Predictive Processing Research Articles

Metacognition and sense of agency

Intelligent agents need to understand how they can change the world, and how they cannot change it, in order to make rational decisions for their forthcoming actions, and to adapt to their current environment. Previous research on the sense of agency, based largely on subjective ratings, failed to dissociate the sensitivity of sense of agency (i.e., the extent to which individual sense of agency tracks actual instrumental control over external events) from judgment criteria (i.e., the extent to which individuals self-attribute agency independent of their actual influence over external events). Furthermore, few studies have examined whether individuals have metacognitive access to the internal processes underlying the sense of agency. We developed a novel two-alternative-forced choice (2FAC) control detection task, in which participants identified which of two visual objects was more strongly controlled by their voluntary movement. The actual level of control over the target object was manipulated by adjusting the proportion of its motion that was driven by the participant's movement, compared to the proportion driven by a pre-recorded movement by another agent, using a staircase to hold 2AFC control detection accuracy at 70%. Participants identified which of the two visual objects they controlled, and also made a binary confidence judgment regarding their control detection judgment. We calculated a bias-free measure of first-order sensitivity (d’) for detection control at any given level of participant's own movement. The proportion of pre-recorded movements determined by the stairecase could then be used as an index of control detection ability. We identified two distinct processes underlying first-order detection of control: one based on instantaneous sensory predictions for the current movement, and one based on detection of a regular motor-visual relation across a series of movements. Further, we found large individual differences across 40 particpants in metacognitive sensitivity (meta-d’) even though first-order sensitivity of control detection was well controlled. Using structural equation modelling (SEM), we showed that metacognition was negatively correlated with the predictive process component of detection of control. This result is inconsistent with previous hypotheses that detection of control relies on metacognitive monitoring of a predictive circuit. Instead, it suggests that predictive mechanisms that compute sense of agency may operate unconsciously.

Literalism in Autistic People: a Predictive Processing Proposal

AbstractAutistic individuals are commonly said – and also consider themselves – to be excessively literalist, in the sense that they tend to prefer literal interpretations of words and utterances. This literalist bias seems to be fairly specific to autism and still lacks a convincing explanation. In this paper we explore a novel hypothesis that has the potential to account for the literalist bias in autism. We argue that literalism results from an atypical functioning of the predictive system: specifically, an atypical balance between predictions and error signals in language processing may make individuals more uncertain about their own predictions. Such uncertainty is then often resolved by resorting to the safest interpretation, that is, the literal one. We start by reviewing existing explanations of other autistic traits that appeal to predictive processing. We then apply these insights to language, by showing that predictions play a key role in everyday comprehension and that a low level of confidence in one’s own predictions is likely to escalate comprehension difficulties. Finally, we take a deeper look at non-literal uses of language by discussing the case of metaphors, to illustrate how a predictive processing account offers a promising explanation of the literalist bias in autism.

Two‐stage deadbeat‐based predictive torque control strategy for modular multilevel converter‐fed three‐phase induction motors

AbstractRecently, Modular Multilevel Converters (MMC) are considered to be extensively used in medium‐voltage AC drives A two‐stage deadbeat‐based strategy is developed by the authors to make the implementation of Predictive Torque Control (PTC) technique for an induction motor that is fed by a three‐phase MMC possible . Despite various benefits of PTC strategy, it has not been widely employed for multilevel topologies due to the huge number of switching states of these converters, and the complex internal dynamics of topologies such as MMC. The proposed control scheme results in significant reduction of the computational burden, using an initial reference voltage that is determined at the first stage of the strategy by the deadbeat technique. Next, the predictive evaluation process is conducted at the second stage to simultaneously control the stator flux, electromagnetic torque and circulating current. The number of admissible vectors that should be evaluated in the cost function is effectively limited by only considering some adjacent vectors of the reference voltage in the first stage. Moreover, the circulating current control capability, which is necessary for effective operation of MMC, is also considered in the developed scheme. Finally, the satisfactory performance of the proposed strategy is validated through various experimental tests.

Stimulus-Specific Prediction Error Neurons in Mouse Auditory Cortex.

Comparing expectation with experience is an important neural computation performed throughout the brain and is a hallmark of predictive processing. Experiments that alter the sensory outcome of an animal's behavior reveal enhanced neural responses to unexpected self-generated stimuli, indicating that populations of neurons in sensory cortex may reflect prediction errors (PEs), mismatches between expectation and experience. However, enhanced neural responses to self-generated stimuli could also arise through nonpredictive mechanisms, such as the movement-based facilitation of a neuron's inherent sound responses. If sensory prediction error neurons exist in sensory cortex, it is unknown whether they manifest as general error responses, or respond with specificity to errors in distinct stimulus dimensions. To answer these questions, we trained mice of either sex to expect the outcome of a simple sound-generating behavior and recorded auditory cortex activity as mice heard either the expected sound or sounds that deviated from expectation in one of multiple distinct dimensions. Our data reveal that the auditory cortex learns to suppress responses to self-generated sounds along multiple acoustic dimensions simultaneously. We identify a distinct population of auditory cortex neurons that are not responsive to passive sounds or to the expected sound but that encode prediction errors. These prediction error neurons are abundant only in animals with a learned motor-sensory expectation, and encode one or two specific violations rather than a generic error signal. Together, these findings reveal that cortical predictions about self-generated sounds have specificity in multiple simultaneous dimensions and that cortical prediction error neurons encode specific violations from expectation.SIGNIFICANCE STATEMENT Audette et. al record neural activity in the auditory cortex while mice perform a sound-generating forelimb movement and measure neural responses to sounds that violate an animal's expectation in different ways. They find that predictions about self-generated sounds are highly specific across multiple stimulus dimensions and that a population of typically nonsound-responsive neurons respond to sounds that violate an animal's expectation in a specific way. These results identify specific prediction error (PE) signals in the mouse auditory cortex and suggest that errors may be calculated early in sensory processing.

Body as First Teacher: The Role of Rhythmic Visceral Dynamics in Early Cognitive Development.

Embodied cognition-the idea that mental states and processes should be understood in relation to one's bodily constitution and interactions with the world-remains a controversial topic within cognitive science. Recently, however, increasing interest in predictive processing theories among proponents and critics of embodiment alike has raised hopes of a reconciliation. This article sets out to appraise the unificatory potential of predictive processing, focusing in particular on embodied formulations of active inference. Our analysis suggests that most active-inference accounts invoke weak, potentially trivial conceptions of embodiment; those making stronger claims do so independently of the theoretical commitments of the active-inference framework. We argue that a more compelling version of embodied active inference can be motivated by adopting a diachronic perspective on the way rhythmic physiological activity shapes neural development in utero. According to this visceral afferent training hypothesis, early-emerging physiological processes are essential not only for supporting the biophysical development of neural structures but also for configuring the cognitive architecture those structures entail. Focusing in particular on the cardiovascular system, we propose three candidate mechanisms through which visceral afferent training might operate: (a) activity-dependent neuronal development, (b) periodic signal modeling, and (c) oscillatory network coordination.

Individual prediction tendencies do not generalize across modalities.

Predictive processing theories, which model the brain as a "prediction machine", explain a wide range of cognitive functions, including learning, perception and action. Furthermore, it is increasingly accepted that aberrant prediction tendencies play a crucial role in psychiatric disorders. Given this explanatory value for clinical psychiatry, prediction tendencies are often implicitly conceptualized as individual traits or as tendencies that generalize across situations. As this has not yet explicitly been shown, in the current study, we quantify to what extent the individual tendency to anticipate sensory features of high probability generalizes across modalities. Using magnetoencephalography (MEG), we recorded brain activity while participants were presented with a sequence of four different (either visual or auditory) stimuli, which changed according to predefined transitional probabilities of two entropy levels: ordered vs. random. Our results show that, on a group-level, under conditions of low entropy, stimulus features of high probability are preactivated in the auditory but not in the visual modality. Crucially, the magnitude of the individual tendency to predict sensory events seems not to correlate between the two modalities. Furthermore, reliability statistics indicate poor internal consistency, suggesting that the measures from the different modalities are unlikely to reflect a single, common cognitive process. In sum, our findings suggest that quantification and interpretation of individual prediction tendencies cannot be generalized across modalities.

Perception and Disjunctive Belief: A New Problem for Ambitious Predictive Processing

ABSTRACT Perception can’t have disjunctive content. Whereas you can think that a box is blue or red, you can’t see a box as being blue or red. Based on this fact, I develop a new problem for the ambitious predictive processing theory, on which the brain is a machine for minimizing prediction error, which approximately implements Bayesian inference. I describe a simple case of updating a disjunctive belief given perceptual experience of one of the disjuncts, in which Bayesian inference and predictive coding pull in opposite directions, with the former implying that one’s confidence in the belief should increase, and the latter implying that it should decrease. Thus, predictive coding fails to approximately implement Bayesian inference across the interface between belief and perception.

The Audiovisual Mismatch Negativity in Predictive and Non-Predictive Speech Stimuli in Older Adults With and Without Hearing Loss.

Adults with aging-related hearing loss (ARHL) experience adaptive neural changes to optimize their sensory experiences; for example, enhanced audiovisual (AV) and predictive processing during speech perception. The mismatch negativity (MMN) event-related potential is an index of central auditory processing; however, it has not been explored as an index of AV and predictive processing in adults with ARHL. In a pilot study we examined the AV MMN in two conditions of a passive oddball paradigm- one AV condition in which the visual aspect of the stimulus can predict the auditory percept and one AV control condition in which the visual aspect of the stimulus cannot predict the auditory percept. In adults with ARHL, evoked responses in the AV conditions occurred in the early MMN time window while the older adults with normal hearing showed a later MMN. Findings suggest that adults with ARHL are sensitive to AV incongruity, even when the visual is not predictive of the auditory signal. This suggests that predictive coding for AV speech processing may be heightened in adults with ARHL. This paradigm can be used in future studies to measure treatment related changes, for example via aural rehabilitation, in older adults with ARHL.

Graded and ungraded expectation patterns: Prediction dynamics during active comprehension.

Language comprehension can be facilitated by the accurate prediction of upcoming words, but prediction effects are not ubiquitous, and comprehenders likely use predictive processing to varying degrees depending on task demands. To ascertain the processing consequences of prioritizing prediction, we here compared ERPs elicited when young adult participants simply read for comprehension with those collected in a subsequent block that required active prediction. We were particularly interested in frontally-distributed post-N400 effects for expected and unexpected words in strongly constraining contexts, which have previously been documented as two distinct patterns: an enhanced positivity ("anterior positivity") observed for prediction violations compared to words that are merely unpredictable (because they occur in weakly constraining sentences) and a distinction between expected endings in more constraining contexts and those same weakly constrained words ("frontal negativity" to the strongly predicted words). We found that the size of the anterior positivity effect was unchanged between passive comprehension and active prediction, suggesting that some processes related to prediction may engage state-like networks. On the other hand, the frontal negativity showed graded patterns from the interaction of task and sentence type. These differing patterns support the hypothesis that there are two separate effects with frontal scalp distributions that occur after the N400 and further suggest that the impact of violating predictions (as long as prediction is engaged at all) is largely stable across varying levels of effort/attention directed toward prediction, whereas other comprehension processes can be modulated by task demands.

Morphosyntactic predictive processing in adult heritage speakers: effects of cue availability and spoken and written language experience

ABSTRACTWe investigated prediction skills of adult heritage speakers and the role of written and spoken language experience on predictive processing. Using visual world eye-tracking, we focused on predictive use of case-marking cues in verb-medial and verb-final sentences in Turkish with adult Turkish heritage speakers (N = 25) and Turkish monolingual speakers (N = 24). Heritage speakers predicted in verb-medial sentences (when verb-semantic and case-marking cues were available), but not in verb-final sentences (when only case-marking cues were available) while monolinguals predicted in both. Prediction skills of heritage speakers were modulated by their spoken language experience in Turkish and written language experience in both languages. Overall, these results strongly suggest that verb-semantic information is needed to scaffold the use of morphosyntactic cues for prediction in heritage speakers. The findings also support the notion that both spoken and written language experience play an important role in predictive spoken language processing.

Arendt in the Metaverse: Four properties of eXtended Reality that imperil factual truth and democracy

This article explores the democratic implications of the wide adoption of XR (eXtended Reality) technologies, drawing on Hannah Arendt’s political theory and her concept of common reality. Arendt argues that the intersubjective construction of a common reality is what enables the establishment of factual truth and that both are prerequisites for a functional democracy. Without them, ideology can be established as truth by anti-democratic forces with power over media outlets, paving a path toward totalitarianism. This article argues that XR technologies can be used to inhibit the construction of a common reality through the same individualization of media experiences that has been shown to impede democratic processes in the social media context. The companies generating increased revenue through individualized micro-segmentation are now also vying for dominance in the XR media arena. It is argued in this article that such individualization can impede the co-construction of a common reality and a factual truth because XR media are hyper-persuasive and capable of altering an individual’s overall perception of reality. The viability of such platform-controlled individualization in XR is demonstrated through an Arendt-based critique of four properties related to XR technologies: Hypertargeted personalization, false memory creation, reality indistinguishability, and predictive processing theory. Based on these analyses, the article concludes that XR policy and regulation must consider how XR gives platforms unprecedented persuasive powers as they become capable of altering a user’s reality perception remotely and how, per Arendt, this may threaten democracy.

The Development of Predictive Gender Processing Strategies During Noun Phrase Decoding: An Eye-Tracking Study With German 5- to 10-Year-Old Children.

Many models of language comprehension assume that listeners predict the continuation of an incoming linguistic stimulus immediately after its onset, based on only partial linguistic and contextual information. Their related developmental models try to determine which cues (e.g., semantic or morphosyntactic) trigger such prediction, and to which extent, during different periods of language acquisition. One morphosyntactic cue utilized predictively in many languages, inter alia German, is grammatical gender. However, studies of the developmental trajectories of the acquisition of predictive gender processing in German remain a few. This study attempts to shed light on such processing strategies used in noun phrase decoding among children acquiring German as their first language by examining their eye movements during a language-picture matching task (N = 78, 5-10 years old). Its aim was to confirm whether the eye movements indicated the presence of age-specific differences in the processing of a gender cue, provided either in isolation or in combination with a semantic cue. The results revealed that German children made use of predictive gender processing strategies from the age of 5 years onward; however, the pace of online gender processing, as well as confidence in the predicted continuation, increased up to the age of 10 years. Predictive processing of gender cues plays a role in German language comprehension even in children younger than 8 years.

Aging attenuates the memory advantage for unexpected objects in real-world scenes

Across the adult lifespan memory processes are subject to pronounced changes. Prior knowledge and expectations might critically shape functional differences; however, corresponding findings have remained ambiguous so far. Here, we chose a tailored approach to scrutinize how schema (in-)congruencies affect older and younger adults’ memory for objects embedded in real-world scenes, a scenario close to everyday life memory demands. A sample of 23 older (52–81 years) and 23 younger adults (18–38 years) freely viewed 60 photographs of scenes in which target objects were included that were either congruent or incongruent with the given context information. After a delay, recognition performance for those objects was determined. In addition, recognized objects had to be matched to the scene context in which they were previously presented. While we found schema violations beneficial for object recognition across age groups, the advantage was significantly less pronounced in older adults. We moreover observed an age-related congruency bias for matching objects to their original scene context. Our findings support a critical role of predictive processes for age-related memory differences and indicate enhanced weighting of predictions with age. We suggest that recent predictive processing theories provide a particularly useful framework to elaborate on age-related functional vulnerabilities as well as stability.

Advancements in Oral Maxillofacial Surgery: A Comprehensive Review on 3D Printing and Virtual Surgical Planning

This comprehensive review explores the advancements in Orthognathic and Oral Maxillofacial Surgery, focusing on the integration of 3D Printing and Virtual Surgical Planning (VSP). Traditional surgical methods, while effective, come with inherent risks and complications, and can lead to variability in outcomes due to the reliance on the surgeon’s skill and experience. The shift towards patient-centric care necessitates personalized surgical methods, which can be achieved through advanced technology. The amalgamation of 3D printing and VSP revolutionizes surgical planning and implementation by providing tactile 3D models for visualization and planning, and accurately designed surgical guides for execution. This convergence of digital planning and physical modeling facilitates a more predictable, personalized, and precise surgical process. However, the adoption of these technologies presents challenges, including the need for extensive software training and the steep learning curve associated with computer-aided design programs. Despite these challenges, the integration of 3D printing and VSP paves the way for advanced patient care in orthognathic and oral maxillofacial surgery.

So Cloze Yet So Far: N400 Amplitude Is Better Predicted by Distributional Information Than Human Predictability Judgements

More predictable words are easier to process—they are read faster and elicit smaller neural signals associated with processing difficulty, most notably, the N400 component of the event-related brain potential. Thus, it has been argued that prediction of upcoming words is a key component of language comprehension, and that studying the amplitude of the N400 is a valuable way to investigate the predictions we make. In this study, we investigate whether the linguistic predictions of computational language models or humans better reflect the way in which natural language stimuli modulate the amplitude of the N400. One important difference in the linguistic predictions of humans versus computational language models is that while language models base their predictions exclusively on the preceding linguistic context, humans may rely on other factors. We find that the predictions of three top-of-the-line contemporary language models—GPT-3, RoBERTa, and ALBERT—match the N400 more closely than human predictions. This suggests that the predictive processes underlying the N400 may be more sensitive to the statistics of language than previously thought.

Molecularly targetable cell types in mouse visual cortex have distinguishable prediction error responses

Predictive processing postulates the existence of prediction error neurons in cortex. Neurons with both negative and positive prediction error response properties have been identified in layer 2/3 of visual cortex, but whether they correspond to transcriptionally defined subpopulations is unclear. Here we used the activity-dependent, photoconvertible marker CaMPARI2 to tag neurons in layer 2/3 of mouse visual cortex during stimuli and behaviors designed to evoke prediction errors. We performed single-cell RNA-sequencing on these populations and found that previously annotated Adamts2 and Rrad layer 2/3 transcriptional cell types were enriched when photolabeling during stimuli that drive negative or positive prediction error responses, respectively. Finally, we validated these results functionally by designing artificial promoters for use in AAV vectors to express genetically encoded calcium indicators. Thus, transcriptionally distinct cell types in layer 2/3 that can be targeted using AAV vectors exhibit distinguishable negative and positive prediction error responses.

Predictive event segmentation and representation with neural networks: A self-supervised model assessed by psychological experiments

People segment complex, ever-changing, and continuous experience into basic, stable, and discrete spatio-temporal experience units, called events. The literature on event segmentation investigates the mechanisms behind this ability. Event segmentation theory points out that people predict ongoing activities and observe prediction error signals to find event boundaries. In this study, we investigated the mechanism giving rise to this ability through a computational model and accompanying psychological experiments. Inspired by event segmentation theory and predictive processing, we introduced a self-supervised model of event segmentation. This model consists of neural networks that predict the sensory signal in the next time-step to represent different events, and a cognitive model that regulates these networks on the basis of their prediction errors. In order to verify the ability of our model in segmenting events, learning them during passive observation, and representing them in its representational space, we prepared a video of human behaviors represented by point-light displays. We compared the event segmentation behaviors of participants and our model with this video in two granularities. Using point-biserial correlation, we demonstrated that the event boundaries of our model correlated with the responses of the participants. Moreover, by approximating the representation space of participants, we showed that our model formed a similar representation space with those of participants. The result suggests that our model that tracks the prediction error signals can produce human-like event boundaries and event representations. Finally, we discuss our contribution to the literature and our understanding of how event segmentation is implemented in the brain.

Mental Gravity: Depression as Spacetime Curvature of the Self, Mind, and Brain.

The principle of mental gravity contends that the mind uses physical gravity as a mental model or simulacrum to express the relation between the inner self and the outer world in terms of "UP"-ness and "DOWN"-ness. The simulation of increased gravity characterises a continuum of mental gravity which states includes depression as the paradigmatic example of being down, low, heavy, and slow. The physics of gravity can also be used to model spacetime curvature in depression, particularly gravitational time dilation as a property of MG analogous to subjective time dilation (i.e., the slowing of temporal flow in conscious experience). The principle has profound implications for the Temporo-spatial Theory of Consciousness (TTC) with regard to temporo-spatial alignment that establishes a "world-brain relation" that is centred on embodiment and the socialisation of conscious states. The principle of mental gravity provides the TTC with a way to incorporate the structure of the world into the structure of the brain, conscious experience, and thought. In concert with other theories of cognitive and neurobiological spacetime, the TTC can also work towards the "common currency" approach that also potentially connects the TTC to predictive processing frameworks such as free energy, neuronal gauge theories, and active inference accounts of depression. It gives the up/down dimension of space, as defined by the gravitational field, a unique status that is connected to both our embodied interaction with the physical world, and also the inverse, reflective, emotional but still embodied experience of ourselves.

Genetic programming-based predictive model for the Cr removal effect of in-situ electrokinetic remediation in contaminated soil

Soil electrokinetic remediation is an emerging and efficient in-situ remediation technology for reducing environmental risks. Promoting the dissolution and migration of Cr in soil under the electric field is crucial to decrease soil toxicity and ecological influences. However, it is difficult to establish strong relationships between soil treatment and impact factors and to quantify their contributions. Machine learning can help establish pollutant migration models, but it is challenging to derive predictive formulas to improve remediation efficiency, describe the predictive model construction process, and reflect the importance of the predictors for better regulation. Therefore, this paper established a predictive model for the electrokinetic remediation of Cr-contaminated soil using genetic programming (GP) after determining the characteristic parameters which influenced the remediation effect, described the model's adaptive optimization process through the algorithm's function, and identified the sensitivity factors affecting the Cr removal effect. Results showed that the Cr(VI) and total Cr concentrations predicted by GP were in satisfactory agreement with the experimental values, 92% of the training data and 90% of the validation data achieved errors within 1%, and could fully reflect the target ions’ content variation in different soil layers. By substituting the above prediction formulas into Sobol sensitivity analysis, it was determined that conductivity, pH, current, and moisture content dramatically affected the Cr content variation in distinguished regions. For overall contaminated area, the system current and soil pH were the most sensitive factors for Cr(VI) and total Cr contents. Remediation efforts throughout the contaminated area should focus on the role of current versus soil pH. GP and sensitivity analysis can provide decision support and operational guidance for in-situ soil electrokinetic treatment by establishing a remediation effect prediction model, expediting the development and innovation of electrokinetic technology.

Robust multisensory deviance detection in the mouse parietal associative area

Context modulates how information is processed in the mammalian brain. For example, brain responses are amplified to contextually unusual stimuli. This phenomenon, known as "deviance detection,"1,2 is well documented in early, primary sensory cortex, where large responses are generated to simple stimuli that deviate from their context in low-order properties, such as line orientation, size, or pitch.2,3,4,5 However, the extent to which neural deviance detection manifests (1) in broader cortical networks and (2) to simple versus complex stimuli, which deviate only in their higher-order, multisensory properties, is not known. Consistent with a predictive processing framework,6,7 we hypothesized that deviance detection manifests in a hierarchical manner across cortical networks,8,9 emerging later and further downstream when stimulus deviance is complex. To test this, we examined brain responses of awake mice to simple unisensory deviants (e.g., visual line gratings, deviating from context in their orientation alone) versus complex multisensory deviants (i.e., audiovisual pairs, deviating from context only in their audiovisual pairing but not visual or auditory content alone). We find that mouse parietal associative area-a higher cortical region-displays robust multisensory deviance detection. In contrast, primary visual cortex exhibits strong unisensory visual deviance detection but weaker multisensory deviance detection. These results suggest that deviance detection signals in the cortex may be conceptualized as "prediction errors," which are primarily fed forward-or downstream-in cortical networks.6,7.