Modelling Peri-Perceptual Brain Processes in a Deep Learning Spiking Neural Network Architecture

Doborjeh, Zohreh Gholami; Kasabov, Nikola; Doborjeh, Maryam Gholami; Sumich, Alexander

doi:10.1038/s41598-018-27169-8

Cited by 31 publications

(5 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The P200 and P300 appear to reflect distinct CES processes, at pre-and post-perceptual stages, during the 3-back task performance. Whilst less is known about the functional significance of the P200 compared to an extensive literature on P300, it has been proposed as reflecting processes underpinning perceptual matching and stimulus classification, such as stimulus detection, evaluation, storage, and encoding ( Crowley & Colrain, 2004;Gholami Doborjeh, Kasabov, Gholami Doborjeh, & Sumich, 2018;Potts, 2004) that facilitate the post-perceptual processing as reflected by the P300 wave ( Chen et al , 2008) . Other studies have implicated P200 mechanisms in stimulus switching ( Karayanidis & Michie, 1997;Nicholson, Karayanidis, Bumak, Poboka, & Michie, 2006;Stefanics, Kremláček, & Czigler, 2014) .…”

Section: Discussionmentioning

confidence: 99%

Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model

Wongupparaj

Sumich

Wickens

et al. 2018

Biological Psychology

Self Cite

View full text Add to dashboard Cite

A robust relationship between working memory (WM) and general intelligence (g) has been well established. Nevertheless, explanations for this relationship in terms of underlying neurocognitive processes are still inadequate. This study addresses this issue using an individual differences approach in which Central Executive System (CES) and Short-Term Storage (STS) c o m p o n e n t s of WM are measured comprehensively and examined for their relationship with g via event-related potentials components (P200 and P300) as mediators. Participants (n=115) completed tests of the WM, CES and STS, as well as g. P200 and P300 components were recorded d u r i n g 3-back WM task performance. Structural equation modelling showed significant negative associations between the P200 latency for target stimuli a n d CES shifting processes, and between the P300 amplitude for target stimuli and CES inhibition and updating processes. The relationship between CES processes and g was mediated in a localized fashion by the P300 amplitude. These findings further support the notion that the CES has a multidimensional structure and, importantly, reveal that the inhibition and updating functions of the CES are crucial in explaining the relationship between WM and g. Negative relations between ERP indices (P200 latency and P300 amplitude for target stimuli) and g support a neural efficiency hypothesis related to high intelligence.

show abstract

Section: Discussionmentioning

confidence: 99%

Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model

Wongupparaj

Sumich

Wickens

et al. 2018

Biological Psychology

Self Cite

View full text Add to dashboard Cite

show abstract

“…7. Recently developed neuromorphic computing techniques suited to spatio-temporal data and specialized deep learning methods for data visualization of cortical connectivity (Kasabov et al, 2016;Doborjeh et al, 2018Doborjeh et al, , 2019 represent promising future approaches. For example, the evolving connectivity of multiple subjects before and after IBS can be measured and visualized in a brain-inspired spiking neural network (SNN) models such as NeuCube (Kasabov, 2014(Kasabov, , 2018, to trace and understand the dynamics and effects of language communication.…”

Section: Apart Frommentioning

confidence: 99%

What has social neuroscience learned from hyperscanning studies of spoken communication? A systematic review

Kelsen

Sumich

Kasabov

et al. 2022

Neuroscience & Biobehavioral Reviews

Self Cite

View full text Add to dashboard Cite

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

show abstract

“…The NeuCube model has been successfully applied and evaluated in several studies. In the context of EEG signal classification, the system was tested to model peri-perceptual brain processes from EEG signal [29]. Doborjeh et al tested the model in a similar study on fMRI data [30].…”

Section: Introductionmentioning

confidence: 99%

Classification and regression of spatio-temporal signals using NeuCube and its realization on SpiNNaker neuromorphic hardware

Behrenbeck¹,

Tayeb²,

Bhiri³

et al. 2019

J. Neural Eng.

Self Cite

View full text Add to dashboard Cite

Objective. The objective of this work is to use the capability of spiking neural networks to capture the spatio-temporal information encoded in timeseries signals and decode them without the use of hand-crafted features and vectorbased learning and the realization of the spiking model on low-power neuromorphic hardware. Approach. The NeuCube spiking model was used to classify different grasp movements directly from raw surface electromyography signals (sEMG), the estimations of the applied finger forces as well as the classification of two motor imagery movements from raw electroencephalography (EEG). In a parallel investigation, the designed spiking decoder was implemented on SpiNNaker neuromorphic hardware, which allows low-energy real-time processing. Results. Experimental results reveal a better classification accuracy using the NeuCube model compared to traditional machine learning methods. For sEMG classification, we reached a training accuracy of 85% and a test accuracy of 84.8%, as well as less than 19% of relative root mean square error (rRMSE) when estimating finger forces from six subjects. For the EEG classification, a mean accuracy of 75% was obtained when tested on raw EEG data from nine subjects from the existing 2b dataset from "BCI Competition IV". Significance. This work provides a proof of concept for a successful implementation of the NeuCube spiking model on the SpiNNaker neuromorphic platform for raw sEMG and EEG decoding, which could chart a route ahead for a new generation of portable closedloop and low-power neuroprostheses.

show abstract

Modelling Peri-Perceptual Brain Processes in a Deep Learning Spiking Neural Network Architecture

Cited by 31 publications

References 64 publications

Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model

Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model

What has social neuroscience learned from hyperscanning studies of spoken communication? A systematic review

Classification and regression of spatio-temporal signals using NeuCube and its realization on SpiNNaker neuromorphic hardware

Contact Info

Product

Resources

About