A clinician-facing account of PDA-style resistance through predictive processing, salience network gating, precision weighting, autonomy, and clinical formulation.
Some clients do not resist social norms because they disagree with them. The norm never became behaviourally relevant in the first place.
For psychologists working with PDA-like traits, autism, or AuDHD, that distinction matters. If the pattern is read as defiance, the intervention usually targets motivation, compliance, or values alignment. If it is read as a salience and precision-weighting difference, the clinical target changes completely.
This article offers a neurobiologically grounded account of PDA-style resistance using predictive processing theory, the tri-network model, and current evidence relating to autistic cognitive profiles. PDA is treated here as a clinically described and contested presentation within autism, not a settled neurobiological entity.
Jump to a section:
A client, often identified with PDA-like traits, autistic, or AuDHD, does not appear to resist social norms because they reject them. Instead, the norms seem to fail to register as behaviourally relevant.
The expectation does not land as something to be prioritised. It is not processed, weighed, and rejected. It simply does not enter the attentional priority system with enough weight to guide action.
That is not a small distinction. It changes the formulation.
If the pattern is interpreted as defiance, the clinician may ask, "How do we increase compliance?" If it is interpreted as a salience allocation difference, the better question is, "What conditions allow this signal to become meaningful inside the person's own predictive model?"
This is where predictive processing becomes clinically useful. It gives us a way to describe why externally assigned importance may not become internal importance, and why imposed relevance can sometimes make engagement worse.
Predictive processing, associated with Karl Friston's free energy principle, describes the brain as a system that continuously generates predictions about incoming sensory and social input, then updates those predictions based on error signals.
When predictions do not match incoming data, a prediction error arises. Learning and attention are driven by these errors, but not all errors are treated equally.
The brain assigns precision weighting to prediction errors. Precision determines how much influence a signal has on perception, learning, and behaviour.
High-precision signals are prioritised. They drive updating. They recruit attention. They change behaviour.
Low-precision signals are down-weighted. They may be noticed, but they do not necessarily alter the person's model of the world. They do not become sticky.
Precision is not fixed. It is shaped by prior experience, perceived reliability of the signal source, context, bodily state, social safety, and autonomy conditions.
In clinical terms, the problem is not always that the client cannot understand the demand. It may be that the demand has not been assigned enough precision to compete with other signals.
Social norms function as externally supplied priors about what is important:
For those priors to influence behaviour, they must be assigned sufficient precision.
In typical social cognition, consensus itself increases precision. If something is widely treated as important, that social convergence acts as a reliability signal. The fact that "everyone knows this" helps the signal become meaningful.
This is how cultural knowledge becomes sticky. Social agreement scaffolds attention and memory.
In some neurodivergent profiles, external consensus does not automatically translate into internal precision weighting. Social importance does not necessarily become cognitive importance. A famous name, a status hierarchy, or a cultural rule may fail to consolidate, not because it is rejected, but because it never becomes a high-precision signal.
That can look like refusal from the outside. Internally, it may be closer to absence.
Understanding how this plays out neurobiologically requires the salience network, a core large-scale brain network identified in Menon's tri-network model.
The salience network is not just a detector of importance. It regulates access to cognitive resources.
The salience network is anchored by two major cortical regions.
Anterior insula. The anterior insula sits deep within the lateral sulcus. It integrates interoceptive signals with cognitive and affective context. It helps translate bodily state into subjective feeling and contributes to the sense that something matters.
Dorsal anterior cingulate cortex. The dorsal anterior cingulate cortex, or dACC, sits on the medial frontal surface. It is involved in conflict monitoring, effort allocation, and mobilisation of cognitive control. It contributes to deciding whether a signal warrants action or further processing.
These regions operate as a tightly coupled system, supported by subcortical structures including the amygdala, striatum, and thalamus. Those structures contribute to motivational and affective salience detection.
The salience network functions as a switch between two other large-scale systems:
When the salience network assigns high importance to a stimulus, it helps suppress DMN activity and recruit the CEN. When it does not, cognition may remain internally oriented or disengaged from the external demand.
What is flagged as salient therefore determines more than attention. It shapes the entire cognitive mode that follows.
Within this architecture, PDA-like presentations can be understood as differences in how precision is assigned across two competing pathways.
In typical social cognition, the pathway often works like this:
This is why names stick. This is why celebrity culture can feel intuitive to many people. This is why social norms propagate. The system is calibrated to treat consensus as a meaningful signal.
A second pathway is driven less by social consensus and more by predictive utility:
Here, salience is not derived from social agreement. It is derived from predictive usefulness.
In PDA-like presentations, what clinicians observe is not a simple absence of social interest. It is an asymmetry in how these two pathways are weighted.
Externally assigned salience does not reliably achieve high precision on its own. It has to earn its way in through predictive utility, personal relevance, relational meaning, or autonomy-consistent engagement.
When social importance is explicitly imposed, the signal may fail to increase in precision. In some cases, the imposition itself appears to reduce its effective weighting.
"You should know this."
"Everyone knows this."
"This matters."
Those statements are intended to increase relevance. For some clients, they may do the opposite.
Within a predictive processing frame, this can be understood as context-dependent down-weighting of externally imposed precision, particularly when autonomy is perceived to be constrained.
This is a theoretical extension. PDA as a distinct neurobiological profile is not established in the research literature. The account here should be understood as a computational formulation of a clinically described pattern, not a confirmed mechanism.
This pattern has a recognisable phenomenology.
Some individuals describe delayed or absent engagement with culturally dominant salience systems, including celebrity culture, social status hierarchies, public figures, and mainstream entertainment references. This is not usually experienced as active rejection. It is experienced as a lack of inherent salience.
The signal does not automatically register as worth encoding.
At the same time, there may be strong and sustained engagement with structured systems, fictional universes, game mechanics, technical taxonomies, or niche domains where internal logic, feedback, and mastery pathways are clear.
The distinction is not capacity. It is precision assignment.
Memory encoding often reflects this. Public figures may not be represented primarily through names or labels, but through functional attributes such as demeanour, role, pattern, values, or behavioural signature. Identity is encoded relationally or functionally rather than nominally.
The name may be the low-utility handle. The person's functional pattern may be what actually consolidates.
Importantly, this weighting is not fixed. Socially derived signals can become highly salient when embedded in personally meaningful, relational, or goal-directed contexts. The issue is not that social information cannot matter. It is that social importance alone may not be enough.
For a broader clinical frame, this sits alongside the wider PsychVault guide to PDA and demand avoidance, the practical guide to low-demand communication scripts, and the discussion of autistic burnout and demand avoidance.
The clinical implications differ sharply from standard behavioural interpretations.
Repetition and social pressure are unlikely to be effective when the underlying issue is precision weighting rather than motivation. Increasing exposure to normative information does not guarantee increased salience. In some cases, it may reduce engagement if the person experiences the repetition as externally imposed demand.
More effective approaches target the conditions under which precision is assigned.
Autonomy appears to modulate whether external signals achieve sufficient precision to guide action. The more a person experiences the demand as imposed, the less likely the signal may be to land.
Choice is not decoration here. It is a salience condition.
Information becomes more available when it connects to the person's own goals, meanings, relationships, or predictive models. "This matters because everyone says it matters" is a weak lever. "This matters because it helps you predict, choose, protect, connect, or build" is stronger.
For some clients, the question is not "what is the expected behaviour?" but "what does this information help me do?" A demand that has a clear functional payoff is more likely to be assigned precision.
Low-demand approaches reduce the autonomy threat attached to the signal. This can allow the underlying information to be processed without the nervous system first having to defend against the demand form.
In this framing, demand avoidance is not the primary problem. It is a readout of how salience is being allocated in real time.
The intervention question shifts from "how do we increase compliance?" to "how do we increase the intrinsic precision of this signal within the individual's predictive model?"
That is slower work. It is also more clinically honest.
The salience network and its role in switching between DMN and CEN are well established in cognitive neuroscience.
Predictive processing accounts of autism are an active area of research, including models of attenuated priors, aberrant precision weighting, and atypical updating dynamics. These models are debated, but they are empirically grounded frameworks.
The extension of these models to PDA-like presentations is theoretical. PDA remains a contested construct within autism research, and its neurobiological specificity has not been established.
This formulation is offered as a computationally coherent interpretation of clinical patterns. It should not be treated as a settled explanatory model or as a replacement for careful individual formulation.
If you are building neurodiversity-affirming resources, PDA support tools, clinical formulation templates, or low-demand psychoeducation handouts, PsychVault is being built as a place to share resources clinicians can actually use. You can browse the resource library, or create a store if you have your own worksheets, group plans, handouts, or templates to share.
Language note: this article uses "PDA-like" and "PDA-style" because PDA is clinically useful for many people but remains contested as a diagnostic construct. The goal is formulation, not labelling for its own sake.
Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138.
Goulden, N., Khusnulina, A., Davis, N. J., Bracewell, R. M., Bokde, A. L., McNulty, J. P., & Mullins, P. G. (2014). The salience network is responsible for switching between the default mode network and the central executive network: replication from DCM. NeuroImage, 99, 180-190.
Lawson, R. P., Rees, G., & Friston, K. J. (2014). An aberrant precision account of autism. Frontiers in Human Neuroscience, 8, 302.
Menon, V. (2011). Large-scale brain networks and psychopathology: a unifying triple network model. Trends in Cognitive Sciences, 15(10), 483-506.
Menon, V., & Uddin, L. Q. (2010). Saliency, switching, attention and control: a network model of insula function. Brain Structure and Function, 214(5-6), 655-667.
Pellicano, E., & Burr, D. (2012). When the world becomes "too real": a Bayesian explanation of autistic perception. Trends in Cognitive Sciences, 16(10), 504-510.
Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., ... & Greicius, M. D. (2007). Dissociable intrinsic connectivity networks for salience processing and executive control. Journal of Neuroscience, 27(9), 2349-2356.
Uddin, L. Q. (2015). Salience processing and insular cortical function and dysfunction. Nature Reviews Neuroscience, 16(1), 55-61.
Share your thoughts and experiences with this resource.
Sign in to leave a comment
Move from strategy into implementation with templates, handouts, and psychoeducation tools already live on the marketplace.
Publish clinician-grade templates, build trust signals, and start growing an evergreen library under your own brand.
A clinical guide to autistic burnout, PDA-profile demand avoidance, masking, shutdown, skill loss, recovery time, and why demand reduction is often treatment.
A neurodiversity-affirming pillar guide to PDA, demand avoidance, the theoretical pillars behind autistic experience, and what visible behaviour does and does not tell us.
A clinical differential guide to PDA, oppositional defiant disorder, conduct disorder, anxiety-driven demand avoidance, intent, empathy, and contraindicated behavioural responses.
