Towards a computational (neuro)phenomenology of mental action: modelling meta-awareness and attentional control with deep-parametric active inference
PsyArXiv June 10, 2020 preprint DOI: 10.31234/osf.io/5jh3c
Summary
This theoretical paper develops a computational model of mental action—the deliberate control of attention and awareness—using deep-parametric active inference. It proposes that meta-awareness, or the ability to monitor and regulate one's own cognitive processes, can be formalized as a higher-order inference mechanism that shapes attentional policies. The model integrates phenomenological insights with computational neuroscience to explain how agents can exert top-down control over their perceptual engagement, offering a framework for understanding mindfulness, self-regulation, and disorders of attention.
Study at a glance
| Design | theoretical or philosophical paper |
|---|---|
| Key finding | Meta-awareness and attentional control can be modeled as higher-order inference within a deep-parametric active inference framework, bridging phenomenology and computational neuroscience. |
Abstract
Towards a computational (neuro)phenomenology of mental action: modelling meta-awareness and attentional control with deep-parametric active inference