Hemodynamic Response Function (HRF)

The hemodynamic response function (HRF) is convolved by the input stimuli represented by a boxcar to generate a more realistic model of the BOLD response. The output of this convolution process results in the design matrix. It is applied since there is a delay from when the stimuli is presented until the BOLD response occurs in a particular voxel, and it is given as a combination of two gamma functions

where time t varies with precision , and the parameters , and depends on user input. The user can alter the peak and the FWHM (full-width-half-maximum) of the gamma functions, as well as the dip c. The peak  and FWHM is then used in the estimation of and  using the two formulas:

The HRF is normalized by the sum itself prior to convolving it with the input boxcar function.

Please note; in most cases it should not be necessary to change the default values of the HRF.

Peak 1

Specify the peak of the first gamma function.

FWHM 1

Specify the FWHM of the first gamma function.

Peak 2

Specify the peak of the second gamma function.

FWHM 2

Specify the FWHM of the second gamma function.

DIP

Specify the dip c.

Duration in Seconds

Specify the duration of the HRF in seconds.

Save Parameters

Check to save parameters when pressing Update Model.

Be aware; if you select this choice, the specified parameters will be applied the next time the BOLD module is used.

Update Model button

Update the design matrix using the new HRF.

Reset button

Reset HRF parameters to the default values, as displayed in the picture above.

Related topics:

Create paradigm in design file
Design Matrix
General linear model (GLM) estimation