A common probabilistic framework for perceptual and statistical learning

József Fiser, Gábor Lengyel

Research output: Review article

Abstract

System-level learning of sensory information is traditionally divided into two domains: perceptual learning that focuses on acquiring knowledge suitable for fine discrimination between similar sensory inputs, and statistical learning that explores the mechanisms that develop complex representations of unfamiliar sensory experiences. The two domains have been typically treated in complete separation both in terms of the underlying computational mechanisms and the brain areas and processes implementing those computations. However, a number of recent findings in both domains call in question this strict separation. We interpret classical and more recent results in the general framework of probabilistic computation, provide a unifying view of how various aspects of the two domains are interlinked, and suggest how the probabilistic approach can also alleviate the problem of dealing with widely different types of neural correlates of learning. Finally, we outline several directions along which our proposed approach fosters new types of experiments that can promote investigations of natural learning in humans and other species.

Original languageEnglish
Pages (from-to)218-228
Number of pages11
JournalCurrent Opinion in Neurobiology
Volume58
DOIs
Publication statusPublished - okt. 2019

Fingerprint

Learning
Brain

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

A common probabilistic framework for perceptual and statistical learning. / Fiser, József; Lengyel, Gábor.

In: Current Opinion in Neurobiology, Vol. 58, 10.2019, p. 218-228.

Research output: Review article

@article{a25e91e592434a00bb65dfee734c61e3,
title = "A common probabilistic framework for perceptual and statistical learning",
abstract = "System-level learning of sensory information is traditionally divided into two domains: perceptual learning that focuses on acquiring knowledge suitable for fine discrimination between similar sensory inputs, and statistical learning that explores the mechanisms that develop complex representations of unfamiliar sensory experiences. The two domains have been typically treated in complete separation both in terms of the underlying computational mechanisms and the brain areas and processes implementing those computations. However, a number of recent findings in both domains call in question this strict separation. We interpret classical and more recent results in the general framework of probabilistic computation, provide a unifying view of how various aspects of the two domains are interlinked, and suggest how the probabilistic approach can also alleviate the problem of dealing with widely different types of neural correlates of learning. Finally, we outline several directions along which our proposed approach fosters new types of experiments that can promote investigations of natural learning in humans and other species.",
author = "J{\'o}zsef Fiser and G{\'a}bor Lengyel",
year = "2019",
month = "10",
doi = "10.1016/j.conb.2019.09.007",
language = "English",
volume = "58",
pages = "218--228",
journal = "Current Opinion in Neurobiology",
issn = "0959-4388",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A common probabilistic framework for perceptual and statistical learning

AU - Fiser, József

AU - Lengyel, Gábor

PY - 2019/10

Y1 - 2019/10

N2 - System-level learning of sensory information is traditionally divided into two domains: perceptual learning that focuses on acquiring knowledge suitable for fine discrimination between similar sensory inputs, and statistical learning that explores the mechanisms that develop complex representations of unfamiliar sensory experiences. The two domains have been typically treated in complete separation both in terms of the underlying computational mechanisms and the brain areas and processes implementing those computations. However, a number of recent findings in both domains call in question this strict separation. We interpret classical and more recent results in the general framework of probabilistic computation, provide a unifying view of how various aspects of the two domains are interlinked, and suggest how the probabilistic approach can also alleviate the problem of dealing with widely different types of neural correlates of learning. Finally, we outline several directions along which our proposed approach fosters new types of experiments that can promote investigations of natural learning in humans and other species.

AB - System-level learning of sensory information is traditionally divided into two domains: perceptual learning that focuses on acquiring knowledge suitable for fine discrimination between similar sensory inputs, and statistical learning that explores the mechanisms that develop complex representations of unfamiliar sensory experiences. The two domains have been typically treated in complete separation both in terms of the underlying computational mechanisms and the brain areas and processes implementing those computations. However, a number of recent findings in both domains call in question this strict separation. We interpret classical and more recent results in the general framework of probabilistic computation, provide a unifying view of how various aspects of the two domains are interlinked, and suggest how the probabilistic approach can also alleviate the problem of dealing with widely different types of neural correlates of learning. Finally, we outline several directions along which our proposed approach fosters new types of experiments that can promote investigations of natural learning in humans and other species.

UR - http://www.scopus.com/inward/record.url?scp=85073968595&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073968595&partnerID=8YFLogxK

U2 - 10.1016/j.conb.2019.09.007

DO - 10.1016/j.conb.2019.09.007

M3 - Review article

AN - SCOPUS:85073968595

VL - 58

SP - 218

EP - 228

JO - Current Opinion in Neurobiology

JF - Current Opinion in Neurobiology

SN - 0959-4388

ER -