1. RNP
  2. Motor Learning
  3. Predict, err, recalibrate: how the system learns
  4. Active inference (Friston 2010)

Motor LearningSeries 6/9

SeriesPredict, err, recalibrate: how the system learns6 of 9

Active inference (Friston 2010)

Friston's 2010 work on active inference links to NIT and to the paradigmatic tension at the heart of RNP.

LabO RNP·3 min read·reviewed on

In brief. In RNP, Friston's active inference (2010) refers to the scientific lineage linked to NIT and to RNP's central paradigmatic tension. In 2010, Karl Friston formalized the free-energy principle and active inference as a unified theoretical framework for brain function (The free-energy principle: a unified brain theory?). It is the founding lineage of contemporary predictive coding, which links sensory prediction, prediction error and active behavior.

Why does Friston 2010 matter to the human movement professional?

Friston 2010 is one of the core contemporary lineages behind NIT and the paradigmatic tension specific to RNP. Its practical value comes from the depth of its unified theoretical framework (the free-energy principle and active inference), which links sensory prediction, prediction error and active behavior. RNP draws on Friston 2010 as the contemporary lineage behind the predictive coding side of NIT and behind the paradigmatic tension with ecological dynamics.

Friston brings several distinctive contributions to RNP. The free-energy principle: the central nervous system minimizes a free-energy function tied to sensory surprise. Active inference: action aims to make the world match predictions. A cortical prediction hierarchy: each level predicts the levels below it and exchanges prediction errors with neighboring levels. Consistency with sensory reweighting: context-dependent Bayesian weighting.

RNP integrates Friston at several points. See also: Clark: Predictive Processing and Surfing Uncertainty (2013-2016), Davids, Button & Bennett: Dynamics of Skill Acquisition (2008).

Friston's active inference (2010) is the cardinal contemporary predictive coding lineage linked to NIT and to the paradigmatic tension specific to RNP.

How Friston applies in practice

Friston applies to reading sensory prediction and the prediction error that goes with it, to adaptive Bayesian weighting, and to the cognitive-motor dual-task NIT modality.

Reading sensory prediction. You read the competitive or functional profile in terms of sensory prediction (anticipating the expected sensory flow) and observable prediction error (the gap between what was predicted and what is actually received). You name the prediction error in profiles where it proves relevant.

Adaptive Bayesian weighting. You bring Friston's perspective into sensory reweighting across the three systems of posturology. You read that weighting as Bayesian weighting that depends on context and predictions.

Cognitive-motor dual-task NIT. You integrate cognitive-motor NIT into the T-S-R (Test–Stimulation–Retest) cycle using Friston's predictive perspective: dual tasks challenge sensory prediction under cognitive load.

The RNP training space, free for life

Get access to RNP training, no strings attached.

Welcome! Introduce yourself

SecureNo cardNo password

By continuing, you agree to our Terms of Use and our Privacy Policy.

What you hear

The first misconception comes from the abstract mathematical theory. You hear "Friston is for mathematicians, not for practitioners on the ground." Active inference has clear practical consequences: sensory reweighting, the cognitive-motor NIT modality, and reading prediction error.

The second misconception comes from a radical opposition to ecological dynamics. You hear "predictive coding and ecological dynamics are incompatible." The paradigmatic tension specific to RNP combines the two approaches sequentially, according to the functional levels of NIT.

The third misconception comes from reducing everything to the brain. You hear "Friston only talks about the brain, not the body." Active inference treats motor action as a way of minimizing prediction error. The moving body is central to Friston's framework.

Friston's active inference (2010) is the central contemporary predictive coding lineage linked to NIT.

Further reading

RNP's paradigmatic tension: Clark: Predictive Processing and Surfing Uncertainty (2013-2016) · Davids, Button & Bennett: Dynamics of Skill Acquisition (2008)

Scientific lineages: Bernstein 1967 · Gibson 1979 · Newell's Constraints Model (1986) · Schmidt & Lee 2019

Posturology systems: Vestibular system · RNP somatosensory system

Going further

Friston 2010 belongs to the contemporary predictive processing tradition and underpins the Bayesian formalization of brain function.

In 2010, Karl Friston published the unified theoretical framework of the free-energy principle and active inference (The free-energy principle: a unified brain theory?, Nature Reviews Neuroscience). His formalization shaped contemporary predictive processing.

Later work in the Friston lineage has extended active inference to motor control, perception, emotion and cognition. RNP connects this lineage to NIT and to the paradigmatic tension specific to RNP.

Scope: a functional, non-medical approach. It complements healthcare follow-up; it does not replace it. Friston's active inference (2010) refers to a contemporary scientific lineage linked to NIT.

€0Your free RNP PathsStart