Stats On Time courses (Beta Weights)
Statistics are done on the time courses (beta weights) to determine the
significance of the components. To use this utility, temporal sorting must be
done on all data-sets. When you click on "Stats On Beta Weights" under
Utilities
drop down box, figure 1 will open to select the regression parameters text file
(obtained after temporal sorting). An option is provided to average beta weights
across runs or sessions (Figure 2). Figure 3 shows the GUI for doing statistics
on beta weights. The parameters in the figure are as follows:
- Select design criteria - Design criteria used to determine the
significance of components. There are six options like one sample t-test,
two sample t-test, one way anova (groups), one way anova (regressors), two way anova (groups, regressors)
and Multiple Regression.
- Groups listbox - You can add groups using + button adjacent to
groups listbox. Figure 4 shows the GUI for entering group name, subject and
session numbers. You need to double click or press enter key on the selected
group to change the group name or subject numbers. To remove a group use
- button.
- Regressors listbox - This listbox is used to select the conditions or
regressors. You can select only one regressor for one sample t-test and two
sample t-test whereas for anova you can select multiple regressors.
- Contrasts listbox - This parameter is optional. You can specify contrast
names and vector using + button adjacent to contrasts listbox. The
length of contrast vector depends on the design criteria used:
- One way anova (groups) - The length of contrast vector is equal to
the number of selected groups.
- One way anova (regressors) - The length of contrast vector is equal
to the number of selected regressors.
- Two way anova (groups, regressors) - The length of contrast vector
is equal to the number of selected groups and regressors. For example if
you have two groups and four conditions, the first two entries
correspond to groups and the next four entries correspond to regressors.
- Calculate - Statistics are calculated after selecting design
criteria, groups, regressors and contrasts (Figure 5). All the results are
printed to a text file with suffix 'summary.txt'. This file also contains
mean and standard deviation for each condition of a group.
Note:
- If you have selected one way anova (groups) as the design
criteria, option is provided to do operation on regressors. For example, if you
have selected targets and novels as regressors. You can subtract novels from
targets and treat it as a single regressor by specifying [1 -1] for equation of
regressors (Figure 6).
- When you select Multiple regression tool, a figure
window will open to select the regressor file or files like age,
test scores, etc. After you selected the regressor files, R-square
statistic is calculated for each component. We also report the
slopes and partial correlations of the regressors.
- If you wish to use your favorite statistics package to do the
statistics on the beta weights, 'icatb_parse_regression_mat.m' function is provided that will write regression parameters in a
excel file or load it in the MATLAB command window. Type help
'icatb_parse_regression_mat.m' for more information.
Figure 1: Figure window used to select regression parameters
text file.
Figure 2: Option for averaging beta weights across runs or
sessions.
Figure 3: GUI for doing stats on beta weights.
Figure 4: Figure window used to enter group name, subject
numbers and session numbers.
Figure 5: Figure shows the completed parameters for doing
stats.
Figure 6: Option provided for doing operation on regressors
when one way anova (groups) design criteria is selected.