PERCEIVED CONTROL OVER BREATHING PERTURBATION DEPENDS ON INTEROSENSORY INPUT AND PREDICTION ERROR

Allostatic control is fundamental for survival. A failure of (prospective) body regulation is indicated by prolonged dyshomeostasis. This persistent imbalance has been associated with impaired mental and physical health through the metacognitive diagnosis of loss of control over bodily states or low allostatic self-efficacy. It is unclear though how the brain monitors control over bodily states. The Allostatic Self-Efficacy Theory (ASE; Stephan et al. 2016) posits that this is achieved by the brain’s monitoring of prediction errors (PEs) concerning bodily states. However, this suggestion has not been tested on data from humans.
In this work, we introduce the Respiratory Metacognition of Control Task (RMCT), a novel behavioural paradigm for manipulation of perceived control over breathing by changing inspiratory resistive loads. We developed and tested a trial-by-trial generative model of perceived control over breathing perturbations in the RMCT using data from 50 healthy individuals in a pre-registered analysis. Our model explains perceived control over breathing perturbations as a function of sensory outcomes (breathing without/with an inspiratory resistive load) and signed PEs quantifying the surprise associated with these outcomes – suggesting a combination of the two is crucial.
Our task and model enable subject-specific parameter estimation providing a mechanistic explanation for inter-individual differences in perceived control over inspiratory resistive loads. It may serve as a tool for patient stratification and to identify potential treatment targets for therapeutic interventions in conditions with altered perceived control over the body, such as chronic fatigue or depression.