Predictive Processing Framework Research Articles

Copredication in Context: A Predictive Processing Approach.

We propose a cognitive‐psychological model of linguistic intuitions about copredication statements. In copredication statements, like “The book is heavy and informative,” the nominal denotes two ontologically distinct entities at the same time. This has been considered a problem for standard truth‐conditional semantics. In this paper, we discuss two questions that have so far received less attention: What kinds of word representations and cognitive mechanisms are responsible for judgments about the felicitousness of copredication statements? Relatedly, why can similar copredication statements have different degrees of felicitousness? We first propose a cognitive‐computational model of copredication within the predictive processing framework. We then suggest that certain asymmetries in felicitousness judgments can be modeled in terms of a set of expectations that are influenced by higher‐order priors associated with discourse context and world knowledge.

The Role of Predictions, Their Confirmation, and Reward in Maintaining the Self-Concept.

The predictive processing framework posits that people continuously use predictive principles when interacting with, learning from, and interpreting their surroundings. Here, we suggest that the same framework may help explain how people process self-relevant knowledge and maintain a stable and positive self-concept. Specifically, we recast two prominent self-relevant motivations, self-verification and self-enhancement, in predictive processing (PP) terms. We suggest that these self-relevant motivations interact with the self-concept (i.e., priors) to create strong predictions. These predictions, in turn, influence how people interpret information about themselves. In particular, we argue that these strong self-relevant predictions dictate how prediction error, the deviation from the original prediction, is processed. In contrast to many implementations of the PP framework, we suggest that predictions and priors emanating from stable constructs (such as the self-concept) cultivate belief-maintaining, rather than belief-updating, dynamics. Based on recent findings, we also postulate that evidence supporting a predicted model of the self (or interpreted as such) triggers subjective reward responses, potentially reinforcing existing beliefs. Characterizing the role of rewards in self-belief maintenance and reframing self-relevant motivations and rewards in predictive processing terms offers novel insights into how the self is maintained in neurotypical adults, as well as in pathological populations, potentially pointing to therapeutic implications.

Counterfactual cognition and psychosis: adding complexity to predictive processing accounts

ABSTRACT Over the last decade or so, several researchers have considered the predictive processing framework (PPF) to be a useful perspective from which to shed some much-needed light on the mechanisms behind psychosis. Most approaches to psychosis within PPF come down to the idea of the “atypical” brain generating inaccurate hypotheses that the “typical” brain does not generate, either due to a systematic top-down processing bias or more general precision weighting breakdown. Strong at explaining common individual symptoms of psychosis, such approaches face some issues when we look at a more general clinical picture. In this paper, we propose an update on the current accounts of psychosis based on the realization that a neurotypical brain constantly generates non-actual, de-coupled, counterfactual hypotheses as part of healthy cognition. We suggest that what is going on in psychosis, at least in some cases, is not so much a generation of erroneous hypotheses, but rather an inability to correctly use the counterfactual ones. This updated view casts “accurate” cognition as more fragile and delicate, but also closes the gap between psychosis and typical cognition.

Schema-Centred Unity and Process-Centred Pluralism of the Predictive Mind

Proponents of the predictive processing (PP) framework often claim that one of the framework’s significant virtues is its unificatory power. What is supposedly unified are predictive processes in the mind, and these are explained in virtue of a common prediction error-minimisation (PEM) schema. In this paper, I argue against the claim that PP currently converges towards a unified explanation of cognitive processes. Although the notion of PEM systematically relates a set of posits such as ‘efficiency’ and ‘hierarchical coding’ into a unified conceptual schema, neither the frameworks’ algorithmic specifications nor its hypotheses about their implementations in the brain are clearly unified. I propose a novel way to understand the fruitfulness of the research program in light of a set of research heuristics that are partly shared with those common to Bayesian reverse engineering. An interesting consequence of this proposal is that pluralism is at least as important as unification to promote the positive development of the predictive mind.

Predictive minds can think: addressing generality and surface compositionality of thought

Predictive processing framework (PP) has found wide applications in cognitive science and philosophy. It is an attractive candidate for a unified account of the mind in which perception, action, and cognition fit together in a single model. However, PP cannot claim this role if it fails to accommodate an essential part of cognition—conceptual thought. Recently, Williams (Synthese 1–27, 2018) argued that PP struggles to address at least two of thought’s core properties—generality and rich compositionality. In this paper, I show that neither necessarily presents a problem for PP. In particular, I argue that because we do not have access to cognitive processes but only to their conscious manifestations, compositionality may be a manifest property of thought, rather than a feature of the thinking process, and result from the interplay of thinking and language. Pace Williams, both of these capacities, constituting parts of a complex and multifarious cognitive system, may be fully based on the architectural principles of PP. Under the assumption that language presents a subsystem separate from conceptual thought, I sketch out one possible way for PP to accommodate both generality and rich compositionality.

Cerebellar Contributions to Social Cognition in ASD: A Predictive Processing Framework.

Functional, structural, and cytoarchitectural differences in the cerebellum are consistently reported in Autism Spectrum Disorders (ASD). Despite this, the mechanisms governing cerebellar contributions to ASD, particularly within the sociocognitive domain, are not well understood. Recently, it has been suggested that several core features of ASD may be associated with challenges creating and using prior expectations or predictions to rapidly adapt to changing stimuli or situations, also known as adaptive prediction. Importantly, neuroimaging, clinical, and animal work find that the cerebellum supports adaptive prediction in both motor and non-motor domains. Perturbations to the cerebellum via injury or neuromodulation have been associated with impairments in predictive skills. Here, we review evidence for a cerebellar role in social cognition and adaptive prediction across individuals with and without ASD.

Predictive Neuronal Adaptation as a Basis for Consciousness.

Being able to correctly predict the future and to adjust own actions accordingly can offer a great survival advantage. In fact, this could be the main reason why brains evolved. Consciousness, the most mysterious feature of brain activity, also seems to be related to predicting the future and detecting surprise: a mismatch between actual and predicted situation. Similarly at a single neuron level, predicting future activity and adapting synaptic inputs accordingly was shown to be the best strategy to maximize the metabolic energy for a neuron. Following on these ideas, here we examined if surprise minimization by single neurons could be a basis for consciousness. First, we showed in simulations that as a neural network learns a new task, then the surprise within neurons (defined as the difference between actual and expected activity) changes similarly to the consciousness of skills in humans. Moreover, implementing adaptation of neuronal activity to minimize surprise at fast time scales (tens of milliseconds) resulted in improved network performance. This improvement is likely because adapting activity based on the internal predictive model allows each neuron to make a more “educated” response to stimuli. Based on those results, we propose that the neuronal predictive adaptation to minimize surprise could be a basic building block of conscious processing. Such adaptation allows neurons to exchange information about own predictions and thus to build more complex predictive models. To be precise, we provide an equation to quantify consciousness as the amount of surprise minus the size of the adaptation error. Since neuronal adaptation can be studied experimentally, this can allow testing directly our hypothesis. Specifically, we postulate that any substance affecting neuronal adaptation will also affect consciousness. Interestingly, our predictive adaptation hypothesis is consistent with multiple ideas presented previously in diverse theories of consciousness, such as global workspace theory, integrated information, attention schema theory, and predictive processing framework. In summary, we present a theoretical, computational, and experimental support for the hypothesis that neuronal adaptation is a possible biological mechanism of conscious processing, and we discuss how this could provide a step toward a unified theory of consciousness.

Self-deception in the predictive mind: cognitive strategies and a challenge from motivation

ABSTRACT In this article, we show how the phenomenon of self-deception when adequately analyzed, can be incorporated into a predictive processing framework. We describe four strategies by which a subject may become self-deceived to account for typical cases of self-deception. We then argue that the four strategies can be modeled within this framework, under the assumption that a satisfying account of motivation is possible within predictive processing. Finally, we outline how we can ground this assumption by discussing how such a systematic model of motivation may be developed using the distinction between a world-model and a self-model.

A robot or a dumper truck? Facilitating play-based social learning across neurotypes.

How can non-autistic adults facilitate social learning with children on the spectrum? A new theoretical understanding of autism is currently emerging that has made this question more relevant than ever. At the intersection of two growing research areas in the field of autism, the borderland that separates the experience of social interaction between neurotypes is increasingly mapped out. By integrating anthropological research on autistic sociality and the neurocognitive framework of predictive processing, this paper explores the question: If autistic people experience the world in a fundamentally different way, what is a meaningful strategy for supporting them in developing their socialities? The paper reports an in-depth analysis of a 2-min sequence in which a non-autistic adult facilitates a collaboration game between three autistic children (8-12 years). The data comes from a participatory research project that develops a new pedagogical approach to social learning based on open-ended construction play. The analytical strategy is informed by conversation analysis. We find that the facilitation supports the children in accomplishing social interaction and collaboration, but it also in several instances gives rise to misunderstandings between the children. Whereas the facilitator aims to support the children's direct verbal communication about the construction task, we observe that the children use a broad repertoire of non-direct communication strategies that enables them to coordinate and align their shared process. We find that the children's actions with their hands in the construction task count as turns in the communication. Regarding the play-based learning environment, we find that the children are engaged in the shared construction task and that they competently navigate social tension when it arises without the facilitator's help. We conclude that the misunderstandings between the children created by the facilitation from a non-autistic adult emerge from a discrepancy of attention in the situation. The facilitator focuses on the words, but the children focus on the task. Even though this discrepancy is not necessarily a result of different neurotypes, we find that it emerges from the social dynamics of facilitation by non-autistic adults that is key in many social intervention settings. Furthermore, we conclude that the play-based learning environment enables the facilitator to support the children without directly instructing them in their social behavior. This appears to give the children an opportunity to acquire complex social experiences through their collaboration. The interaction dynamics in the data clip is shaped by the non-autistic adult's expectations of the children's interaction. This made us wonder whether we can establish a learning environment that begins from the learners' perspectives instead. The analysis caused us to change the facilitation strategy that we employ in our project. It is our hope that our approach will inspire reflection and curiosity in researchers and practitioners who develop social interventions targeting autistic people.

The Predictive Dynamics of Happiness and Well-Being

We offer an account of mental health and well-being using the predictive processing framework (PPF). According to this framework, the difference between mental health and psychopathology can be located in the goodness of the predictive model as a regulator of action. What is crucial for avoiding the rigid patterns of thinking, feeling and acting associated with psychopathology is the regulation of action based on the valence of affective states. In PPF, valence is modelled as error dynamics—the change in prediction errors over time . Our aim in this paper is to show how error dynamics can account for both momentary happiness and longer term well-being. What will emerge is a new neurocomputational framework for making sense of human flourishing.

Provoking thought: A predictive processing account of critical thinking and the effects of education

In this paper, we propose that an increasingly regarded theoretical framework in neuroscience—the predictive processing framework—can help to advance an understanding of the foundations of critical thinking as well as provide a mechanistic hypothesis for how education may increase a learner’s subsequent use of critical thinking outside of an educational context. We begin by identifying a lacuna in the understanding of critical thinking: a causal account of the internal triggers to think critically. We then introduce the predictive processing framework to address this lacuna, and unpack the implications of this framework for understanding the deployment of critical thinking skills in situ, in the real world.

Inattention and Uncertainty in the Predictive Brain.

Negative effects of inattention on task performance can be seen in many contexts of society and human behavior, such as traffic, work, and sports. In traffic, inattention is one of the most frequently cited causal factors in accidents. In order to identify inattention and mitigate its negative effects, there is a need for quantifying attentional demands of dynamic tasks, with a credible basis in cognitive modeling and neuroscience. Recent developments in cognitive science have led to theories of cognition suggesting that brains are an advanced prediction engine. The function of this prediction engine is to support perception and action by continuously matching incoming sensory input with top-down predictions of the input, generated by hierarchical models of the statistical regularities and causal relationships in the world. Based on the capacity of this predictive processing framework to explain various mental phenomena and neural data, we suggest it also provides a plausible theoretical and neural basis for modeling attentional demand and attentional capacity "in the wild" in terms of uncertainty and prediction error. We outline a predictive processing approach to the study of attentional demand and inattention in driving, based on neurologically-inspired theories of uncertainty processing and experimental research combining brain imaging, visual occlusion and computational modeling. A proper understanding of uncertainty processing would enable comparison of driver's uncertainty to a normative level of appropriate uncertainty, and thereby improve definition and detection of inattentive driving. This is the necessary first step toward applications such as attention monitoring systems for conventional and semi-automated driving.

Neural network modeling of altered facial expression recognition in autism spectrum disorders based on predictive processing framework

The mechanism underlying the emergence of emotional categories from visual facial expression information during the developmental process is largely unknown. Therefore, this study proposes a system-level explanation for understanding the facial emotion recognition process and its alteration in autism spectrum disorder (ASD) from the perspective of predictive processing theory. Predictive processing for facial emotion recognition was implemented as a hierarchical recurrent neural network (RNN). The RNNs were trained to predict the dynamic changes of facial expression movies for six basic emotions without explicit emotion labels as a developmental learning process, and were evaluated by the performance of recognizing unseen facial expressions for the test phase. In addition, the causal relationship between the network characteristics assumed in ASD and ASD-like cognition was investigated. After the developmental learning process, emotional clusters emerged in the natural course of self-organization in higher-level neurons, even though emotional labels were not explicitly instructed. In addition, the network successfully recognized unseen test facial sequences by adjusting higher-level activity through the process of minimizing precision-weighted prediction error. In contrast, the network simulating altered intrinsic neural excitability demonstrated reduced generalization capability and impaired emotional clustering in higher-level neurons. Consistent with previous findings from human behavioral studies, an excessive precision estimation of noisy details underlies this ASD-like cognition. These results support the idea that impaired facial emotion recognition in ASD can be explained by altered predictive processing, and provide possible insight for investigating the neurophysiological basis of affective contact.

You can't always get what you want

The predictive processing framework has gained significant popularity across disciplines investigating the mind and brain. In this article we critically examine two of the recently made claims about the kind of headway that the framework can make in the neuroscientific and philosophical investigation of consciousness. Firstly, we argue that predictive processing is unlikely to yield significant breakthroughs in the search for the neural correlates of consciousness as it is still too vague to individuate neural mechanisms at a fine enough scale. Despite its unifying ambitions, the framework harbors a diverse family of competing computational models which rely on different assumptions and are under-constrained by neurological data. Secondly, we argue that the framework is also ill suited to provide a unifying theory of consciousness. Here, we focus on the tension between the claim that predictive processing is compatible with all of the leading neuroscientific models of consciousness with the fact that most attempts explaining consciousness within the framework rely heavily on external assumptions.

Predictive processing and anti-representationalism

Many philosophers claim that the neurocomputational framework of predictive processing entails a globally inferentialist and representationalist view of cognition. Here, I contend that this is not correct. I argue that, given the theoretical commitments these philosophers endorse, no structure within predictive processing systems can be rightfully identified as a representational vehicle. To do so, I first examine some of the theoretical commitments these philosophers share, and show that these commitments provide a set of necessary conditions the satisfaction of which allows us to identify representational vehicles. Having done so, I introduce a predictive processing system capable of active inference, in the form of a simple robotic “brain”. I examine it thoroughly, and show that, given the necessary conditions highlighted above, none of its components qualifies as a representational vehicle. I then consider and allay some worries my claim could raise. I consider whether the anti-representationalist verdict thus obtained could be generalized, and provide some reasons favoring a positive answer. I further consider whether my arguments here could be blocked by allowing the same representational vehicle to possess multiple contents, and whether my arguments entail some extreme form of revisionism, answering in the negative in both cases. A quick conclusion follows.

Perceiving as knowing in the predictive mind

On an ‘internalist’ picture, knowledge isn’t necessary for understanding the nature of perception and perceptual experience. This contrasts with the ‘knowledge first’ picture, according to which it’s essential to the nature of successful perceiving as a mental state that it’s a way of knowing. It’s often thought that naturalistic theorizing about the mind should adopt the internalist picture. However, I argue that a powerful, recently prominent framework for scientific study of the mind, ‘predictive processing,’ instead supports the knowledge first picture. Under predictive processing, it’s intrinsic to successful perceiving that it’s a state with the kind of modal robustness that’s distinctive of knowledge, which gives us reason to think of successful perceiving along knowledge first lines. Furthermore, I argue that the predictive processing framework encourages us to conceptualize experiences which don’t amount to knowledge along knowledge first lines, as states which by their nature fall short of knowledge.

Unilateral neglect within the predictive processing framework.

Unilateral neglect within the predictive processing framework.

Expectancy-based rhythmic entrainment as continuous Bayesian inference.

When presented with complex rhythmic auditory stimuli, humans are able to track underlying temporal structure (e.g., a "beat"), both covertly and with their movements. This capacity goes far beyond that of a simple entrained oscillator, drawing on contextual and enculturated timing expectations and adjusting rapidly to perturbations in event timing, phase, and tempo. Previous modeling work has described how entrainment to rhythms may be shaped by event timing expectations, but sheds little light on any underlying computational principles that could unify the phenomenon of expectation-based entrainment with other brain processes. Inspired by the predictive processing framework, we propose that the problem of rhythm tracking is naturally characterized as a problem of continuously estimating an underlying phase and tempo based on precise event times and their correspondence to timing expectations. We present two inference problems formalizing this insight: PIPPET (Phase Inference from Point Process Event Timing) and PATIPPET (Phase and Tempo Inference). Variational solutions to these inference problems resemble previous "Dynamic Attending" models of perceptual entrainment, but introduce new terms representing the dynamics of uncertainty and the influence of expectations in the absence of sensory events. These terms allow us to model multiple characteristics of covert and motor human rhythm tracking not addressed by other models, including sensitivity of error corrections to inter-event interval and perceived tempo changes induced by event omissions. We show that positing these novel influences in human entrainment yields a range of testable behavioral predictions. Guided by recent neurophysiological observations, we attempt to align the phase inference framework with a specific brain implementation. We also explore the potential of this normative framework to guide the interpretation of experimental data and serve as building blocks for even richer predictive processing and active inference models of timing.

Explanation Within Arm’s Reach: A Predictive Processing Framework for Single Arm Use in Octopuses

Octopuses are highly intelligent animals with vertebrate-like cognitive and behavioural repertoires. Despite these similarities, vertebrate-based models of cognition and behaviour cannot always be successfully applied to octopuses, due to the structural and functional characteristics that have evolved in their nervous system in response to the unique challenges posed by octopus morphology. For instance, the octopus brain does not support a somatotopic or point-for-point spatial map of the body—an important feature of vertebrate nervous systems. Thus, while octopuses are capable of motor tasks whose vertebrate counterparts require detailed interoceptive monitoring, these movements may not be explainable using motor control frameworks premised on internal spatial representation. One such motor task is the extension of a single arm. The ability of octopuses to select and use a single arm without the guidance of a somatotopic map has been regarded as a motor control puzzle. In an attempt at a solution, this paper develops a predictive processing account of single-arm extension in octopuses.

The first prior: From co-embodiment to co-homeostasis in early life

The idea that our perceptions in the here and now are influenced by prior events and experiences has recently received substantial support and attention from the proponents of the Predictive Processing (PP) and Active Inference framework in philosophy and computational neuroscience. In this paper we look at how perceptual experiences get off the ground from the outset, in utero. One basic yet overlooked aspect of current PP approaches is that human organisms first develop within another human body. Crucially, while not all humans will have the experience of being pregnant or carrying a baby, the experience of being carried and growing within another person’s body is universal. Specifically, we focus on the development of minimal selfhood in utero as a process co-embodiment and co-homeostasis, and highlight their close relationship. We conclude with some implications on several critical questions fuelling current debates on the nature of conscious experiences, minimal self and social cognition.