Nonmodular architectures of cognitive systems based on active inference
Manuel Baltieri, Christopher L. Buckley
arXiv Preprint Archive March 22, 2019 Peer reviewed via arXiv
Summary
This paper argues that even modern closed-loop models in cognitive science maintain a strong separation between motor and perceptual functions, echoing the modularity of traditional input/output views. The authors present a minimal sensorimotor model based on the separation principle of control theory and show its limitations when external forces, such as environmental perturbations or interference from other agents, are not modeled. As an alternative, they propose a nonmodular architecture grounded in active inference, demonstrating its robustness to unknown external inputs and showing that, in linear models, this mechanism is equivalent to integral control.
Study at a glance
| Design | theoretical or philosophical paper |
|---|---|
| Key finding | A nonmodular architecture based on active inference is robust to unknown external inputs, and in linear models this mechanism is equivalent to integral control. |
Abstract
In psychology and neuroscience it is common to describe cognitive systems as input/output devices where perceptual and motor functions are implemented in a purely feedforward, open-loop fashion. On this view, perception and action are often seen as encapsulated modules with limited interaction between them. While embodied and enactive approaches to cognitive science have challenged the idealisation of the brain as an input/output device, we argue that even the more recent attempts to model systems using closed-loop architectures still heavily rely on a strong separation between motor and perceptual functions. Previously, we have suggested that the mainstream notion of modularity strongly resonates with the separation principle of control theory. In this work we present a minimal model of a sensorimotor loop implementing an architecture based on the separation principle. We link this to popular formulations of perception and action in the cognitive sciences, and show its limitations when, for instance, external forces are not modelled by an agent. These forces can be seen as variables that an agent cannot directly control, i.e., a perturbation from the environment or an interference caused by other agents. As an alternative approach inspired by embodied cognitive science, we then propose a nonmodular architecture based on the active inference framework. We demonstrate the robustness of this architecture to unknown external inputs and show that the mechanism with which this is achieved in linear models is equivalent to integral control.