Working-memory evaluation based on EEG signals during n-back tasks

J Integr Neurosci. 2018;17(3-4):695-707. doi: 10.3233/JIN-180096.

Abstract

Working memory is the cognitive process of receiving, processing, and communicating information. Early evaluation and training may help to prevent a marked decline in working-memory ability. The aim of this study was to establish an n-back task system for objectively evaluating the working-memory ability based on the θ power, the γ power, and the degree of θ-γ synchronization. Our experiments were divided into four memory loads (1-back to 4-back) and further divided into digital and matrix stimulation modes. The γ band was divided into three components (γ1: 30-50 Hz, γ2: 50-70 Hz, and γ3: 70-90 Hz). The study recruited 23 healthy young subjects. The obtained results indicate that accuracy rates might be inappropriate for assessing the working memory. The θ power relates to working-memory recognition while the γ power varied with the memory load. Besides, a higher load level should be more appropriate for assessing the working-memory ability. Finally, the number of θ-γ3 couplings was highest between the frontal area and Pz and O2. In summary, the results of this study may be useful in multiple evaluations of working memory, and may lead to a wider variety of clinical applications in the future.

Keywords: EEG; cross-frequency coupling; gamma band; n-back task; theta band; working memory.

MeSH terms

  • Adult
  • Brain / physiology*
  • Electroencephalography* / methods
  • Female
  • Humans
  • Male
  • Memory, Short-Term / physiology*
  • Signal Processing, Computer-Assisted
  • Young Adult