Volume 22, Issue 1 (Spring 2020)                   Advances in Cognitive Sciences 2020, 22(1): 81-92 | Back to browse issues page


XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Haratiannezhadi A, Setayeshi S, Hatami J. Design an intelligent system based on a computational cognitive model using attention network task. Advances in Cognitive Sciences 2020; 22 (1) :81-92
URL: http://icssjournal.ir/article-1-928-en.html
1- PhD Student of Cognitive Modeling, Institute for Cognitive Science Studies, Tehran, Iran
2- Associate Professor of Physics, Amirkabir University of Technology, Tehran, Iran
3- Associate Professor Psychology, Institute for Cognitive Science Studies, Tehran, Iran
Abstract:   (2949 Views)
Introduction: The attention is a gateway for learning and a limited resource. Attention cognitive model helps to perceive and use it efficiently. This research aimed to find an intelligent model for a different level of attention.
Methods: Developing a cognitive model based on the attention network task and brain signals. The model builds on machine learning techniques. The initial research population consists of 92 adult volunteers who completed the Depression Anxiety Stress Scales test (DASS-21). Based on the test results, 31 subjects selected and invited to take Attention Network test and during the test, brain signals were captured for this purpose, Brain-computer Interface (BCI) was used and a model constructed based on different levels which used subject’s brain signal, reaction time and test result.
Results: Data were classified based on different machine learning methods such as Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Adaboost. The correct classification rate for these classifiers is 68, 90, and 87 percent.
Conclusion: The final model is selected based on the accuracy. So the KNN classifier has better generalization and it estimates test data better than other classifers. The desired Nero-cognitive model is based on the results and KNN classifiers are the best option for these types of cognitive models which is difficult to gather data and the dataset’s size are small.

Full-Text [PDF 1614 kb]   (966 Downloads)    
Type of Study: Research |
Received: 2019/07/16 | Accepted: 2019/11/28 | Published: 2020/06/10

Add your comments about this article : Your username or Email:
CAPTCHA

Send email to the article author


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Designed & Developed by : Yektaweb