Does ADHD Exist? A Predictive Processing Account of Executive Dysregulation
Toward a Precision-Based Integrative Model
Palabras clave: ADHD, predictive processing, active inference, executive functions, inhibition, neurodevelopment, precision weighting
Abstract
Attention-Deficit/Hyperactivity Disorder (ADHD) remains a controversial construct regarding its ontological status, etiology, and treatment. This paper proposes an integrative framework grounded in predictive processing theory, conceptualizing ADHD as a disorder of precision regulation in hierarchical inference systems.
This approach unifies classical models, including inhibitory control deficits and reward-based (delay aversion) accounts, while also incorporating evolutionary perspectives such as the “hunter hypothesis.” We introduce the Predictive Executive Dysregulation Model (PEDM), which posits that ADHD arises from instability in the allocation of precision to predictions and prediction errors, particularly in goal-directed behavior. The model generates testable predictions and offers clinical implications for multimodal interventions.1. Introduction
The question of whether ADHD “exists” reflects a deeper epistemological issue: the lack of a unified definition of mental disorder. While the behavioral phenotype of ADHD—characterized by inattention, impulsivity, and executive dysfunction—is well documented, its interpretation varies across theoretical frameworks.
Traditional debates contrast biological versus psychosocial explanations. However, such dichotomies may obscure the need for mechanistic models capable of integrating multiple levels of analysis.
This paper argues that predictive processing provides such a framework.
2. ADHD: Competing Models and Fragmentation
Historically, ADHD has been explained through distinct but overlapping models:
2.1. Inhibitory Control Deficit
This model emphasizes dysfunction in fronto-striatal circuits affecting behavioral inhibition and executive control.
2.2. Dual Pathway / Reward Models
Alternative accounts highlight altered reward processing, particularly delay aversion, implicating mesolimbic dopaminergic systems.
2.3. State Regulation and Variability
ADHD is also associated with fluctuations in attention and inefficient information processing rather than pure impulsivity.
These models, while empirically supported, remain fragmented and lack a unifying computational framework.
3. Predictive Processing as a Unifying Framework
Predictive processing conceptualizes the brain as a hierarchical inference system that:
- generates predictions about sensory input
- minimizes prediction error
- assigns precision (confidence weighting) to signals
Perception, action, and cognition emerge from minimizing prediction error under uncertainty.
A central mechanism is precision weighting, which determines whether the system trusts:
- top-down predictions
- bottom-up sensory input
4. The Predictive Executive Dysregulation Model (PEDM)
We propose that ADHD can be understood as:
A disorder of precision regulation affecting the stabilization of goal-directed predictions over time.
Core Mechanisms
-
Reduced stability of high-level predictions
- difficulty maintaining task goals
-
Aberrant precision allocation
- excessive weighting of incoming stimuli → distractibility
-
Increased prediction error volatility
- behavioral variability
-
Impaired temporal integration
- difficulty linking present actions to future outcomes
5. Integration of Classical Theories
5.1. Inhibition Reinterpreted
In predictive terms:
- inhibition = maintaining a prediction despite competing inputs
- ADHD = failure to stabilize predictions
Thus, inhibitory deficits emerge as a secondary phenomenon of unstable precision control.
5.2. Delay Aversion and Reward
Reward sensitivity reflects altered precision in value estimation:
- immediate rewards = high precision
- delayed rewards = low precision
This aligns with findings on motivational dysfunction in ADHD.
5.3. The Hunter Hypothesis
From a predictive perspective:
- ADHD-like cognition favors exploration over exploitation
- prioritizes novel or unexpected stimuli
This may be adaptive in uncertain environments, but maladaptive in structured contexts.
6. Empirical Predictions (Falsifiability)
The PEDM generates testable hypotheses:
-
Increased intra-individual variability
- due to unstable precision weighting
-
Context-dependent performance
- improvement under structured, low-uncertainty conditions
-
Enhanced sensitivity to novelty and salience
- over-weighting of prediction errors
-
Motivation-dependent cognition
- performance modulated by reward salience
-
Pharmacological effects
- dopaminergic agents improve signal-to-noise (precision regulation)
7. Clinical Implications
This model reframes treatment as precision modulation:
7.1. Pharmacotherapy
Enhances signal stability and reduces noise in neural processing.
7.2. Behavioral Interventions
External structure reduces uncertainty → stabilizes predictions.
7.3. Exercise and Arousal Regulation
Modulate dopaminergic tone and attentional stability.
7.4. Psychotherapy
Strengthens internal models and goal representation.
8. Discussion
The PEDM resolves key tensions:
- Biological vs. psychosocial → both modulate precision
- Deficit vs. difference → context-dependent functionality
- Heterogeneity → explained by variability in precision regulation
Rather than a discrete disease entity, ADHD emerges as a dynamic systems-level dysfunction.
9. Conclusion
ADHD does not exist as a clearly bounded natural kind, but as a reliable pattern of dysregulated predictive processing.
The Predictive Executive Dysregulation Model integrates:
- inhibitory control theories
- reward-based models
- evolutionary perspectives
and reframes ADHD as:
A disorder of stabilizing predictions in time under conditions of uncertainty.
References (APA 7)
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204.
Da Costa, L., Parr, T., Sengupta, B., & Friston, K. (2020). Neural dynamics under active inference. Neural Computation.
Faraone, S. V., & Larsson, H. (2019). Genetics of attention deficit hyperactivity disorder. Molecular Psychiatry, 24, 562–575.
Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.
Metin, B., et al. (2013). ADHD performance reflects inefficient information processing. Neuropsychology, 27(2), 193–200.
Sonuga-Barke, E. J. S. (2005). Causal models of ADHD. Biological Psychiatry, 57(11), 1231–1238.
Sonuga-Barke, E. J. S. (2010). Delay aversion and dual pathway models. Neuropsychology Review, 20(1), 86–102.
Sonuga-Barke, E. J. S., & Sergeant, J. A. (2005). The neuroscience of ADHD. Developmental Science, 8(2), 103–104.
Sonuga-Barke, E. J. S., & Castellanos, F. X. (2007). Spontaneous attentional fluctuations. Neuroscience & Biobehavioral Reviews, 31(7), 977–986.
