrees, and in ordinary conscious experience we find a hierarchy of experienced wholes. This chapter provides a preliminary taxonomy of experienced wholes, with many examples. Drawing on formal concepts of the predictive processing framework, a formal description of possible computational underpinnings of experienced wholeness is given. Finally, a rigorous formulation of the mélange model (first proposed in chapter 4) is provided.

Cultural psychiatry is concerned with understanding the implications of human cultural diversity for psychopathology, illness experience, and intervention. The emerging paradigms of embodiment and enactment in cognitive science provide ways to approach this diversity in terms of variations in bodily and intersubjective experience, narrative practices, and discursive formations. This chapter will outline an approach to cultural neurophenomenology and psychopathology through metaphor theory, which examines the interplay of culturally shaped developmental processes of embodied experience and narrative practices structured by ideologies of personhood and social positioning. The new paradigm has broad implications for psychiatric theory, research, and practice, which will be illustrated with examples from the cross-cultural study of delusions.

Chapter 4 discusses how the immediacy of language processing provides a fundamental constraint on theories of language acquisition and evolution. Language happens in the here-and-now. Because memory is fleeting, new material will rapidly obliterate previous material, creating a Now-or-Never bottleneck. To successfully deal with the continual deluge of linguistic information, the brain must compress and recode its input into “chunks” as rapidly as possible. It must deploy all available information predictively to ensure that local linguistic ambiguities are dealt with Right-First-Time; once the original input is lost, there is no way to recover it. Similarly, language learning must also occur in the here-and-now. This implies that language acquisition involves learning how to process linguistic structure, rather than inducing a grammar. Incoming language is recoded incrementally into chunks of increasing granularity, from sounds to constructions, and beyond. Importantly, several key properties of language follow naturally from this perspective, including the local nature of linguistic dependencies, the quasi-regular nature of linguistic structure, multiple levels of linguistic representation, and duality of patterning (i.e., that meaningful units are composed of smaller elements).

This chapter provides an introduction to predictive processing (PP), with a focus on its relation to Bayesian inference and the more general framework provided by Karl Friston’s free-energy principle (FEP). Philosophical implications of PP and FEP are discussed. In particular, it is argued that PP posits genuine representations, and that FEP questions whether the conceptual distinction between action and perception is useful when it comes to understanding their neuro-computational underpinnings. Finally, it is argued that PP supports the view that the mind is to some extent secluded from the external world and that perception is indirect because it is based on genuinely inferential processes.

Over the last thirty years the comparator hypothesis has emerged as a prominent account of inner speech pathology. This chapter discusses a number of cognitive accounts broadly derived from this approach, highlighting the existence of two importantly distinct notions of inner speech in the literature; one as a prediction in the absence of sensory input, the other as an act with sensory consequences that are themselves predicted. Under earlier frameworks in which inner speech is described in the context of classic models of motor control, I argue that these two notions may be compatible, providing two routes to inner speech pathology. Under more recent accounts grounded in the architecture of Bayesian predictive processing, I argue that “active inference” approaches to action generation pose serious challenges to the plausibility of the latter notion of inner speech, while providing the former notion with rich explanatory possibilities for inner speech pathology.

The ubiquity of predictive processing in the brain suggests that it is functionally oriented toward the future. Mechanisms for predictive processing such as internal generative models can give rise to internally generated brain dynamics, thus permitting the brain to “detach,” at least partially, from the here and now of the current sensorimotor context. One example are internally generated sequences of neural activity in the rodent hippocampus, which can be produced and replayed in the absence of external cues, and have been linked to flexible decisions, planning, and memory functions. In this chapter the author considers the idea that other, more sophisticated kinds of detached cognition, such as counterfactual thinking and some forms of episodic simulation, might also be based on internally generated dynamics and use an internal model that originally supported predictive processing.

The introduction to this volume presents the approach of predictive processing, outlines its applicability to literary texts, and provides a sketch of the overall argument of the book. The predictions relevant for predictive processing are defined as not necessarily conscious. They unfold over a cascade of feedback loops from embodied states and movements through predictions about thoughts and larger, culturally embedded presuppositions. With this new definition, the treatment of prediction in Probability Designs extends beyond the approaches from genre and ideology. It not only links to recent developments in cognitive literary studies with embodied and historically, culturally situated cognition but also provides the flexibility of constructed, changeable predictions and probabilities.

Predictive processing (PP) is not a theory of consciousness. Hence, it is not obvious that PP should have any relevance to research on consciousness. A first promising possibility opens up if we consider the ambitious assumption that PP is a general theory of brain function. If the brain’s function is to minimize prediction error (just as the heart’s function is to pump blood), as Jakob Hohwy (2015) suggests, then it might well be that the computational processes underlying consciousness can usefully be described within the PP framework. This chapter focuses on (i) how PP accounts for attention, and what this suggests with regards to the relation between attention and consciousness (e.g., how volitional attention may change the contents of consciousness); (ii) furthermore, it is suggested that PP can provide a unifying perspective on some proposed functions and theories of consciousness (such as global workspace theory, attention schema theory, and integrated information theoy).

This chapter discusses Andy Clark’s recent explorations of Bayesian perceptual models and predictive processing. In the first part, the chapter discusses the predictive processing (PP) framework, explicating its relationship with hierarchical Bayesian models in theories of perception. In the second part, it examines the relationship between perception and action in the PP model. The overarching goal is twofold: first, to get clearer on the picture of mental activity that Clark is presenting, including what exactly is represented at the levels of the perception/action hierarchy, and the nature of the information processing it postulates; second, although the framework presented by Clark certainly has interesting novel features, some of his glosses on it are misleading. In particular, Clark’s interpretation of predictive processing as essentially a top-down, expectation-driven process, on which perception is aptly thought of as “controlled hallucination,” exaggerates the contrast with the traditional picture of perception as bottom-up and stimulus-driven. Additionally, despite the rhetoric, Clark’s PP model substantially preserves the traditional distinction between perception and action.

‘What role do art and neuroaesthetics play in disclosing mental models?’ asks Karl Friston in a (2013) review of a book on the interactions between the sciences and the humanities in Vienna around 1900. The chapter proposes some answers to Friston’s question, with particular respect to literary narrative and to the Bayesian probabilistic model of cognition that Friston himself developed. It suggests that the fantastic, defined as epistemic hesitation between the strange and the marvellous, constitutes a revealing defamiliarization of the ways in which our mind calibrates the probabilities of the real world in Bayesian inferences. According to the recent Bayesian approaches to the ways in which minds and bodies move in, perceive, and establish causal patterns for their natural and cultural environment, they develop and continuously update probabilistic models which give them a good enough grasp of these environments. In the fantastic, these probabilistic models are challenged and, arguably, made available to observation. On the example of Jacques Cazotte’s Le Diable amoureux (The Devil in Love, 1772/1776), this article shows how the analysis of narrative strategies in the fantastic can contribute to the ‘found science’ of artistic short-cuts that reveals underlying cognitive processes. It also proposes that the changes between different versions of the same narrative could serve as an experimental repository for the cognitive study of literature.

The chapter details how probability designs build the perception of coincidence and surprise by creating a ‘height of drop’ before plot events. It investigates more closely the role of the reader. The perception of the probability of a certain prediction is manipulated along a range of textual devices, which are discussed on the example of Jane Austen’s Pemberley scene in Pride and Prejudice. These manipulations of precision (that is, the perceived reliability of prediction errors), it is argued, also contribute to readers’ (illusory) sense of agency and their explorative mental moves. While readers follow the probability design, they are actively configuring the predictions of the narrative, which is related to phenomena like anomalous suspense and the paradox of tragedy.

Internal states of bodily arousal contribute to emotional feeling states and behaviors. This chapter details the influence of interoceptive processing on emotion and describes how deficits in interoceptive ability may underpin aberrant emotional processes characteristic of clinical conditions. The representation and control of bodily physiology (e.g. heart rate and blood pressure) and the encoding of emotional experience and behavior share neural substrates within forebrain regions coupled to ascending neuromodulatory systems. This functional architecture provides a basis for dynamic embodiment of emotion. This chapter will approach the relationship between interoception and emotion within the interoceptive predictive processing framework and describe how emotional states could be the product of interoceptive prediction error minimization.

Embodied predictive processing accounts place the visceral milieu, its homeostatic functioning, and our interoceptive awareness thereof on the center stage of self-awareness. Starting from the privileged status that homeostatic priors have within the cortical hierarchy of an organism whose main imperative is to maintain homeostasis, we focus on the mechanisms that underlie interoceptive precision and its impact on embodiment and cognition. Beyond their privileged status for ensuring the stability of organism, this chapter considers the psychological importance that interoceptive priors and interoceptive precision have for self-awareness and the grounding of a coherent self-model. In a manner analogous to the role that interoception plays for homeostasis, interoception at the psychological level seems to contribute to the stability of self-awareness. This psychological role of interoception is illustrated by a growing body of research that considers the antagonism but also the integration between exteroceptive and interoceptive models of the self.

In this chapter I further develop interaction theory and the concept of primary intersubjectivity by providing evidence for our ability to directly perceive intentions and emotions. Intentions and emotions can be understood at least in part as composed of perceivable patterns of contextualized embodied behaviors. I argue that perception is “smart” and in no need of inferential or simulational supplementation in most instances of social interaction. I consider that even some theory theorists have acknowledged the role of perception but not without giving up the idea of a subpersonal processing that amounts to an inferential mindreading. I also consider recent predictive processing accounts and argue for an embodied-enactive interpretation of such processes. Finally, I consider concerns about direct social perception raised by research in social psychology.

Many commentators on Clark’s writings on predictive processing have wondered how well the predictive processing model actually fits with embodied and extended cognition. The former seems to imply that cognition is secluded from the environment, while the latter implies that cognition is in and of the environment. This chapter argues that a reconciliation with embodied and extended cognition is possible but requires that predictive processing proponents reject environmental seclusion. To do so means adopting ecological information in place of the Shannon information most typically invoked by proponents of predictive processing, and giving many of the other semantic-sounding terms they use (e.g., “prediction,” “model,” “representation”) deflationary understandings.

Andy Clark’s exciting work on predictive processing provides the umbrella under which his hugely influential previous work on embodied and extended cognition seeks a unified home. This chapter argues that in fact predictive processing harbours internalist, inferentialist and epistemic tenets that cannot leave embodied and extended cognition unchanged. Predictive processing cannot do the work Clark requires of it without relying on rich, preconstructive internal representations of the world, nor without engaging in paradigmatically rational integration of prior knowledge and new sensory input. Hence, next to Clark’s image of fluid “uncertainty surfing” is an equally valid image of more emaciated and plodding world-modelling. Rather than underpinning orthodox embodied and extended approches, predictive processing therefore presents an opportunity for a potentially fruitful new synthesis of cognitivist and embodied approaches to cognition.

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. This chapter develops this position in the context of interoception (the sense of the body from within) and physiological regulation. A key idea here, which recalls twentieth-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. The chapter explores how this perspective helps connect control-oriented interoceptive inference to phenomenological properties of embodied selfhood and subjectivity. The upshot echoes (or perhaps subverts) a classic philosophical trope of the Enlightenment philosopher Julien de La Mettrie: to find the origins of our conscious selves in our nature as beast machines.

In the chapters that follow, the first-order probability design around narrative plot is developed. I.1: Plot and Probability Transformations concerns itself with plot events and prediction errors. I.2: Probability Designs discusses the links between design, the creative process, and the author’s intentionality. Finally, I.3: The Height of Drop addresses how readers’ perception of probabilities is manipulated.