Help Understanding F-tests using AM2

AFNI Message Board
1

Hello all!

My name is Michael and I’m a new AFNI user. I’ve tried to search for some literature from AFNI regarding the use of AM2 for stim_types and understanding the output of AM2. I know that using AM2 gives you 2 regressors per condition (Beta for mean, and Beta for amplitude modulation), but I’m still unsure as to the statistical results of it.

My question is: When viewing F-stats for a single condition, what exactly do those stats mean (how are the mean & modulation regressors interpreted together)?

Also, if there’s any literature you can direct me to for understanding the F-stats it would be much appreciated!

2

When viewing F-stats for a single condition, what exactly do those stats mean (how are the mean
& modulation regressors interpreted together)?

If the two effects are b0 and b1, the null hypothesis for the F-test is

H[sub]0[/sub]: b0 = 0 and b1 = 0

So, when the null hypothesis is rejected, it means that at least one of two effects are statistically significant.

3

Gang Chen,

I apologize for the delayed response but thank you for your help! You’ve made AM2 make much more sense to me now and I appreciate it.