usage plot_coupling(data,locs,pars); makes head-in-head plots data is an NxN matrix where N is the number of channels locs is (ideally) an Nx5 matrix: 1st column: a channel i gets a circle only if locs(i,1)>.5 if locs is an Mx5 matrix with M<N the absolute values of the first column are interpreted as indices of scalp-electrodes. If the values themselves are smaller than 0 then these electrodes are only included within each small circle but do not correspond to a small circle itself. 2nd and 3rd column: x,y coordinates of electrodes in 2D 4th and 5th column: x,y coordinates of centers of spheres slight deformation of electrode locations to avoid overlapping spheres if locs is Nx2, the 2 columns are interpreted as x,y coordinates for both electrodes and circle-centers. if locs is Nx3, the 2nd and 3rd column are interpreted as x,y coordinates for both electrodes and sphere-centers. The first as interpreted as for 5xN case. pars sets parameters; pars.scale sets color-scale; it is a 1x2 vector denoting min and max of colorbar. defaults is pars.scale=[-max(max(abs(data))),max(max(abs(data)))]; pars.resolution sets resolution. Default is pars.resolution=25 Increasing it makes better pictures but is slower pars.global_size sets global size factor of small circles and distancec between circles. default: global_size=1 pars.relative_size sets size factor for distances leaving circle-size unchanged default: relative_size=1 pars.head_up moves the big circle (the head) up. Default: head_up=0 pars.head_right moves the big circle (the head) to the right. Default: head_right=0 REFERENCES:  Nolte G, Bai O, Wheaton L, Mari Z, Vorbach S, Hallett M. Identifying true brain interaction from EEG data using the imaginary part of coherency. Clin Neurophysiol. 2004 Oct;115(10):2292-307.