Prediction Gone Awry: A Computational Framework for Emotion Dysregulation Across Autism and...

Area of Science:

• Neuroscience
• Developmental Psychology
• Computational Psychiatry

Background:

• Emotion dysregulation is prevalent in childhood and adolescence, particularly in autism and attention-deficit/hyperactivity disorder (ADHD).
• Existing research often describes symptoms like irritability and slow return to baseline without detailing the underlying computational mechanisms.
• A deeper understanding of the brain's predictive processes is needed to explain and address emotion regulation challenges.

Purpose of the Study:

• To propose a predictive-coding framework for understanding emotion regulation and dysregulation.
• To hypothesize specific computational differences in autism and ADHD related to prediction error processing.
• To identify potential physiological markers for testing these hypotheses.

Main Methods:

• The study proposes a theoretical framework based on predictive coding principles.
• It hypothesizes how altered precision of prediction errors and expectation updating contribute to emotion dysregulation.
• Candidate physiological correlates such as heart-rate variability and pupillary variability are outlined for empirical investigation.

Main Results:

• Emotion dysregulation may arise from overweighted prediction errors (excessive precision) or sluggish updating of prior expectations.
• Autism is hypothesized to involve heightened confidence in surprising inputs and slow expectation updating, leading to reactivity and inertia.
• ADHD may be characterized by weaker expectations and unstable gain control, resulting in rapid affective swings.

Conclusions:

• A predictive-coding account offers a mechanistic explanation for emotion dysregulation in developmental disorders.
• This framework suggests interventions should focus on restoring flexible inference rather than direct emotion suppression.
• Further empirical validation in developmental samples is required to confirm these hypotheses.