Association between EEG spectral power dynamics and event related potential amplitude on a P300 speller

被引:2
作者
Bojorges-Valdez, Erik [1 ]
Yanez-Suarez, Oscar [2 ]
机构
[1] Univ Iberoamer AC, Dept Engn, Mexico City, DF, Mexico
[2] Univ Autonoma Metropolitana, Dept Elect Engn, Mexico City, DF, Mexico
来源
BIOMEDICAL PHYSICS & ENGINEERING EXPRESS | 2018年 / 4卷 / 02期
关键词
event related potentials; detrended fluctuation analysis; power spectral density time series; scaling exponent; BCI; EEG;
D O I
10.1088/2057-1976/aaa15e
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Objective: Event related potentials (ERPs) are phenomena that produce alterations on the spectral power of different bands synced with stimuli presentation. Such a perspective overlooks that the spectral power time series present spontaneous oscillations, the dynamics of which have been associated with the state of health and could be also linked to superior processes such as attention, memory, or cognition. In this work, amplitude of the ERP has been associated with dynamics of the power spectral density (PSD) time series. Approach: Data from a public P3 speller database were analyzed. Statistics about latency, amplitude and scalp distribution of the ERP confirm correct task realization. The spectral power over alpha, beta and gamma bands were correlated with ERP amplitude, considering such estimations time-locked with stimuli presentation. Additionally, PSD time series over the same bands were constructed using a sliding window not synced with stimulation. From these time series, a scaling exponent was computed using a detrended fluctuation analysis approach, and such indices were correlated with ERP amplitude of all registered electrodes. Main results: The correlation between ERP amplitude and scaling exponents of PSD series achieve statistical significance in almost all channels and for all analyzed bands (excepting alpha at C4). In particular, beta and gamma PSD series achieve R-2 values above 0.6 for Cz. Meanwhile, the correlation analysis with time-locked PSD achieves a statistically significant relationship with ERP amplitude (p < 0.05) but with a coefficient of determination (R-2) below 0.2, only for gamma and alpha over the C4 and Pz channels. Significance: The ERP amplitude and scaling exponent of the PSD time series show an anticorrelated behavior, suggesting that larger ERP amplitudes are elicited when the PSD series are less autocorrelated, similar to white noise, further suggesting that the processes involved in the generation of such dynamics could be related to the mechanisms of cognitive responses.
引用
收藏
页数:9
相关论文
共 42 条
  • [1] Altenmuller E.O., 2005, ELECTROENCEPHALOGRAP
  • [2] Event-related potentials in the auditory oddball as a function of EEG alpha phase at stimulus onset
    Barry, RJ
    Rushby, JA
    Johnstone, SJ
    Clarke, AR
    Croft, RJ
    Lawrence, CA
    [J]. CLINICAL NEUROPHYSIOLOGY, 2004, 115 (11) : 2593 - 2601
  • [3] EEG alpha activity and the ERP to target stimuli in an auditory oddball paradigm
    Barry, RJ
    Kirkaikul, S
    Hodder, D
    [J]. INTERNATIONAL JOURNAL OF PSYCHOPHYSIOLOGY, 2000, 39 (01) : 39 - 50
  • [4] BASAR E, 1985, INT J NEUROSCI, V26, P161
  • [5] Are cognitive processes manifested in event-related gamma, alpha, theta and delta oscillations in the EEG?
    Basar, E
    Basar-Eroglu, C
    Karakas, S
    Schürmann, M
    [J]. NEUROSCIENCE LETTERS, 1999, 259 (03) : 165 - 168
  • [6] Human EEG shows long-range temporal correlations of oscillation amplitude in Theta, Alpha and Beta bands across a wide age range
    Berthouze, Luc
    James, Leon M.
    Farmer, Simon F.
    [J]. CLINICAL NEUROPHYSIOLOGY, 2010, 121 (08) : 1187 - 1197
  • [7] Blasio F M D, 2013, J PSYCHOPHYSIOL, V89, P9
  • [8] Bojorges-Valdez E, 2014, P 6 INT BRAIN COMP I
  • [9] Bojorges-Valdez E R, 2013, P 5 INT BRAIN COMPUT, V264-5, P133
  • [10] Evaluation of the continuous detection of mental calculation episodes as a BCI control input
    Bojorges-Valdez, Erik
    Echeverria, Juan C.
    Yanez-Suarez, Oscar
    [J]. COMPUTERS IN BIOLOGY AND MEDICINE, 2015, 64 : 155 - 162