Hi,
I just tried running 3dDeconvolve, and it shut down due to collinearity problems (see message below). I'm not totally positive, but I was thinking perhaps the problem is with the number of stim types I have. Essentially, the experiment is a memory experiment, in which participants can respond in three ways to three different stimuli, giving 9 stimulus types. In the end, 4 of these stim types are of no interest.
Sorry this message will be so long... but I wanted to make sure you had all the info...
++ Input polort=3; Longest run=444.0 s; Recommended minimum polort=3 ++ OK ++
++ -stim_times using TR=2 seconds
++ '-stim_times 1' using LOCAL times
++ '-stim_times 2' using LOCAL times
++ '-stim_times 3' using LOCAL times
++ '-stim_times 4' using LOCAL times
++ '-stim_times 5' using LOCAL times
++ '-stim_times 6' using LOCAL times
++ '-stim_times 7' using LOCAL times
++ '-stim_times 8' using LOCAL times
++ '-stim_times 9' using LOCAL times
GLT matrix from 'glt/glt1.txt':
: +RFACE -KFACE +RSCR -KSCR
0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -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
GLT matrix from 'glt/glt2.txt':
: +RFACE +RSCR +KFACE +KSCR -NNEW
0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 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 -1 -1 -1 -1 -1 -1 -1 -1
++ Wrote matrix values to file ret_decon_results.xmat.1D
++ Signal+Baseline matrix condition [X] (648x84): 8.53189 ++ VERY GOOD ++
*+ WARNING: !! in Signal+Baseline matrix:
* Largest singular value=5.10984
* 1 singular value is less than cutoff=5.10984e-07
* Implies strong collinearity in the matrix columns!
++ Signal+Baseline matrix singular values:
-2.95087e-16 0.0701967 0.199864 0.211716 0.284032
0.298257 0.338876 0.350676 0.37686 0.409233
0.421726 0.437665 0.463918 0.486843 0.497015
0.512038 0.53285 0.559236 0.562001 0.581472
0.595749 0.612754 0.632418 0.644834 0.65886
0.677422 0.685308 0.706352 0.715775 0.722027
0.724429 0.741199 0.753925 0.771943 0.78733
0.80129 0.80892 0.827448 0.83183 0.842812
0.863628 0.866832 0.893006 0.909522 0.924186
0.940607 0.954296 0.958593 0.978412 0.994004
1.01637 1.0361 1.04034 1.05203 1.08484
1.09058 1.09322 1.12536 1.14289 1.17844
1.19817 1.22097 1.26877 1.29511 1.3082
1.32638 1.34283 1.37446 1.38234 1.44778
1.45931 1.52157 1.54959 1.59263 1.64142
1.67575 1.76894 1.81007 1.96537 1.97371
2.01913 2.13862 2.32869 5.10984
++ Signal-only matrix condition [X] (648x72): 4.58671 ++ VERY GOOD ++
*+ WARNING: !! in Signal-only matrix:
* Largest singular value=4.2094
* 1 singular value is less than cutoff=4.2094e-07
* Implies strong collinearity in the matrix columns!
++ Signal-only matrix singular values:
-1.45422e-15 0.200086 0.212581 0.286088 0.299191
0.350519 0.354744 0.380394 0.421339 0.425179
0.440974 0.474207 0.49603 0.514439 0.534943
0.56066 0.577257 0.581916 0.612387 0.62524
0.651944 0.668181 0.684465 0.713034 0.720462
0.729966 0.746677 0.75623 0.78042 0.799158
0.824881 0.828854 0.844207 0.861192 0.862288
0.881344 0.919898 0.925006 0.933023 0.95249
0.967979 0.998924 1.00486 1.02281 1.0566
1.10267 1.11551 1.12209 1.15387 1.20546
1.21184 1.22023 1.24224 1.2817 1.29369
1.31052 1.33772 1.36792 1.42362 1.4257
1.53811 1.55005 1.56342 1.63663 1.67929
1.74389 1.79914 1.90604 1.92857 2.05314
2.09448 4.2094
++ Baseline-only matrix condition [X] (648x12): 1.2519 ++ VERY GOOD ++
++ -polort-only matrix condition [X] (648x12): 1.2519 ++ VERY GOOD ++
++ Matrix inverse average error = 0.000755857 ++ OK ++
++ Matrix setup time = 5.91 s
** ERROR: !! 3dDeconvolve: Can't run past 2 matrix warnings without -GOFORIT 0
** ERROR: !! Currently at -GOFORIT 0
** ERROR: !! See file 3dDeconvolve.err for all WARNING and ERROR messages !!
** ERROR: !! Be sure you understand what you are doing before using -GOFORIT !!
** ERROR: !! If in doubt, consult with someone or with the AFNI message board !!
** FATAL ERROR: !! 3dDeconvolve (regretfully) shuts itself down !!
If someone could please explain the meaning of this error to me... I would very much appreciate it!
Thanks so much!
