A neural network model for the study of the relationship between language processing, topographical representation and γ-band synchronization
-
1
Universiyt of Bologna, Italy
Introduction – Synchronisation of neural activity in the gamma-band is assumed to play a relevant role in several high cognitive tasks, such as object recognition, associative memory, or music perception. It is generally thought that temporal correlation is exploited by the brain for the solution of the binding and segmentation problem, allowing activity in disparate brain areas to be grouped together. In particular, recent experiments suggest that a distributed neural system may be at the basis of the processing of language (Gonzales et al., 2006). Pulvermüller (2005) hypothesized that topographically organized regions of the cortex directly connect different features of a word. Aim of the present work was to develop a mathematical model of neural oscillators, to analyze the possible mechanisms relating language, perception and gamma-band activity via computer simulations techniques.
Method – The model includes four areas of Wilson-Cowan oscillators (“feature areas”). Each area is devoted to the representation of a particular feature of an object, according to a topological representation which implements a similarity rule. Hence, an object is completely represented by the synchronized activation of four neural groups in four areas. Furthermore, a fifth area (without topological organization) is devoted to the representation of words (“semantic area”). Synapses linking the four “feature areas” and the “semantic area” are initially set to zero, and subsequently trained via a time dependent Hebbian rule.
Results – The simulations consisted of a training phase and a recovery phase. During the training phase, several objects (represented by four features each) are separately given to the network together with the corresponding word, so as to build long-range synapses. The recovery phase consisted in two different sets of trials: i) Simulations were performed giving various simultaneous objects to the network (from 1 to 4), with some missing and/or modified properties. The results suggest that the network can reconstruct objects, and segment them from the other simultaneously present objects, even in case of deteriorated information. Moreover, the reconstructed object automatically evokes activity in the semantic network, which codes for the correct word. ii) Simulations were performed by giving 2 or 3 words simultaneously as input to the semantic area. The results show that the presence of the words can evoke activity in the four “feature areas”. Moreover, as shown in Fig. 1, the activity in feature areas coding for different objects oscillate out of phase, achieving a correct segmentation.
Conclusions - The model represents an original attempt to link binding, segmentation, gamma-band synchronization and language processing into a single theoretical structure. In particular, according to recent data, it stresses the existence of a strict relationship between word coding and perception coding, and supports the hypothesis that a semantic process may engage many cortical areas.
Conference:
10th International Conference on Cognitive Neuroscience, Bodrum, Türkiye, 1 Sep - 5 Sep, 2008.
Presentation Type:
Poster Presentation
Topic:
Brain Electrical Oscillations in Cognition
Citation:
Cuppini
C,
Magosso
E and
Ursino
M
(2008). A neural network model for the study of the relationship between language processing, topographical representation and γ-band synchronization.
Conference Abstract:
10th International Conference on Cognitive Neuroscience.
doi: 10.3389/conf.neuro.09.2009.01.120
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
03 Dec 2008;
Published Online:
03 Dec 2008.
*
Correspondence:
Elisa Magosso, Universiyt of Bologna, Cesena, Italy, elisa.magosso@unibo.it