NITRC Generalized PPI Toolbox Forum: help

P.concatR not working?

Madeline Ragland — Tue, 24 Feb 2026 19:51:00 GMT

<p>I am trying to use the [P.ConcatR = 1] parameter in gPPI toolbox, but I continue to get an error that the matrix dimensions do not match when running the spm time series extraction. When extracting the time series, "y" is not equal to the total number of concatenated volumes. </p>
<p>For context, my first level GLM in SPM contains 6 different runs as independent regressors (with matching task and motion regressors per run). I also have predefined ROIs from retinotopy that are in the same space as the bold runs (MNI space). Instead of extracting the time series prior to using the gPPI toobox, I am using a .nii file for the VOI parameter which is allowed. However, it seems that the time series is not being extracted for all of the volumes and that the P.concatR is not working properly. Any advice on how to fix this problem? Should I concatenate the runs at the first level? Should I extract the time series prior to gPPI for the VOI and if so, what is the recommended method? </p>

Error with timeseries extraction using gppi

Madeline Ragland — Tue, 24 Feb 2026 19:50:04 GMT

<p>Hi!</p>
<p> I am trying to use the gppi toolbox (PPPI Version: 13.1.4-17-2014 with SPM25) but have come across issues related to the timeseries extraction function  (y=spm_filter(SPM.xX.K,SPM.xX.W*y)). In our study, we have 6 runs per subject and individualized ROIs already created using retinotopy data. Everything is in MNI space and has the same voxel dimensions. </p>
<p> For the first level model, I included each run as an independent regressor. I wasn't sure how to create the one VOI.mat file/subject as the model contains 6 runs, so instead I used the nii file for the VOI parameter (P.VOI= VOI.nii). I also included P.ConcatR = 1; % Enables concatenation of sessions so that the 6 sessions are concatenated at this level of analysis. I also included P.equalroi=0; and P.FLmask=1 to offset an error related to "VOI larger than dataset". </p>
<p>When running the PPI, I get the following outputs: VOI has 670 voxels in 1.500000e+00x1.500000e+00x1.500000e+00 space. This is in the the same space as the input data and functional mask. But then receive the error message: Error using  *  multiplication. Error in timeseries_extract (line 355) y=spm_filter(SPM.xX.K,SPM.xX.W*y);</p>
<p>Please advise on how to setup the proper first-level GLM, VOI extraction and gPPI setup to help solve this problem!</p>
<p>Thanks!</p>
<p>Madeline</p>
<p> </p>

PPI analysis, nothing works :)

Ekaterina Pronizius — Tue, 26 Aug 2025 11:23:57 GMT

<p>Dear all, could you please tell me what am I doing wrong? </p>
<div class="rtcContent">
<div class="lineNode">clear all</div>
<div class="lineNode">clc</div>
<div class="lineNode"></div>
<div class="lineNode">skipNum = [14,26,29,37,40,41,55,56]; %array of the participants, which are not existing or problematic</div>
<div class="lineNode">subjects = [10:10];</div>
<div class="lineNode"></div>
<div class="lineNode">rois = {'lOFC'}; %one ROI for now; , 'rTPJ'</div>
<div class="lineNode"></div>
<div class="lineNode">% choose correct basepath</div>
<div class="lineNode">basepath = fullfile[BLINDED];</div>
<div class="lineNode">addpath(fullfile(basepath, fullfile('04_PPI', 'PPI', 'PPPIv13')));</div>
<div class="lineNode">subjpath = fullfile(basepath,fullfile('07_MRI_DATA'));</div>
<div class="lineNode"></div>
<div class="lineNode">for r = 1:length(rois)</div>
<div class="lineNode"> </div>
<div class="lineNode">clear roi; roi = char(rois(r));</div>
<div class="lineNode"> </div>
<div class="lineNode">for i = 1:length(subjects)</div>
<div class="lineNode"> </div>
<div class="lineNode">A = subjects(i);</div>
<div class="lineNode">if ~ismember(A, skipNum) %if NOT a member</div>
<div class="lineNode">P = [];</div>
<div class="lineNode">id = "MP_0" + subjects(i);</div>
<div class="lineNode"> </div>
<div class="lineNode">clear P</div>
<div class="lineNode">P.subject = char(id);</div>
<div class="lineNode"> </div>
<div class="lineNode">% directory of first level spm.mat file</div>
<div class="lineNode">% create a copy just in case</div>
<div class="lineNode">firstlevel_dir = fullfile(subjpath, P.subject, 'first_level');</div>
<div class="lineNode">ppi_firstlevel_dir = fullfile(subjpath, P.subject, 'first_level_PPI');</div>
<div class="lineNode"></div>
<div class="lineNode">% copy everything ONLY if the PPI copy does not exist yet</div>
<div class="lineNode">if ~isfolder(ppi_firstlevel_dir)</div>
<div class="lineNode">mkdir(ppi_firstlevel_dir); % create empty folder</div>
<div class="lineNode">fprintf('Created empty first_level_PPI for %s\n', P.subject);</div>
<div class="lineNode">else</div>
<div class="lineNode">fprintf('Using existing first_level_PPI for %s\n', P.subject);</div>
<div class="lineNode">end</div>
<div class="lineNode"></div>
<div class="lineNode">P.directory = firstlevel_dir;</div>
<div class="lineNode">P.outdir = ppi_firstlevel_dir; %output directory if you want to store the PPI analysis in a different location than the first-level SPM.mat file.</div>
<div class="lineNode">cd(P.directory);</div>
<div class="lineNode"> </div>
<div class="lineNode">% --- PPPI PARAMETERS ---</div>
<div class="lineNode">% path to VOI (roi.nii) + roiname</div>
<div class="lineNode">P.VOI = fullfile(basepath,'04_PPI','ROIs', [roi,'.nii']);</div>
<div class="lineNode">P.Region = roi; % roi name without '.nii'</div>
<div class="lineNode"> </div>
<div class="lineNode">% F-contrast to remove noise from the data before deconvolution</div>
<div class="lineNode">% (will be created)</div>
<div class="lineNode">% preprocessing</div>
<div class="lineNode">P.contrast = 'Omnibus F-test for PPI Analyses';</div>
<div class="lineNode">P.analysis = 'psy'; %stands for psychophysiological interaction</div>
<div class="lineNode">P.extract = 'eig'; % method of roi extraction: default = eigenvariate</div>
<div class="lineNode">% Specifies traditional SPM PPI ('trad') or generalized condition-specific PPI ('cond')</div>
<div class="lineNode">P.method = 'cond';</div>
<div class="lineNode">P.equalroi = 0; % specifies that ROIs must be same size in all subjects, 0 allows roi to be trimmed by first level mask</div>
<div class="lineNode">P.FLmask = 1; % 0 - no mask (no such image), default. 1 = specifies that the ROI should be restricted using the mask.img from the first-level statistics.</div>
<div class="lineNode"> </div>
<div class="lineNode">% Tasks = name of regressors (= the name you put in first-level</div>
<div class="lineNode">% specification. you only need to specify the conditions that you are interested</div>
<div class="lineNode">% in for your PPI analysis. These conditions should match the names of</div>
<div class="lineNode">% the regressors (or conditions) that you used in your first-level analysis.</div>
<div class="lineNode"></div>
<div class="lineNode">% regressors / tasks /</div>
<div class="lineNode">P.Tasks = {'1' 'IMM' 'MOR' 'NEU'}; %1 = tasks in all (this case 2) sessions.</div>
<div class="lineNode">% specify contrasts</div>
<div class="lineNode">P.CompContrasts = 1;</div>
<div class="lineNode">P.Contrasts(1).left = {'IMM'};</div>
<div class="lineNode">P.Contrasts(1).right = {'MOR'};</div>
<div class="lineNode">P.Contrasts(1).STAT = 'T';</div>
<div class="lineNode">P.Contrasts(1).name = 'IMM > MOR';</div>
<div class="lineNode"></div>
<div class="lineNode">P.Contrasts(2).left = {'MOR'};</div>
<div class="lineNode">P.Contrasts(2).right = {'IMM'};</div>
<div class="lineNode">P.Contrasts(2).STAT = 'T';</div>
<div class="lineNode">P.Contrasts(2).name = 'IMM < MOR';</div>
<div class="lineNode"></div>
<div class="lineNode">% if 'cond' as method:</div>
<div class="lineNode">P.Weights = []; % would have to be entered if traditional SPM PPI (n*3 matrix; n = number of conditions)</div>
<div class="lineNode">P.SPMver = 12;</div>
<div class="lineNode"></div>
<div class="lineNode">% --- RUN PPPI ---</div>
<div class="lineNode">try</div>
<div class="lineNode">fprintf('Running full estimation for %s\n', P.subject);</div>
<div class="lineNode">P.Estimate = 1;</div>
<div class="lineNode">PPPI(P);</div>
<div class="lineNode">catch ME</div>
<div class="lineNode">warning('Full estimation failed for %s: %s', P.subject, ME.message);</div>
<div class="lineNode">% continue anyway</div>
<div class="lineNode">end</div>
<div class="lineNode">end</div>
<div class="lineNode">end</div>
<div class="lineNode">end<br><br>My problem:<br>The analysis runs but gives me an error <br>
<div>Created empty first_level_PPI for MP_010</div>
<div>Running full estimation for MP_010</div>
<div>Log File: MP_010_PPPI_8_26_2025.log</div>
<div> </div>
<div>PPPI Version: 13.1.4-17-2014</div>
<div> </div>
<div> </div>
<div>Parameters used:</div>
<div>     Processing subject: MP_010</div>
<div>     VOI file          : X:\userdata\ekaterinap85\data\Publikation\Moral_Pletti\Pletti_code\04_PPI\ROIs\lOFC.nii</div>
<div>     Output file will be: MP_010_lOFC_session#_cond_PPI_regressors.txt</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Valid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>Warning: Missing conditions!!! Invalid Contrast</div>
<div>Invalid Contrast</div>
<div>ESS image  7                    :        ...written ess_0007.nii</div>
<div>spm{F} image  7                 :       ...written spmF_0007.nii</div>
<div>Saving SPM.mat                  :               ...SPM.mat saved</div>
<div>     Contrast          : Omnibus F-test for PPI Analyses</div>
<div>     Analysis          : Psychophysiological Interactions</div>
<div>     Extraction        : eigenvariate</div>
<div>     Tasks             : _1_IMM_MOR_NEU</div>
<div>     Method            : Condition Specific<br><br>No output is saved in the PPI folder<br>No contrasts<br>The results of the omnibus tests is just a whole-blob activation. I really dont know what to do. I want to investigate connectivity from the seed region lOPC for the contrast IMM>MOR and IMM<MOR<br><br>Any help is appreciated <br>Ekaterina</div>
</div>
</div>

Issue with PPI Code: "Error Moving Contrast"

Nobody — Mon, 21 Oct 2024 14:53:13 GMT

<p>Dear all,</p>
<p>I'm running ppi code  and unfortunately encountering an ‘’error moving contrast‘’ issue. I've checked potential problems from the threads : </p>
<p><a title="Valid Contrast - error moving contrast Images" href="forum.php?thread_id=14096&forum_id=1970">https://www.nitrc.org/forum/forum.php?thread_id=14096&forum_id=1970</a></p>
<p>and </p>
<p><a href="message.php?msg_id=24950">https://www.nitrc.org/forum/message.php?msg_id=24950</a></p>
<p>but to no avail(if ~strcmp(spm('Ver'),'SPM8') && ~strncmpi(spm('Ver'),'SPM12',5) ,checked)</p>
<p>Here's my matlab history:</p>
<p data-pm-slice="0 0 []">PPPI Version: 13.1.4-17-2014</p>
<p>Parameters used:</p>
<p>Processing subject: sub-8002</p>
<p>VOI file : D:\fmri2\ROI\right_amygdala.nii</p>
<p>Output file will be: sub-8002_ramy_session#_cond_PPI_regressors.txt</p>
<p>Contrast : No adjustment</p>
<p>Analysis : Psychophysiological Interactions</p>
<p>Extraction : eigenvariate</p>
<p>Tasks : ALL</p>
<p>Method : Condition Specific</p>
<p>VOI has 123 voxels in 2x2x2 space</p>
<p>VOI has 119 voxels in 2x2x2 space. This is in the the same space as the input data and functional mask.</p>
<p>VOI has 123 voxels in 2x2x2 space</p>
<p>VOI has 119 voxels in 2x2x2 space. This is in the the same space as the input data and functional mask.</p>
<p>VOI has 123 voxels in 2x2x2 space</p>
<p>VOI has 119 voxels in 2x2x2 space. This is in the the same space as the input data and functional mask.</p>
<p>Generate Contrast Vectors</p>
<p>Valid Contrast</p>
<p>Generate xCon</p>
<p>Generate Contrasts</p>
<p>contrast image 1 : ...written con_0001.nii</p>
<p>spm{T} image 1 : ...written spmT_0001.nii</p>
<p>Moving Contrast Images</p>
<p>Time to Move</p>
<p>Will move con_0001.nii to con_PPI_face > shape_sub-8002.nii</p>
<p>Now attempting move</p>
<p>error moving contrast PPI_face > shape</p>
<p>error moving spmT/F PPI_face > shape</p>
<p> </p>
<p>Why can't Mr.nii move right? wait for expert's answer</p>

gPPI Group Level Analysis - SPM level-2 or something different?

jebezek — Fri, 16 Aug 2024 12:51:58 GMT

<p>Hello,</p>
<p> </p>
<p>I have recently completed gPPI analyses in MATLAB utilizing the PPPI.m script downloaded from NITRC and I have generated individual-level files for each subject in my analysis. I am using the ventral striatum (VS) as a seed region and I have three contrasts specified within my analysis based on task condition.</p>
<p> </p>
<p>Now that I have individual-level output, I am interested in completing a group-level analysis in SPM; however, I am not sure if the individual gPPI output is appropriate for a regular level 2 analysis. I have two SPM.mat files per subject, one for the left VS and one for the right VS. How can I examine these results at the group level to identify the regions significantly coupled with my seed region of interest during my conditions of interest across the entire sample?</p>
<p> </p>
<p>Thank you in advance for your assistance.</p>

RE: Subject-specific VOIs created before

Qian Liu — Tue, 16 Apr 2024 19:38:06 GMT

<p>Dear Dr. McLaren,</p>
<p>When I use the settings about constraining the mask to be within the subject and group clusters you mentioned, I get this error "Field assignment is assigned to a non-structure array object." Could you please help me to check what is wrong with my settings?</p>
<p>Thank you!</p>
<p>Qian Liu</p>

RE: PPI between two ROIS and PPI connectivity values.

Qian Liu — Sat, 13 Apr 2024 21:31:59 GMT

<p>I think you could use CONN functional toolbox to get the matrix between two ROIs.</p>

PPI analysis-how to do ROI-to-mask (not whole brain) analysis?

Qian Liu — Sat, 13 Apr 2024 21:29:22 GMT

<p>I am performing the task modulated PPI analysis, and I want to perform the analysis between ROIs and a small mask, to determine which part of this small mask(not the whole brain) related to the ROIs. I am using the gPPI toolbox. Sincerely hope someone can answer. Thanks</p>

xY field name error

Rachel Ferry — Fri, 05 Apr 2024 18:05:31 GMT

<p>Hello,</p>
<p>I'm trying to run the SPM PPI function, but when I select the VOI that I created in marsbar it gives an error saying that the 'xY' variable is not found (see below). I've read through multiple posts re: the toolbox, but can't seem to find anyone else that has encountered this specific issue. Any suggests on steps I could take to address this would be greatly appreciated!</p>
<p>Error:</p>
<p>Warning: Variable 'xY' not found. <br>> In spm_peb_ppi (line 237) <br>Unrecognized field name "xY".</p>
<p>Error in spm_peb_ppi (line 240)<br>    xY(1) = p.xY;<br> <br>Error while evaluating UIControl Callback.</p>
<p> </p>
<p>-Rachel</p>

gPPI -SPM estimation fails - please help!

Sophie Lev — Tue, 02 Apr 2024 0:32:17 GMT

Hi Dr McLaren (and everyone),

Thank you for providing the gPPI toolbox.

I have an event-related task with 4 conditions, where each trial randomly presented and is separated with an intertrial interval (rest/fixation-cross) period. The 1^st level model space has conditions specified as the regressors of interest, and noise regressors of no interest.

I have successfully managed to run the example data upon downloading the toolbox and changing a couple of the scripts with the scripts I collected from the posts in this forum. However, when I run the toolbox on my data, I ran into errors. I managed to fix some of them but the issue still continues.

In running the data, I used the following P structure;

P.subject: '152'

P.directory: '/path/to/subject/1st_level'

P.VOI:'/path/to/subject /PPI_1st/masks/vmPFC_mask.nii'

P.Region: 'vmPFC'

P.Estimate: 1

P.extract: 'eig'

P.Tasks: {'0' 'CondA' 'CondB' 'CondC' 'CondD'}

P.Weights: []

P.maskdir: '/path/to/subject/1st_level'

P.equalroi: 0

P.FLmask: 0

P.analysis: 'psy'

P.method: 'cond'

P.CompContrasts: 0

P.Contrasts(1).name='PPI_CondB_CondA

P.Contrasts(1).left='CondB'

P.Contrasts(1).right='CondA'

P.Contrasts(1).MinEvents=20

P.Contrasts(1).STAT='T'

P.Contrasts(2).name='PPI_CondC_CondA'

P.Contrasts(2).left='CondC'

P.Contrasts(2).right='CondA'

P.Contrasts(2).MinEvents=20

P.Contrasts(2).STAT='T'

P.Contrasts(3).name='PPI_CondD_CondA'

P.Contrasts(3).left='CondD'

P.Contrasts(3).right='CondA'

P.Contrasts(3).MinEvents=20

1) When I run gPPI on my data, I got the output below, then estimation failed. It creates SPM.mat file with missing fields. Looking at the SPM fields created I realized the toolbox seemed not to extract the timeseries from the mask (Timeseries regressors for the mask is all NaN).

2) Adding P.preservevarcorr=1 to the ‘P’ parameter structure (preserving variance correction estimated from the 1^st level), lets the estimation go through a little bit more, but the estimation still gets stuck at the estimation of spatial non-sphericity.

I am not sure whether estimation fails because of a problem with the time-series extraction or the timeseries extraction fails because of the estimation problem.

Any advice, insight and help would be greatly appreciated!!

###################################

Valid Contrast