Dresden 2026 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DD: Fachverband Didaktik der Physik

DD 28: Poster – Neue Konzepte

DD 28.2: Poster

Tuesday, March 10, 2026, 14:00–16:00, P5

Predictive Coding in the Human Brain and its possible applications for a Shared Universe Engine — •André Bresges — University of Cologne, Albertus-Magnus-Platz, 50931 Cologne

Predictive coding theory proposes the brain isn't a passive receiver but an active prediction machine, constantly generating internal models to anticipate sensory input, minimizing surprise by only processing the "prediction errors" (mismatches) that occur. This leads to efficient perception, learning, and action by updating its world model to better match reality. Predictive Coding is a Bayesian brain hypothesis where top-down predictions (expectations from past experience) meet bottom-up sensory data, with errors signaling the need for adjustment of the model. Predictive Coding explains consciousness not as a static model, but dynamically as *the brain's best guess about the world*. Educational Methods like POE (Predict Observe Explain) are in good alignment with this theoretical framework. We argue that predictive coding might also be a good theory to build a Shared Universe Engine upon because of its: Efficiency: Reduces redundant processing, saving cognitive resources. Precision: The Brain aims to minimize prediction errors. Speed: Unifies perception, cognition, and action (sensor direction and tuning) under a single principle.

Keywords: Astrophysics; Dynaverse; Excellence Cluster; Predict Observe Explain; Artificial Intelligence