» Articles » PMID: 32868772

Unconscious Reinforcement Learning of Hidden Brain States Supported by Confidence

Overview
Journal Nat Commun
Specialty Biology
Date 2020 Sep 2
PMID 32868772
Citations 13
Authors
Affiliations
Soon will be listed here.
Abstract

Can humans be trained to make strategic use of latent representations in their own brains? We investigate how human subjects can derive reward-maximizing choices from intrinsic high-dimensional information represented stochastically in neural activity. Reward contingencies are defined in real-time by fMRI multivoxel patterns; optimal action policies thereby depend on multidimensional brain activity taking place below the threshold of consciousness, by design. We find that subjects can solve the task within two hundred trials and errors, as their reinforcement learning processes interact with metacognitive functions (quantified as the meaningfulness of their decision confidence). Computational modelling and multivariate analyses identify a frontostriatal neural mechanism by which the brain may untangle the 'curse of dimensionality': synchronization of confidence representations in prefrontal cortex with reward prediction errors in basal ganglia support exploration of latent task representations. These results may provide an alternative starting point for future investigations into unconscious learning and functions of metacognition.

Citing Articles

Touch-driven advantages in reaction time but not in performance in a cross-sensory comparison of reinforcement learning.

Sun W, Ripp I, Borrmann A, Moll M, Fairhurst M Heliyon. 2025; 11(1):e41330.

PMID: 39839521 PMC: 11748724. DOI: 10.1016/j.heliyon.2024.e41330.


Time-dependent neural arbitration between cue associative and episodic fear memories.

Cortese A, Ohata R, Alemany-Gonzalez M, Kitagawa N, Imamizu H, Koizumi A Nat Commun. 2024; 15(1):8706.

PMID: 39433735 PMC: 11494204. DOI: 10.1038/s41467-024-52733-4.


Decoding and modifying dynamic attentional bias in gaming disorder.

Oka T, Kubo T, Kobayashi N, Murakami M, Chiba T, Cortese A Philos Trans R Soc Lond B Biol Sci. 2024; 379(1915):20230090.

PMID: 39428882 PMC: 11491851. DOI: 10.1098/rstb.2023.0090.


Mechanisms of brain self-regulation: psychological factors, mechanistic models and neural substrates.

Sitaram R, Sanchez-Corzo A, Vargas G, Cortese A, El-Deredy W, Jackson A Philos Trans R Soc Lond B Biol Sci. 2024; 379(1915):20230093.

PMID: 39428875 PMC: 11491850. DOI: 10.1098/rstb.2023.0093.


Interaction between the prefrontal and visual cortices supports subjective fear.

Taschereau-Dumouchel V, Cote M, Manuel S, Valevicius D, Cushing C, Cortese A Philos Trans R Soc Lond B Biol Sci. 2024; 379(1908):20230245.

PMID: 39005034 PMC: 11444220. DOI: 10.1098/rstb.2023.0245.


References
1.
Moutard C, Dehaene S, Malach R . Spontaneous Fluctuations and Non-linear Ignitions: Two Dynamic Faces of Cortical Recurrent Loops. Neuron. 2015; 88(1):194-206. DOI: 10.1016/j.neuron.2015.09.018. View

2.
Pessiglione M, Petrovic P, Daunizeau J, Palminteri S, Dolan R, Frith C . Subliminal instrumental conditioning demonstrated in the human brain. Neuron. 2008; 59(4):561-7. PMC: 2572733. DOI: 10.1016/j.neuron.2008.07.005. View

3.
Seitz A, Kim D, Watanabe T . Rewards evoke learning of unconsciously processed visual stimuli in adult humans. Neuron. 2009; 61(5):700-7. PMC: 2683263. DOI: 10.1016/j.neuron.2009.01.016. View

4.
Seitz A, Watanabe T . Psychophysics: Is subliminal learning really passive?. Nature. 2003; 422(6927):36. DOI: 10.1038/422036a. View

5.
Bechara A, Damasio H, Tranel D, Damasio A . Deciding advantageously before knowing the advantageous strategy. Science. 1997; 275(5304):1293-5. DOI: 10.1126/science.275.5304.1293. View