Embodiment Enables Non-Predictive Ways of Coping with Self-Caused Sensory Stimuli

TL;DR

This study demonstrates through a computational model that embodied systems can employ non-predictive strategies to cope with self-caused sensory stimuli, challenging the traditional predictive attenuation hypothesis.

Contribution

The paper introduces a computational model showing that embodied systems can use non-predictive methods to manage self-caused sensory inputs, highlighting alternative coping mechanisms.

Findings

Solutions regulating self-caused sensory inputs often emerge in simple embodied systems.

Some systems rely on self-caused inputs rather than predictive filtering.

Non-predictive strategies can be effective in managing sensory stimuli in embodied agents.

Abstract

Living systems process sensory data to facilitate adaptive behaviour. A given sensor can be stimulated as the result of internally driven activity, or by purely external (environmental) sources. It is clear that these inputs are processed differently - have you ever tried tickling yourself? The canonical explanation of this difference is that when the brain sends a signal that would result in motor activity, it uses a copy of that signal to predict the sensory consequences of the resulting motor activity. The predicted sensory input is then subtracted from the actual sensory input, resulting in attenuation of the stimuli. To critically evaluate this idea, and investigate when non-predictive solutions may be viable, we implement a computational model of a simple embodied system with self-caused sensorimotor dynamics, and analyse how controllers successfully accomplish tasks in this model. We find that in these simple systems, solutions that regulate behaviour to control self-caused sensory inputs tend to emerge, rather than solutions which predict and filter out self-caused inputs. In some cases, solutions depend on the presence of these self-caused inputs.