FieldTrip

Note that this reference documentation is identical to the help that is displayed in Matlab when you type “help ft_spike_plot_isireturn”.

  FT_SPIKE_PLOT_ISIRETURN makes a return plot from ISIH structure. A return
  plot (or Poincare plots) plots the isi to the next spike versus the isi
  from the next spike to the second next spike, and thus gives insight in
  the second order isi statistics. This func also plots the raw
  isi-histogram on left and bottom and thereby give a rather complete
  visualization of the spike-train interval statistics.
 
  Use as
    ft_spike_plot_isireturn(cfg, data) 
 
  Inputs:
    ISIH must be the output structure from FT_SPIKE_ISIH and contain the field
    ISIH.isi. If cfg.isihist = 'yes', the field ISIH.isih and ISIH.time must
    be present as well.
 
  General configurations:
    cfg.spikechannel     = string or index of single spike channel to trigger on (default = 1)
                           Only one spikechannel can be plotted at a time.
    cfg.density          = 'yes' or 'no', if 'yes', we will use color shading on top of
                           the individual datapoints to indicate the density.
    cfg.scatter          = 'yes' (default) or 'no'. If 'yes', we plot the individual values.
    
  General configurations related to smoothing the scatterplot:
    cfg.smoothmethod     = 'kernel' (default) or 'hist'.
                            If 'kernel', we overlay a smooth density plot calculated by 
                            non-parametric kernel smoothing with cfg.kernel.
                            If 'hist', we overlay a 2-D histogram.
    cfg.dt               =  resolution of the 2-D histogram, or of the kernel plot. Since we 
                            have to smooth for a finite number of values, cfg.dt determines
                            the resolution of our smooth density plot.
    cfg.colormap         = N-by-3 colormap (see COLORMAP). Default = hot(256);
    cfg.interpolate      = 'yes' or 'no', determines whether we interpolate the density
                            plot
 
  Specific configurations related to kernel smoothing of the scatterplot:
    cfg.kernel           = 'gausswin' or 'boxcar', or N-by-N matrix containing window
                            values with which we convolve the scatterplot that is binned
                            with resolution cfg.dt. N should be uneven, so it can be centered
                            at each point of the lattice.
                            'gausswin' is N-by-N multivariate gaussian, where the diagonal of the 
                            covariance matrix is set by cfg.gaussvar.
                            'boxcar' is N-by-N rectangular window.
                            If cfg.kernel is numeric, it should be of size N-by-N.
    cfg.gaussvar         =  variance  (default = 1/16 of window length in sec).
    cfg.winlen           =  window length in seconds (default = 5*cfg.dt). The total
                            length of our window is 2*round*(cfg.winlen/cfg.dt) +1;