From self-organizing systems to subjective temporal extension

Area of Science:

• Cognitive Science
• Neuroscience
• Physics

Background:

• Subjective temporality's emergence is not fully understood.
• The free energy principle offers a framework for self-organizing systems.
• Information geometry provides tools for formalizing complex dynamics.

Purpose of the Study:

• To formalize subjective temporal extension using the free energy principle and information geometry.
• To computationally model temporal inference via active inference.
• To explain the emergence of subjective time's
• width
• from self-modeling architectures.

Main Methods:

• Adopting the free energy principle perspective.
• Utilizing information geometry for formalization.
• Simulating a deep parametric generative model of temporal inference with active inference.

Main Results:

• Two biases in time perception emerged: intentional binding and altered experience in meditative states.
• The model computationally explains numerous alterations in subjective temporal experience.
• The self-simulational theory of temporal extension (SST) provides a unified computational account.

Conclusions:

• SST, grounded in the free energy principle, explains subjective time's variable "width".
• The computational mechanism accounts for subjective time seeming to fly or drag.
• The model is deployable for synthetic applications in understanding temporal perception.