Domain generality versus modality specificity: The paradox of statistical learning

Ram Frost, Blair Armstrong, Noam Siegelman, Morten Hyllekvist Christiansen

Research output: Contribution to journalReviewResearchpeer-review

Abstract

Statistical learning (SL) is typically considered to be a domain-general mechanism by which cognitive systems discover the underlying distributional properties of the input. However, recent studies examining whether there are commonalities in the learning of distributional information across different domains or modalities consistently reveal modality and stimulus specificity. Therefore, important questions are how and why a hypothesized domain-general learning mechanism systematically produces such effects. Here, we offer a theoretical framework according to which SL is not a unitary mechanism, but a set of domain-general computational principles that operate in different modalities and, therefore, are subject to the specific constraints characteristic of their respective brain regions. This framework offers testable predictions and we discuss its computational and neurobiological plausibility.

Original languageEnglish
JournalTrends in Cognitive Sciences
Volume19
Issue number3
Pages (from-to)117-125
ISSN1364-6613
DOIs
Publication statusPublished - 2015

Keywords

  • Domain-general mechanisms
  • Modality specificity
  • Neurobiologically plausible models
  • Statistical learning
  • Stimulus specificity

Cite this

Frost, Ram ; Armstrong, Blair ; Siegelman, Noam ; Christiansen, Morten Hyllekvist. / Domain generality versus modality specificity : The paradox of statistical learning. In: Trends in Cognitive Sciences. 2015 ; Vol. 19, No. 3. pp. 117-125.
@article{a8da46c29e944f3dbe32813a095b7858,
title = "Domain generality versus modality specificity: The paradox of statistical learning",
abstract = "Statistical learning (SL) is typically considered to be a domain-general mechanism by which cognitive systems discover the underlying distributional properties of the input. However, recent studies examining whether there are commonalities in the learning of distributional information across different domains or modalities consistently reveal modality and stimulus specificity. Therefore, important questions are how and why a hypothesized domain-general learning mechanism systematically produces such effects. Here, we offer a theoretical framework according to which SL is not a unitary mechanism, but a set of domain-general computational principles that operate in different modalities and, therefore, are subject to the specific constraints characteristic of their respective brain regions. This framework offers testable predictions and we discuss its computational and neurobiological plausibility.",
keywords = "Domain-general mechanisms, Modality specificity, Neurobiologically plausible models, Statistical learning, Stimulus specificity",
author = "Ram Frost and Blair Armstrong and Noam Siegelman and Christiansen, {Morten Hyllekvist}",
year = "2015",
doi = "10.1016/j.tics.2014.12.010",
language = "English",
volume = "19",
pages = "117--125",
journal = "Trends in Cognitive Sciences",
issn = "1364-6613",
publisher = "Elsevier",
number = "3",

}

Domain generality versus modality specificity : The paradox of statistical learning. / Frost, Ram; Armstrong, Blair; Siegelman, Noam; Christiansen, Morten Hyllekvist.

In: Trends in Cognitive Sciences, Vol. 19, No. 3, 2015, p. 117-125.

Research output: Contribution to journalReviewResearchpeer-review

TY - JOUR

T1 - Domain generality versus modality specificity

T2 - The paradox of statistical learning

AU - Frost, Ram

AU - Armstrong, Blair

AU - Siegelman, Noam

AU - Christiansen, Morten Hyllekvist

PY - 2015

Y1 - 2015

N2 - Statistical learning (SL) is typically considered to be a domain-general mechanism by which cognitive systems discover the underlying distributional properties of the input. However, recent studies examining whether there are commonalities in the learning of distributional information across different domains or modalities consistently reveal modality and stimulus specificity. Therefore, important questions are how and why a hypothesized domain-general learning mechanism systematically produces such effects. Here, we offer a theoretical framework according to which SL is not a unitary mechanism, but a set of domain-general computational principles that operate in different modalities and, therefore, are subject to the specific constraints characteristic of their respective brain regions. This framework offers testable predictions and we discuss its computational and neurobiological plausibility.

AB - Statistical learning (SL) is typically considered to be a domain-general mechanism by which cognitive systems discover the underlying distributional properties of the input. However, recent studies examining whether there are commonalities in the learning of distributional information across different domains or modalities consistently reveal modality and stimulus specificity. Therefore, important questions are how and why a hypothesized domain-general learning mechanism systematically produces such effects. Here, we offer a theoretical framework according to which SL is not a unitary mechanism, but a set of domain-general computational principles that operate in different modalities and, therefore, are subject to the specific constraints characteristic of their respective brain regions. This framework offers testable predictions and we discuss its computational and neurobiological plausibility.

KW - Domain-general mechanisms

KW - Modality specificity

KW - Neurobiologically plausible models

KW - Statistical learning

KW - Stimulus specificity

U2 - 10.1016/j.tics.2014.12.010

DO - 10.1016/j.tics.2014.12.010

M3 - Review

VL - 19

SP - 117

EP - 125

JO - Trends in Cognitive Sciences

JF - Trends in Cognitive Sciences

SN - 1364-6613

IS - 3

ER -