Decomposing response inhibition: A POMDP model
- 1Department of Computational Neuroscience, Max Planck Institute for Biological Cybernetics, Tübingen, Germany
- 2Department of Psychiatry and Psychotherapy, Tübingen Center for Mental Health, University of Tübingen, Tübingen, Germany
- 3International Max Planck Research School (IMPRS) for The Mechanisms of Mental Function and Dysfunction (MMFD), Graduate Training Centre of Neuroscience (GTC), Tübingen, Germany
Abstract
Inhibitory control is a core cognitive function whose competence varies across the population, with impairments often observed in psychiatric conditions such as attention deficit hyperactivity disorder (ADHD). The Stop Signal Task (SST) is a widely used paradigm for assessing this ability. However, conventional formalizations of SST performance, such as the independent race model, rely on assumptions that are frequently violated in modern experimental designs. Furthermore, they typically fit only mean reaction times, overlooking crucial trial-by-trial dynamics. To address these limitations, we formalize the SST as a partially observable Markov decision process (POMDP). This framework characterizes inhibitory control with two components: noisy perceptual inference regarding stimuli and optimal control balanced against potential costs. To fit this model to the Adolescent Brain Cognitive Development (ABCD) study baseline cohort (N = 3,567), we introduce Transformer-encoded Simulation-Based Inference (TeSBI). This end-to-end architecture learns compact, sequence-aware embeddings from raw behavioral data. It enables efficient, amortized inference of individual-level posteriors. Extensive validation confirms it extracts reliable and identifiable parameters. We identify distinct latent computational attributes associated with scores on ADHD questionnaires. Controlling for sex, IQ, and medication status, children with higher ADHD scores exhibit subtle but robust shifts in computational attributes. They show a reduction in go cue directional precision, alongside a blunted sensitivity to stop error, go error and time costs. The learned embedding space reveals a continuous manifold in which children with higher ADHD scores are heterogeneously distributed, rather than forming distinct disorder clusters. This indicates that similar clinical characteristics can emerge from diverse combinations of computational mechanisms, supporting a dimensional perspective on neurodiversity. Our end-to-end framework can be extended to a broader range of cognitive tasks. It offers a scalable, theory-driven solution for analyzing large-scale behavioral data. Author summary: Inhibitory control is essential for adjusting thoughts and behavior and is often impaired in conditions like ADHD. Traditional assessments often oversimplify the underlying decision-making. We address this using a biologically grounded mathematical framework that separates how we perceive signals from how we execute control. This framework remains valid in diverse experimental designs where traditional models fail. To apply this complex model at scale, we develop a specialized machine learning approach (TeSBI) to reverse-engineer individual cognitive profiles. Analyzing data from over 3,000 children, we linked higher ADHD scores to subtle, consistent computational shifts: a reduction in go cue directional precision, a blunted sensitivity to the costs of making mistakes or taking too much time. These alterations remain robust regardless of sex, IQ, or medication status. Rather than forming a single disorder cluster, these children displayed diverse cognitive profiles, supporting a dimensional view of neurodiversity. By combining theory-driven cognitive modeling with scalable data-driven inference, our framework effectively captures complex inhibitory mechanisms and enables the precise analysis of large-scale behavioral datasets. This approach paves the way for more personalized strategies in computational psychiatry by recognizing the inherent heterogeneity within clinical characteristics.
Related
- Decomposing alarm-calling behaviour to investigate context-dependent threat discrimination and response escalation in a social mongoose
- Chronic Alcohol Exposure Induces Persistent DNA Damage Response Signaling and Sensitizes Breast Epithelial Cells to PARP Inhibition
- Reversing the Hypoxic Adaptive Response in Breast Cancer Cells through Small-Molecule Inhibition of Oncogenic MicroRNA-210
- Pharmacologically altered androgens, brain activation and response inhibition in a stop-signal task in male heavy drinkers
- A Visually Interpretable Histopathology-Based Immune Model Predicts T-effector Biology and Response to Immune checkpoint inhibition in Clear Cell Renal Cell Carcinoma Clinical Trial and Contemporary Real-World Datasets
BibTeX
@article{wang2026decomposing,
title = {{Decomposing response inhibition: A POMDP model}},
author = {Wenting Wang and Tobias Kaufmann and Peter Dayan},
journal = {PLOS Computational Biology},
year = {2026},
month = sep,
doi = {10.1371/journal.pcbi.1014068},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014068},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Decomposing response inhibition: A POMDP model
AU - Wenting Wang
AU - Tobias Kaufmann
AU - Peter Dayan
PY - 2026
DA - 2026/09/25
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014068
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014068
AB - Inhibitory control is a core cognitive function whose competence varies across the population, with impairments often observed in psychiatric conditions such as attention deficit hyperactivity disorder (ADHD). The Stop Signal Task (SST) is a widely used paradigm for assessing this ability. However, conventional formalizations of SST performance, such as the independent race model, rely on assumptions that are frequently violated in modern experimental designs. Furthermore, they typically fit only mean reaction times, overlooking crucial trial-by-trial dynamics. To address these limitations, we formalize the SST as a partially observable Markov decision process (POMDP). This framework characterizes inhibitory control with two components: noisy perceptual inference regarding stimuli and optimal control balanced against potential costs. To fit this model to the Adolescent Brain Cognitive Development (ABCD) study baseline cohort (N = 3,567), we introduce Transformer-encoded Simulation-Based Inference (TeSBI). This end-to-end architecture learns compact, sequence-aware embeddings from raw behavioral data. It enables efficient, amortized inference of individual-level posteriors. Extensive validation confirms it extracts reliable and identifiable parameters. We identify distinct latent computational attributes associated with scores on ADHD questionnaires. Controlling for sex, IQ, and medication status, children with higher ADHD scores exhibit subtle but robust shifts in computational attributes. They show a reduction in go cue directional precision, alongside a blunted sensitivity to stop error, go error and time costs. The learned embedding space reveals a continuous manifold in which children with higher ADHD scores are heterogeneously distributed, rather than forming distinct disorder clusters. This indicates that similar clinical characteristics can emerge from diverse combinations of computational mechanisms, supporting a dimensional perspective on neurodiversity. Our end-to-end framework can be extended to a broader range of cognitive tasks. It offers a scalable, theory-driven solution for analyzing large-scale behavioral data. Author summary: Inhibitory control is essential for adjusting thoughts and behavior and is often impaired in conditions like ADHD. Traditional assessments often oversimplify the underlying decision-making. We address this using a biologically grounded mathematical framework that separates how we perceive signals from how we execute control. This framework remains valid in diverse experimental designs where traditional models fail. To apply this complex model at scale, we develop a specialized machine learning approach (TeSBI) to reverse-engineer individual cognitive profiles. Analyzing data from over 3,000 children, we linked higher ADHD scores to subtle, consistent computational shifts: a reduction in go cue directional precision, a blunted sensitivity to the costs of making mistakes or taking too much time. These alterations remain robust regardless of sex, IQ, or medication status. Rather than forming a single disorder cluster, these children displayed diverse cognitive profiles, supporting a dimensional view of neurodiversity. By combining theory-driven cognitive modeling with scalable data-driven inference, our framework effectively captures complex inhibitory mechanisms and enables the precise analysis of large-scale behavioral datasets. This approach paves the way for more personalized strategies in computational psychiatry by recognizing the inherent heterogeneity within clinical characteristics.
SN - 1553-7358
ER -