# Habitual and Reflective Control in Hierarchical Predictive Coding

arXiv: https://arxiv.org/abs/2109.00866 · HTML: https://mlsanity.com/paper/2109.00866/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 34 references that are in the map.
- First posted on arXiv 2 Sep 2021.

## Builds on (most cited first, 3 of 3)

- [The free energy principle for action and perception: A mathematical review](https://mlsanity.com/paper/1705.09156/) (2017)
- [Thinking Fast and Slow: Optimization Decomposition Across Timescales](https://mlsanity.com/paper/1704.07785/) (2017)
- [Generative models as parsimonious descriptions of sensorimotor loops](https://mlsanity.com/paper/1904.12937/) (2019)

## Built on by (most cited first, 3 of 3)

- [Predictive Coding: Towards a Future of Deep Learning beyond Backpropagation?](https://mlsanity.com/paper/2202.09467/) (2022)
- [Backpropagation at the Infinitesimal Inference Limit of Energy-Based Models: Unifying Predictive Coding, Equilibrium Propagation, and Contrastive Hebbian Learning](https://mlsanity.com/paper/2206.02629/) (2022)
- [Preventing Deterioration of Classification Accuracy in Predictive Coding Networks](https://mlsanity.com/paper/2208.07114/) (2022)

## Abstract (arXiv)

In cognitive science, behaviour is often separated into two types. Reflexive control is habitual and immediate, whereas reflective is deliberative and time consuming. We examine the argument that Hierarchical Predictive Coding (HPC) can explain both types of behaviour as a continuum operating across a multi-layered network, removing the need for separate circuits in the brain. On this view, "fast" actions may be triggered using only the lower layers of the HPC schema, whereas more deliberative actions need higher layers. We demonstrate that HPC can distribute learning throughout its hierarchy, with higher layers called into use only as required.