Post event time histogram
Purpose
Description
Tips
Example
Macro Synopsis
Modules
Related Functions
References
Purpose
Post event time histogram.
Description
"Post event time histogram", also refered to as (event-related) averaging, or
interspike-interval histogram, generates a signal y showing the average
course of the probe signal x after a positive or negative peak (event)
in the trigger signal trig, i.e. the Post event time histogram is computed as the averaged signal of all time intervals
of the probe signal x, where 
is a
user-defined time delay, and 
is the time equivalent to
the i-th peak of the trigger signal.
Parameters are:
- window size (time interval) in x-axis units ws
For a detailed description of this parameter, refer to Peak-to-peak intervals.
- peaktype, the type of peaks, which is either
- "Maxima" (0) or
- "Minima" (1)
- crossing, determinig the choice whether a threshold
crossing between peaks is
- "not checked" (0) or
- "checked" (1)
- threshold thresh for the detection of peaks
- time delay tau (see above)
To get proper results, the algorithm should only be applied to stationary time series.
Tips
It is possible to trigger the averaging of a signal by its own peaks by creating a copy of it (e.g. by adding a constant 0 to the signal (Add constant) or by rescaling it with a factor 1 (Rescale)) and using these "twins" as input signals x and trig, respectively.
Example
Post event time histogram is used e.g. for the analysis of evoked potentials in EEGs.
Macro Synopsis
y =
PETH([trig,x],ws,peaktype,crossing,thresh,tau);
signal trig,x,y;
float ws;
option peaktype,crossing;
float thresh, tau;
Modules
Nonlinear
Related Functions
Peak-to-peak intervals, Post event scan.
References
Perkel/Gerstein/Moore [51]