Multichannel Systems NeuroExplorer - Manual (Page 111)

TopLeft Specifies the Excel cell where the results are copied. Should be in the form
CR where C is Excel column name, R is the row number. For example, A1
is the top-left cell in the worksheet.
Summary of Numerical Results
The following information is available in the Summary of Numerical Results
Column Description
Variable Variable name.
Y Min Y axis minimum.
Y Max Y axis maximum.
Spikes The number of spikes used (for neurons or events).
Rate Hist Bin Bin size of the rate histograms used for the calculations (for neurons or
events).
Filter Length The length of the interval filter.
Mean Freq. Mean firing rate (for neurons or events).
Confidence Level Confidence level for the coherence values.
Algorithm
NeuroExplorer uses rate histograms to estimate the power spectra of the spike trains. The
parameters of rate histograms are calculated using the following formulas:
Bin = 1./(2.* Maximum_Frequency )
Number_of_Bins = 2* Number_of_Frequency_Values
For two variables X (reference) and Y (target) the following entities are calculated. The time intervals
of length Bin*NumberOfBins with specified overlaps are defined. For each interval, rate histograms
are calculated and detrended, Hanning window is applied and FFTs are calculated. Then, individual
and cross-densities are calculated:
Pxx = FFT(X)*Conj(FFT(X)),
Pyy = FFT(Y)*Conj(FFT(Y))
Pxy = FFT(X)*Conj(FFT(Y)).
Here Conj(z) is a complex conjugate of z. Pxx, Pyy and Pxy values are averaged across all intervals
and coherence values are calculated as
Mean(Pxy)*Mean(Pxy) / (Mean(Pxx)*Mean(Pyy)).
Coherence phase values are calculated as phase of Mean(Pxy).
For continuous variables, if both the reference and the target variables have the same digitizing
frequency, the FFTs of variable values (after detrending and applying Hanning tapering) are
calculated and then resampled to the specified frequency steps. If the reference is a timestamped
variable or two continuous variables have different digitizing rates, the values of continuous variables
are averaged within the specified bins and then FFTs of these averages are calculated.
Calculation of the Confidence Level
Confidence levels are calculated as described in Kattla and Lowery (2010). The confidence level Z is
calculated as
Page 109