Rethinking Predictive Processing.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Predictive coding theory suggests the brain uses internal models to predict sensory input and updates them via prediction errors.
• Neurophysiological studies show patterns supporting this framework, but definitions and evidence vary.
• Questions remain regarding the validity and scope of predictive processing.

Purpose of the Study:

• To provide a historical overview of the predictive processing framework.
• To evaluate empirical support, focusing on sensory prediction error signals.
• To clarify the information encoded by these signals and explore alternative accounts.

Main Methods:

• Literature review of neurophysiological studies on predictive coding.
• Analysis of sensory prediction error signals and their computational underpinnings.
• Discussion of alternative models for information encoding in prediction errors.

Main Results:

• Empirical evidence for predictive coding is mixed, with varied definitions and interpretations.
• Sensory prediction error signals may reflect diverse computations despite similar response patterns.
• Alternative accounts exist for how these signals encode information.

Conclusions:

• Clarifying the precise information represented by sensory prediction error signals is critical.
• Further research is needed to refine the predictive processing framework.
• Understanding neuronal algorithms for perception and cognition requires advancing predictive coding research.