New theoretical framework for perceptual-motion integration based on active simulation function

Perception and motion are the basis of all biological activities and are strongly connected. This relationship is referred to as “perceptual-motion integration,” and it is crucial for the ability of living organisms to adapt their behaviour to the environment. Many experimental and theoretical studies of the mechanism of perceptual-motion integration have been conducted, but the overall operation of the mechanism has not yet been clarified. In this article, we propose a new theoretical framework of perceptual-motion integration in living organisms and a mathematical model within this framework by extending the simulation function, which is a fundamental form of the prediction function. This model describes the dynamics of a perception operator as an algebraic constraint condition that the perception of a living organism should satisfy in simulating the sensory information received from the environment based on an internal model of self-motion. According to the model, a living organism actively perceives and simulates the environment using such an internal model, thereby eliciting a time-invariant quantity that characterizes the environment in terms of self-states that serve as a moving frame. From a biological perspective, living organisms must identify cognitively stable features in the ever-changing environment. Such features are represented by time-invariant quantities that appear when living organisms actively perceive the environment and make it possible for living organisms to adapt to it. Furthermore, the derivation of such time-invariant quantities indicates that this theoretical framework can uniformly describe biologically interesting phenomena such as the mirroring of the intentional motions of others and body extension during tool use.