Reverse Correlation
This analysis is used for estimation of receptive fields in vision research. The analysis calculates the average visual stimulus that preceded the spike.
Parameters
Parameter |
Description |
|---|---|
Reference |
the variable that contains timestamps of the stimuli. |
Stim. Sequence File |
The path of the text file that contains the sequence of images used for stimulation. See Algorithm below for file format description. |
Char. per pixel |
The number of characters per pixel in image file. |
Pixels in Row |
The number of pixels in a row for each stimulus image. |
Rows in Image |
The number of rows of pixels in each stimulus image. |
Cache Stim. Data |
An option to cache stimulus image data. If this option is selected, the image data is loaded once and stored in memory (until a different stimulus sequence file is specified). To clear cache, run this analysis once with this option disabled. |
XMin (deg) |
Minimum of the X axis of the average stimulus display (in degrees). |
XMax (deg) |
Maximum of the X axis of the average stimulus display (in degrees). |
YMin (deg) |
Minimum of the Y axis of the average stimulus display (in degrees). |
YMax (deg) |
Maximum of the Y axis of the average stimulus display (in degrees). |
Time Min (sec) |
Time minimum in seconds. |
Time Max (sec) |
Time maximum. For example, if Time Min = -0.4 and Time Max = 0, NeuroExplorer will analyze all the stimuli that were presented up to 400 msec before each spike. |
Time Bin (sec) |
Time bin. |
Display |
This option specifies what view of the average stimulus will be displayed. See Algorithm below. |
Show Slice At |
This option specifies the slice that will be shown. The units and valid range depend on the Display option See Algorithm below. |
Z Min |
Color scale minimum. |
Z Max |
Color scale maximum. |
Smooth Colors |
An option to smooth colors of the average stimulus matrix. |
Matrix Scale |
An option on what color scale to use when drawing the matrix. |
Send to Matlab |
An option to send the matrix of numerical results to Matlab. See also Matlab Options. |
Matrix Name |
Specifies the name of the results matrix in Matlab workspace. |
Matlab command |
Specifies a Matlab command that is executed after the numerical results are sent to Matlab. |
Send to Excel |
An option to send numerical results or summary of numerical results to Excel. See also Excel Options. |
Sheet Name |
The name of the worksheet in Excel where to copy the numerical results. |
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. |
Save 3D Matrix |
An option to save 3D matrix of numerical results in a .csv file. 3D matrix (in X vs. Y display case) is saved starting from the first slice (at Time Min), then the second slice, etc. Within the slice, the first row corresponds to Y Min, the first column corresponds to X Min. |
Save File |
The name of the file that will contain 3D matrix data. |
Summary of Numerical Results
The following information is available in the Summary of Numerical Results
Column |
Description |
|---|---|
Variable |
Variable name. |
YMin |
Y axis minimum. |
YMax |
Y axis maximum. |
RefCount |
Number of reference events (and images used) |
Color Scale Min |
Minimum of color scale. |
Color Scale Max |
Maximum of color scale. |
Spikes |
The number of spikes used in calculation. |
Algorithm
Stimulus File Format
NeuroExplorer assumes that the sequence of images used for stimulation is saved in a text file. The file has the following format:
each line of the file represents a row of pixels in the stimulus image
each pixel is represented by an integer
each pixel has the same number of characters used for its description
images are saved in the file in the order they were presented
images can be (optionally) separated by blank lines
Here is an example of an image file with 2 images (16 pixels per row, 12 rows per image):
0 1 1 1 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 1 1 1 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
The computational algorithm works the following way. For each spike of a selected neuron, NeuroExplorer identifies the images that precede the spike and are within the specified time interval (between Tmin and Tmax). For example, if Tmin = -0.4 and Tmax = 0, NeuroExplorer will analyze all the stimuli that were presented up to 400 msec before each spike.
Then, NeuroExplorer calculates the average of the presented stimuli over all the spikes. The result is a 3-dimensional matrix (with X, Y, and Time dimensions, time axis is divided into bins of the specified size).
Display option identifies what 2-dimensional view of this matrix is presented:
X vs. Y display shows an average stimulus image for the specified time slice
X vs. Time display shows average X pixels of the stimuli for the specified Y value
Y vs. Time display shows average Y pixels of the stimuli for the specified X value
Reference
Izumi Ohzawa, Gregory C. DeAngelis, and Ralph D. Freeman (1996). Encoding of binocular disparity by simple cells in the cat’s visual cortex. J. Neurophysiol. 75: 1779-1805.