Verbal Symbols Support Concrete but Enable Abstract Concept Formation: Evidence From Brain‐Constrained Deep Neural Networks

Concrete symbols (e.g., sun, run) can be learned in the context of objects and actions, thereby grounding their meaning in the world. However, it is controversial whether a comparable avenue to semantic learning exists for symbols (e.g., democracy). When we simulated the putative brain mechanisms of...

Full description

Saved in:
Bibliographic Details
Published in:Language learning Vol. 74; no. S1; pp. 258 - 295
Main Authors: Dobler, Fynn R., Henningsen‐Schomers, Malte R., Pulvermüller, Friedemann
Format: Journal Article
Language:English
Published: Hoboken Wiley Subscription Services, Inc 01.06.2024
Subjects:
abstract concepts
Brain
Brain mechanisms
Circuits
Concept formation
Conceptual semantics
deep neural networks
Democracy
grounding
Learning
Meaning
Neural networks
neurocomputational modeling
Neurolinguistics
Semantics
Semiotics
Symbolism
Symbols
ISSN:0023-8333, 1467-9922
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Concrete symbols (e.g., sun, run) can be learned in the context of objects and actions, thereby grounding their meaning in the world. However, it is controversial whether a comparable avenue to semantic learning exists for symbols (e.g., democracy). When we simulated the putative brain mechanisms of conceptual/semantic grounding using brain‐constrained deep neural networks, the learning of instances of concrete concepts outside of language contexts led to robust neural circuits generating substantial and prolonged activations. In contrast, the learning of instances of concepts yielded much reduced and only short‐lived activity. Crucially, when conceptual instances were learned in the context of wordforms, circuit activations became robust and long‐lasting for both concrete and meanings. These results indicate that, although the neural correlates of concrete conceptual representations can be built from grounding experiences alone, concept formation at the neurobiological level is enabled by and requires the correlated presence of linguistic forms. A one‐page Accessible Summary of this article in nontechnical language is freely available in the Supporting Information online and at https://oasis‐database.org.
Bibliography:Malte R. Henningsen‐Schomers
Accessible Summary
Fynn R. Dobler
https://oasis‐database.org
CRediT author statement
conceptualization; methodology; writing – original draft preparation; writing – review and editing; funding acquisition.
conceptualization; methodology; writing – original draft preparation; writing – review and editing.
Friedemann Pulvermüller
The data that support the findings of this study and the analysis code are available via the Open Science Framework at
https://osf.io/m8dg5
Supporting Information
online and at
.
investigation; formal analysis; writing – original draft preparation; writing – review and editing.
This work was supported by the European Research Council (ERC) through the Advanced Grant “Material constraints enabling human cognition, MatCo” (ERC‐2019‐ADG 883811) and by the Deutsche Forschungsgemeinschaft under Germany's Excellence Strategy through the Cluster of Excellence “Matters of Activity. Image Space Material” (DFG EXC 2025/1 – 390648296). We would like to thank the high‐performance computing service of Freie Universität Berlin and Martin Freyer and Phillip Krause for technical support. The authors declare no conflict of interest.
of this article in nontechnical language is freely available in the
bll_berlin
Author Twitter information
The handling editor for this manuscript was Guillaume Thierry.
A one‐page
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0023-8333
1467-9922
DOI:10.1111/lang.12646