eegsim {eegkit}

Simulate Event-Related Potential EEG Data

Description

Simulates event-related potential EEG data from hypothetical visual-stimulus ERP study. Data are simulated using a linear combination of five spatiotemporal component functions: P100, N100, P200, N200, and P300 components. User can control the coefficient (weight) given to each component, as well as the time shift (delay) of each component.

Usage

eegsim(channel, time, coefs = rep(1,5), tshift = rep(0,5))

Arguments

channel

Character vector of length n giving EEG channel of simulated data.

time

Numeric vector of length n giving time point of simulated data (should be in interval [0,1]).

coefs

Numeric vector of length 5 giving the coefficients (weights) to use for P100, N100, P200, N200, and P300 components (respectively).

tshift

Numeric vector of length 5 giving the time shifts (delays) to use for P100, N100, P200, N200, and P300 components (respectively).

Value

Returns a vector of simulated EEG data corresponding to the input channel(s), time point(s), coefficients, and time shifts.

Note

Simulates data for 39 parietal and occipital electrodes: CP1 CP2 CP3 CP4 CP5 CP6 CPZ I1 I2 IZ O1 O2 OZ P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 POZ PZ TP7 TP8 TP9 TP10

Returns simulated value of 0 for other electrodes.

Author(s)

Nathaniel E. Helwig <[e-mail]>

References

Created by Nathaniel E. Helwig (2014) using data from:

Bache, K. & Lichman, M. (2013). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science.

Begleiter, H. Neurodynamics Laboratory. State University of New York Health Center at Brooklyn. http://www.downstate.edu/hbnl/

Ingber, L. (1997). Statistical mechanics of neocortical interactions: Canonical momenta indicatros of electroencephalography. Physical Review E, 55, 4578-4593.

Ingber, L. (1998). Statistical mechanics of neocortical interactions: Training and testing canonical momenta indicators of EEG. Mathematical Computer Modelling, 27, 33-64.

Examples

##########   EXAMPLE   ##########

### plot spatiotemporal component functions

# data(eegcoord)
# chnames <- rownames(eegcoord)
# tseq <- seq(0,1,length.out=200)

# quartz(width=18,height=6)
# layout(matrix(c(1,2,3,4,5,6,7,8,9,10,11,11), 2, 6, byrow = TRUE))

# eegspace(eegcoord[,4:5],p1s(chnames),cex.point=1,main=expression(psi[p1]),cex.main=2,vlim=c(-3,9))
# eegtime(tseq,p1t(tseq),ylim=c(-1,1),asp=1/2,main=expression(tau[p1]),cex.main=2,
#         xlab="Time After Stimulus (sec)")
# eegspace(eegcoord[,4:5],p2s(chnames),cex.point=1,main=expression(psi[p2]),cex.main=2,vlim=c(-3,9))
# eegtime(tseq,p2t(tseq),ylim=c(-1,1),asp=1/2,main=expression(tau[p2]),cex.main=2,
#         xlab="Time After Stimulus (sec)")
# eegspace(eegcoord[,4:5],p3s(chnames),cex.point=1,main=expression(psi[p3]),cex.main=2,vlim=c(-3,9))
# eegtime(tseq,p3t(tseq),ylim=c(-1,1),asp=1/2,main=expression(tau[p3]),cex.main=2,
#         xlab="Time After Stimulus (sec)")
# eegspace(eegcoord[,4:5],n1s(chnames),cex.point=1,main=expression(psi[n1]),cex.main=2,vlim=c(-3,9))
# eegtime(tseq,n1t(tseq),ylim=c(-1,1),asp=1/2,main=expression(tau[n1]),cex.main=2,
#         xlab="Time After Stimulus (sec)")
# eegspace(eegcoord[,4:5],n2s(chnames),cex.point=1,main=expression(psi[n2]),cex.main=2,vlim=c(-3,9))
# eegtime(tseq,n2t(tseq),ylim=c(-1,1),asp=1/2,main=expression(tau[n2]),cex.main=2,
#         xlab="Time After Stimulus (sec)")
# plot(seq(-10,10),seq(-10,10),type="n",axes=FALSE,xlab="",ylab="")
# text(0,8,labels=expression(omega[p1]*" = "*psi[p1]*tau[p1]),cex=2)
# text(0,4,labels=expression(omega[n1]*" = "*psi[n1]*tau[n1]),cex=2)
# text(0,0,labels=expression(omega[p2]*" = "*psi[p2]*tau[p2]),cex=2)
# text(0,-4,labels=expression(omega[n2]*" = "*psi[n2]*tau[n2]),cex=2)
# text(0,-8,labels=expression(omega[p3]*" = "*psi[p3]*tau[p3]),cex=2)



### plot simulated data at various time points

# quartz(width=15,height=3)
# tseq <- c(50,150,250,350,450)/1000
# par(mfrow=c(1,5))
# for(j in 1:5){
#   eegspace(eegcoord[,4:5],eegsim(chnames,rep(tseq[j],87)),vlim=c(-6.8,5.5),
#            main=paste(tseq[j]*1000," ms"),cex.main=2)
# }