Index

Name

mkcontrast-sess - creates a contrast matrix that is then used as input to test hypotheses.

Synopsis

mkcontrast-sess -contrast <contrastname> -analysis <analysisname> [options]

Arguments

Positional Arguments

None

Required Flagged Arguments

None

Optional Flagged Arguments

-analysis analysisname (or use -an)

name of session-level functional analysis

Name of the analysis to create the contrast for. The analysis must have been defined by mkanalysis-sess. Also, selxavg-sess must have been run at least once.

-contrast contrastname (or use -co)

contrast name

The name of the contrast. This name is then used in later programs (eg, stxgrinder-sess). mkcontrast-sess will create a matlab file in the study analysis directory. This file will be called contrastname.mat. When stxgrinder-sess is run, it will create a subdirectory called contrastname under the analysis directory for each sessions.

-a a1 <-a a2 ...>

positive contrast conditions

These specify which conditions to use in the contrast, where 1, a2, c1, c2, etc, are the condition numbers as found in the paradigm file. The -a conditions will be weighted positively. The -c conditions will be weighted negatively. Eg, -a 2 -c 1 -c 3 will weight conditions 1 and 3 by -0.5 and condition 2 by +1.

-c c1 <-c c2 ...>

negative contrast conditions

These specify which conditions to use in the contrast, where 1, a2, c1, c2, etc, are the condition numbers as found in the paradigm file. The -a conditions will be weighted positively. The -c conditions will be weighted negatively. Eg, -a 2 -c 1 -c 3 will weight conditions 1 and 3 by -0.5 and condition 2 by +1.

-nosumconds

don't sum conditions (for f-test)

Specifies that the conditions be tested separately

-ircorr nircorr

correlate with gamma function

-deltarange dmin dmax

range of delta for gamma function

-taurange tmin tmax

range of tau for gamma function

-umask umask

set unix file permission mask

-scriptonly

don't run, just generate a script

-version

print version and exit

-help

For FIR Designs

-setwdelay

prompt for setting of delay weights

For FIR designs, allows the user to set the weight for each post-stimulus delay. When the program runs, it will prompt the user to enter the weights (eg, if there are 10 points in the FIR, the the user will enter a list of 10 numbers). If you do not want to include a point, set its weight to 0. Eg, if thre are 10 points, but you only want to consider points 3-7, then enter 0 0 1 1 1 1 1 0 0 0. These weights are applied on a condition-by-condition basis. See also -sumdelays.

-sumdelays

sum weighted delays

For FIR designs, sum the points inside the FIR window so that each condition is represented by a single number (ie, the sum). If delay weights have been specified (see -setwdelay), then the weighted sum is computed.

-rmprestim

subtract prestimulus baseline

For FIR designs, compute a contrast such that the average hemodynamicresponse prior to simulus onset is subtracted from the hemodynamic response waveform. This is good for removing baseline shifts.

Outputs

??

??

Description

mkcontrast-sesscreates a contrast matrix that is then used as input to test hypotheses. The actual voxel-by-voxel statistical test is performed by stxgrinder-sess.

Examples

Example 1

??

Example 2

??

Bugs

??

See Also

stxgrinder-sess

Links

FreeSurfer, FsFast

Methods Description

description
description

References

References/Lastname###

Reporting Bugs

Report bugs to <analysis-bugs@nmr.mgh.harvard.edu>

Author/s

DougGreve