-
Notifications
You must be signed in to change notification settings - Fork 0
/
savePermPop_expand.m
216 lines (189 loc) · 10 KB
/
savePermPop_expand.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
% savePermPop_expand.m
%
% associated with the following publication: Roth, ZN, and Merriam, EP (2023).
% Representations in human primary visual cortex drift over time
% DOI:
%
% usage: savePermPop_expand(1,0,1000)
% by: zvi roth
% date: 3/10/2022
% purpose: permute the session order and compute measures of population representational drift
% uses files created by: simPopResponse_expand.m
% creates files used by:
function res = savePermPop_expand(rois,toZscore,nperms,r2thresh,fixedFirst)
tic
if ieNotDefined('rois'), rois = [1]; end
if ieNotDefined('toZscore'), toZscore = 0; end%
if ieNotDefined('nperms'), nperms = 2; end
if ieNotDefined('r2thresh'), r2thresh = 0; end
if ieNotDefined('fixedFirst'), fixedFirst = 0; end
fixedFirstStr='';
if fixedFirst
fixedFirstStr = '_fixedFirst_';
end
rng(1)
saveFolder = fullfile('~','misc','data18','rothzn','nsd','repDrift_expand','/');
if ~isfolder(saveFolder)
saveFolder = ['/misc/data18/rothzn/nsd/repDrift_expand/'];
end
corrTypeBetweenRDMs = 'Spearman';%'Pearson'; %for comparing RDMs across sessions
distTypeRDM = 'correlation';%'correlation', 'euclidean', 'mahalanobis', 'spearman'. For creating RDM
nrois = 4;
subjects = 1:8;%[5:8];
nsubjects = length(subjects);
nimg = 100;
nsessions=30;
maxSessions = nsessions;
minSessions=nsessions;
r2threshStr = '';
if r2thresh>0
r2threshStr = ['r2thresh' num2str(r2thresh,'%4.2f')];
end
errorbarColor = [0.2 0.2 0.2];
surfaceAlpha = 0.1;
linewidth = 2;
zscoreStr='';
if toZscore==1
zscoreStr = '_zscore';
elseif toZscore==2
zscoreStr = '_zeroMean';
elseif toZscore==3
zscoreStr = '_equalStd';
elseif toZscore==4
zscoreStr = '_zeroROImean';
end
imgCorrMat = NaN(nrois,length(subjects),nimg,nsessions,nsessions);
imgCorrMatOri = NaN(nrois,length(subjects),nimg,nsessions,nsessions);
imgCorrVec = NaN(nrois,length(subjects),nimg,nsessions*(nsessions-1)/2);
imgCorrVecOri = NaN(nrois,length(subjects),nimg,nsessions*(nsessions-1)/2);
sessCorrMat = NaN(nrois,length(subjects),nsessions,nimg,nimg);
sessCorrMatOri = NaN(nrois,length(subjects),nsessions,nimg,nimg);
sessCorrVec = NaN(nrois,length(subjects),nsessions,nimg*(nimg-1)/2);
sessCorrVecOri = NaN(nrois,length(subjects),nsessions,nimg*(nimg-1)/2);
betweenSessCorr = NaN(nrois,length(subjects),nsessions,nsessions);
betweenSessCorrOri = NaN(nrois,length(subjects),nsessions,nsessions);
for isub=1:length(subjects)
isub
load(fullfile(saveFolder,['regressSessCombineROI_sub' num2str(subjects(isub)) zscoreStr '.mat']),'allRoiPrf','nsplits');
nsessions = nsplits-1; %maxnumber of sessions. nsplits includes the mean, and is <41 for some subjects
subSessions(isub,1:nsessions) = ones;
for iroi=rois%1:nrois
load(fullfile(saveFolder,['simPopResp_v' num2str(iroi) '_sub' num2str(subjects(isub)) zscoreStr '.mat']),...
'voxSessResp','voxSessRespOri', 'simImgs','nsessions');
subNumSessions(isub) = nsessions;
goodVox = allRoiPrf{iroi}.r2>r2thresh;
r2GoodVox(iroi,isub) = r2thresh;
numGoodVox(iroi,isub) = sum(goodVox);
for iimg=1:nimg
%for each image, a correlation matrix between sessions
imgCorrMat(iroi,isub,iimg,1:nsessions,1:nsessions) = corr(squeeze(voxSessResp(goodVox,:,iimg)));
imgCorrMatOri(iroi,isub,iimg,1:nsessions,1:nsessions) = corr(squeeze(voxSessRespOri(goodVox,:,iimg)));
imgCorrVec(iroi,isub,iimg,1:nsessions*(nsessions-1)/2) = pdist(squeeze(voxSessResp(goodVox,:,iimg))',distTypeRDM);
imgCorrVecOri(iroi,isub,iimg,1:nsessions*(nsessions-1)/2) = pdist(squeeze(voxSessRespOri(goodVox,:,iimg))',distTypeRDM);
end
for isess=1:nsessions
%RDM of all images in a single session
sessCorrMat(iroi,isub,isess,:,:) = corr(squeeze(voxSessResp(goodVox,isess,:)));
sessCorrMatOri(iroi,isub,isess,:,:) = corr(squeeze(voxSessRespOri(goodVox,isess,:)));
sessCorrVec(iroi,isub,isess,:) = pdist(squeeze(voxSessResp(goodVox,isess,:))',distTypeRDM);
sessCorrVecOri(iroi,isub,isess,:) = pdist(squeeze(voxSessRespOri(goodVox,isess,:))',distTypeRDM);
end
%correlate between RDMs of different sessions
betweenSessCorr(iroi,isub,1:minSessions,1:minSessions) = corr(squeeze(sessCorrVec(iroi,isub,1:minSessions,1:minSessions))','Type',corrTypeBetweenRDMs);
betweenSessCorrOri(iroi,isub,1:minSessions,1:minSessions) = corr(squeeze(sessCorrVecOri(iroi,isub,1:minSessions,1:minSessions))','Type',corrTypeBetweenRDMs);
end
end
minSessions = min(subNumSessions);
%average across images
avgImgCorrMat = squeeze(mean(imgCorrMat,3));
avgImgCorrMatOri = squeeze(mean(imgCorrMatOri,3));
%initialize permutation matrices of average across images.
imgCorrMatPerm = NaN(nrois,length(subjects),nperms,minSessions,minSessions);
imgCorrMatOriPerm = NaN(nrois,length(subjects),nperms,minSessions,minSessions);
%initialize permutation matrices of correlations between RDMs.
betweenSessCorrPerm = NaN(nrois,length(subjects),nperms,minSessions,minSessions);
betweenSessCorrOriPerm = NaN(nrois,length(subjects),nperms,minSessions,minSessions);
%%
distMatrix = toeplitz(0:nsessions-1);
betweenSessImg = NaN(nrois,length(subjects),minSessions-1);
betweenSessImgOri = NaN(nrois,length(subjects),minSessions-1);
betweenSessImgPerm = NaN(nrois,length(subjects),nperms,minSessions-1);
betweenSessImgOriPerm = NaN(nrois,length(subjects),nperms,minSessions-1);
betweenSessDist = NaN(nrois,length(subjects),minSessions-1);
betweenSessDistOri = NaN(nrois,length(subjects),minSessions-1);
betweenSessDistPerm = NaN(nrois,length(subjects),nperms,minSessions-1);
betweenSessDistOriPerm = NaN(nrois,length(subjects),nperms,minSessions-1);
for iroi=rois%1:nrois
%correlations between population responses
for isub=1:length(subjects)
nsessions = subNumSessions(isub);
if isub>1%use subject 1 randomized orders
permOrders{isub} = permOrders{1};
else%randomize for subject 1
for iperm=1:nperms
if fixedFirst
permOrders{isub}(iperm,:) = [1 1+randperm(minSessions-1)];%ONLY PERMUTING FIRST 30 SESSIONS
else
permOrders{isub}(iperm,:) = randperm(minSessions);%ONLY PERMUTING FIRST 30 SESSIONS
end
end
end
temp = squeeze(avgImgCorrMat(iroi,isub,1:minSessions,1:minSessions));
tempOri = squeeze(avgImgCorrMatOri(iroi,isub,1:minSessions,1:minSessions));
for idist=1:minSessions-1
betweenSessImg(iroi,isub,idist) = nanmean(temp(distMatrix==idist));
betweenSessImgOri(iroi,isub,idist) = nanmean(tempOri(distMatrix==idist));
end
if toNormalize
initIndex = 1;
betweenSessImg(iroi,isub,:) = (betweenSessImg(iroi,isub,:) - betweenSessImg(iroi,isub,initIndex))./abs(betweenSessImg(iroi,isub,:) + betweenSessImg(iroi,isub,initIndex));
betweenSessImgOri(iroi,isub,:) = (betweenSessImgOri(iroi,isub,:) - betweenSessImgOri(iroi,isub,initIndex))./abs(betweenSessImgOri(iroi,isub,:) + betweenSessImgOri(iroi,isub,initIndex));
end
for iperm=1:nperms
permOrder = permOrders{isub}(iperm,:);%different for each subject, but same for all ROIs
tempPerm = squeeze(betweenSessCorr(iroi,isub,permOrder,permOrder));
tempOriPerm = squeeze(betweenSessCorrOri(iroi,isub,permOrder,permOrder));
for idist=1:minSessions-1
betweenSessImgPerm(iroi,isub,iperm,idist) = nanmean(tempPerm(distMatrix==idist));
betweenSessImgOriPerm(iroi,isub,iperm,idist) = nanmean(tempOriPerm(distMatrix==idist));
end
imgCorrMatPerm(iroi,isub,iperm,:,:) = tempPerm;
imgCorrMatOriPerm(iroi,isub,iperm,:,:) = tempOriPerm;
end
end
%correlations between RDMs
for isub=1:length(subjects)
temp = squeeze(betweenSessCorr(iroi,isub,1:minSessions,1:minSessions)); %includes NaNs for sessinos that don't exist
tempOri = squeeze(betweenSessCorrOri(iroi,isub,1:minSessions,1:minSessions));%includes NaNs for sessinos that don't exist
for idist=1:minSessions-1
betweenSessDist(iroi,isub,idist) = nanmean(temp(distMatrix==idist));
betweenSessDistOri(iroi,isub,idist) = nanmean(tempOri(distMatrix==idist));
end
if toNormalize
initIndex = 1;
betweenSessDist(iroi,isub,:) = (betweenSessDist(iroi,isub,:) - betweenSessDist(iroi,isub,initIndex))./abs(betweenSessDist(iroi,isub,:) + betweenSessDist(iroi,isub,initIndex));
betweenSessDistOri(iroi,isub,:) = (betweenSessDistOri(iroi,isub,:) - betweenSessDistOri(iroi,isub,initIndex))./abs(betweenSessDistOri(iroi,isub,:) + betweenSessDistOri(iroi,isub,initIndex));
end
for iperm=1:nperms
permOrder = permOrders{isub}(iperm,:);%different for each subject, but same for all ROIs
tempPerm = squeeze(betweenSessCorr(iroi,isub,permOrder,permOrder));
tempOriPerm = squeeze(betweenSessCorrOri(iroi,isub,permOrder,permOrder));
for idist=1:nsessions-1
betweenSessDistPerm(iroi,isub,iperm,idist) = nanmean(tempPerm(distMatrix==idist));
betweenSessDistOriPerm(iroi,isub,iperm,idist) = nanmean(tempOriPerm(distMatrix==idist));
end
betweenSessCorrPerm(iroi,isub,iperm,:,:) = tempPerm;
betweenSessCorrOriPerm(iroi,isub,iperm,:,:) = tempOriPerm;
end
end
end
save(fullfile(saveFolder, ['permPop' fixedFirstStr num2str(nperms) zscoreStr r2threshStr '.mat']),...
'rois','toNormalize','toZscore','r2thresh','nrois', ...
'permOrders', 'subSessions', 'subjects', 'minSessions','distMatrix',...
'betweenSessCorr','betweenSessCorrOri',...
'avgImgCorrMat','avgImgCorrMatOri',...
'betweenSessImg','betweenSessImgOri','betweenSessImgPerm','betweenSessImgOriPerm',...
'betweenSessDist','betweenSessDistOri','betweenSessDistPerm','betweenSessDistOriPerm',...
'numGoodVox',...
'imgCorrMatPerm','imgCorrMatOriPerm','betweenSessCorrPerm','betweenSessCorrOriPerm');
toc