OSCR

Aberrant Neural Entrainment to Word-Level Speech Patterns in Fragile X Syndrome: Evidence for a Statistical Learning Deficit.

Code ↔ Paper

5 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 5 matches
  1. [1] § Methods › EEG Recording and Analysis › Preprocessing ↔ statistical_learning_analysis_scripts.zip/MainITCScript.m, lines 102–134 · score 0.79 · Artifact Blocking, artifact corrected, outer ring, algorithm, threshold, AB
  2. [2] § Methods › EEG Recording and Analysis › Time Course of Neural Entrainment Over Learning ↔ statistical_learning_analysis_scripts.zip/MainITCScript.m, lines 1541–1584 · score 0.62 · ITC pseudovalue, normalized jackknifed, formula, syllable, word
  3. [3] § Methods › EEG Recording and Analysis › Computation of Intertrial Phase Coherence and Statistical Analysis › Scalp‐Level Statistical Analysis ↔ statistical_learning_analysis_scripts.zip/MainITCScript.m, lines 1340–1409 · score 0.61 · Shaded error bars, standard error, scored ITC, syllable frequency, Topographical, word frequency
  4. [4] § Methods › EEG Recording and Analysis › Computation of Intertrial Phase Coherence and Statistical Analysis ↔ statistical_learning_analysis_scripts.zip/MainITCScript.m, lines 1586–1660 · score 0.59 · phase locked, word structure, coherence, epochs, ITC, syllable
  5. [5] § Results › Source‐Level Analysis › Word Entrainment ↔ statistical_learning_analysis_scripts.zip/MainITCScript.m, lines 1340–1409 · score 0.52 · Shaded error bars, standard error, scored ITC, word frequency, zITC, FXS

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 1,661 lines · 74 KB · CC-BY-4.0 · 5 matches

  1. % Apply the Infant Pipeline initially applied in Choi, Batterink 2020 to
  2. % the child data from FXS Project
  4. %datafile_names must remain consistent each time script is run.
  6. %notes
  7. %0770 AB algorithm seemed to create completely tiny voltages - excluded anyway though
  9. %% Read files to analyse
  11. clear % clear matlab workspace
  12. clc % clear matlab command window
  14. CurrentDirectory = '/Volumes/Laura_data/Laura Backup/eeglab 2024/';
  15. cd(CurrentDirectory)
  16. eeglab
  18. %Enter the path of the folder that has the raw data to be analyzed
  19. rawdata_location = [CurrentDirectory 'FXS Project/incoming/structured'];
  21. datafile_names=dir(rawdata_location);
  22. datafile_names=datafile_names(~ismember({datafile_names.name},{'.', '..', '.DS_Store'}));
  23. datafile_names={datafile_names.name};
  24. [filepath,name,ext] = fileparts(char(datafile_names{1}));
  26. %Channel Locations
  27. channel_locations = [CurrentDirectory 'FXS Project/channel location file/GSN-HydroCel-129.sfp'];
  29. %Outer ring of channels:
  30. outerlayer_channel = {'E17' 'E38' 'E43' 'E44' 'E48' 'E49' 'E113' 'E114' 'E119' 'E120' 'E121' 'E125' 'E126' 'E127' 'E128' 'E56' 'E63' 'E68' 'E73' 'E81' 'E88' 'E94' 'E99' 'E107'}; % list of channels
  32. %Codes:
  33. %1-12-10
  34. %11-2-4
  35. %8-3-5
  36. %9-7-6
  38. SyllableCodeStrings = {'DIN1', 'DIN2', 'DIN3', 'DIN4', 'DIN5', 'DIN6', 'DIN7', 'DIN8', 'DIN9', 'DI10', 'DI11', 'DI12', ...
  39. 'D101', 'D102', 'D103', 'D104', 'D105', 'D106', 'D107', 'D108', 'D109', 'D110', 'D111', 'D112'}; %most subjects have the top row of codes, 2310 has bottom row
  40. WordOnsetStrings = {'DIN1', 'DIN8', 'DIN9', 'DI11', ...
  41. 'D101', 'D108', 'D109', 'D111'};
  44. %algorithm parameters for AB algorithm.
  45. threshold = 50; %50 originally; %0 code for didn't apply AB algorithm at all. 999 code for filter from 0.1 to 30 and no alogirthm.
  46. approach = 'total'; %single smoothing matrix used
  49. % ---- Subject Information ---
  50. TDC = {'2481', '1646', '3367', '2860', '1844', '2920', '1136', '2808', '1498', '1655', '2577', '2802', '1343', '1430', ...
  51. '3260', '2181', '1025', '1692', '0076', '0439', '2835', '3186', '0694', '1466', '3107', '2723', '1008', '2313', '1864', '1340', '2805'};
  53. FXS = {'2978', '0836', '1721', '0649', '0468', '0871', '2195', '0913', '1649', '2151', '2472', '2604', '1037', '0392', '1841', '2238', '1798'};
  54. %ASD = {'3463', '2219', '0409', '0502', '0114', '1292', '2410', '2866', '2339', '2100', '0878', '1963','2215', '1995', '2695'};
  56. AllSubjects = [TDC, FXS, ASD];
  58. NewSubjectIDs = {'1798', '1841', '2238', '1963', '1995', '2215', '2695'};
  60. NewSubjectIndices = find(contains(datafile_names, NewSubjectIDs));
  61. AllSubjectIndices = find(contains(datafile_names, [TDC, FXS])); %for now just run TDC and FXS, run ASD later
  64. %% Preprocessing following Fujioka et al. 2011
  66. for subjecti = NewSubjectIndices
  68. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[]; %clear any dataset currently loaded in GUI.
  69. EEG=mff_import([rawdata_location filesep datafile_names{subjecti}]);
  70. [ALLEEG EEG CURRENTSET] = pop_newset( ALLEEG, EEG, CURRENTSET, 'gui','off');
  72. %import channel locations
  73. EEG=pop_chanedit(EEG, 'load',{channel_locations 'filetype' 'autodetect'});
  74. EEG = eeg_checkset( EEG );
  76. % Check whether the channel locations were properly imported. The EEG signals and channel numbers should be same.
  77. if size(EEG.data, 1) ~= length(EEG.chanlocs)
  78. error('The size of the data does not match with channel numbers.');
  79. end
  81. %remove outer ring of channels (see Fujioka et al. 2011)
  82. EEG = pop_select( EEG,'nochannel', outerlayer_channel);
  84. %filter between 0.5 and 20 Hz - same as before.
  86. %ton of 60 hz noise
  87. %filter first with notch, then with buttersworth filter?
  88. EEG = pop_basicfilter( EEG, 1:EEG.nbchan, 'Boundary', 'boundary', 'Cutoff', 60, 'Design', 'notch', 'Filter', 'PMnotch', 'Order', 180, 'RemoveDC', 'on' ); % GUI: 07-Jun-2017 18:55:44
  90. %filter the data using ERPLAB [ 0.5 20]
  91. EEG = pop_basicfilter( EEG, 1:EEG.nbchan, 'Cutoff', [ 0.5 20], 'Design', 'butter', 'Filter', 'bandpass', 'Order', 2, 'RemoveDC', 'on' ); % GUI: 15-Jan-2014 16:49:27
  93. %save the dataset into the appropriate folder
  94. SubjectName = datafile_names{subjecti}(1:4);
  96. EEG.setname = [SubjectName '_preprocessed'];
  97. EEG = pop_saveset( EEG, 'filename', [SubjectName '_preprocessed'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/preprocessed/']); %
  99. end
  102. %% APPLY ARTIFACT BLOCKING (AB) ALGORITHM
  105. for subjecti = NewSubjectIndices
  107. SubjectName = datafile_names{subjecti}(1:4);
  109. %load preprocessed dataset
  110. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  111. EEG = pop_loadset( 'filename', [SubjectName '_preprocessed.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/preprocessed/']);
  112. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  114. %outer ring of channels already removed. %remove CZ - algorithm doesn't
  115. %work if ref channel is included.
  117. EEG = pop_select( EEG,'rmchannel', {'Cz'});
  119. %call algorithm
  120. CleanEEG = AB_alg_window(EEG.data,EEG.srate,threshold, approach); %algorithm won't work properly if ref channel is included.
  122. %EEGDataOriginal = EEG.data;
  123. EEG.data = CleanEEG;
  125. %rerefernce the data - I didn't retain the Cz channel because it gave me an
  126. %warning about the same channel label.
  127. EEG = pop_reref( EEG, []);
  130. %save the dataset into the appropriate folder
  131. EEG.setname = [SubjectName '_artifact corrected'];
  132. EEG = pop_saveset( EEG, 'filename', [SubjectName '_artifact corrected'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/ABCleaned/']); %"real" subjects (not pilot subjects) will be named Infant1, Infant2, Infant3, etc.
  134. end
  136. %% CREATE EPOCHS
  138. NumSyllablesPerEpoch = 30; %syllables as in PsychSci Paper corresponding to 9 seconds.
  139. EpochLength = NumSyllablesPerEpoch*0.3; %in seconds --> this assumes presentation rate of 300ms per syllable
  142. for subjecti = NewSubjectIndices %11:length(datafile_names)
  144. SubjectName = datafile_names{subjecti}(1:4);
  146. %load preprocessed dataset
  147. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  148. EEG = pop_loadset( 'filename', [SubjectName '_artifact corrected.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/ABCleaned/']);
  149. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  151. UnacceptableJitter = 320; %1000hz sampling rate --> 320 msec rather than 300 msec SOA between syllables is not acceptable. Expect 300
  153. EventsToRemove = [];
  154. for x = 2:size(EEG.event,2) %for each code beginning at the second code (first few are start codes anyway)
  156. if strcmp(EEG.event(x).type, 'DI64')
  157. EventsToRemove = [EventsToRemove, x];
  158. end
  160. end
  162. if size(EventsToRemove, 1) > 0
  163. EEG = pop_selectevent( EEG, 'event', EventsToRemove,'select','inverse','deleteevents','on');
  164. end
  166. %epoch to every 10th word (every 30 syllables in the random condition)
  168. y = 5; %don't include the very first syllable due to sharp auditory onset response; skip over three "start codes" plus first syllable
  170. while y < size(EEG.event,2) %number of events
  172. if contains(EEG.event(y).type, WordOnsetStrings) %ismember( EEG.event(y).type, WordOnsetStrings)
  173. CandidateEpochOnset = y; %found a candidate code to onset to.
  175. syllablecounter = 0;
  176. for z = y:size(EEG.event,2) %candidate code is syllable 1, and so on.
  178. %if a pause or break in auditory stimulation is detected
  179. if EEG.event(z).latency - EEG.event(z-1).latency > UnacceptableJitter %155 %unacceptable level of stim jitter or pause. %should typically have an interstim interval of 154 sampling points (300 ms) --> if not, pause is detected (165 originally)
  180. disp('break in auditory stimulation detected!')
  181. disp(EEG.event(z))
  182. syllablecounter = 0; %reset
  183. y = z;
  184. break
  185. end
  187. if contains( EEG.event(z).type, SyllableCodeStrings)
  188. syllablecounter = syllablecounter + 1;
  190. if syllablecounter == NumSyllablesPerEpoch % 36 syllables without running into a boundary or other unexpected code
  191. EEG = pop_editeventvals(EEG,'changefield',{CandidateEpochOnset,'type','NE'});%NE - New Epoch
  192. disp(CandidateEpochOnset)
  193. y = z; %search for next candidate epoch onset 30 positions later.
  194. break
  195. end
  196. elseif ~ismember( EEG.event(z).type, SyllableCodeStrings) %a 'boundary' event --> reset the counter in this case as it needs to be 36 continuous syllables.
  197. disp('other code found')
  198. disp(EEG.event(z).type)
  199. syllablecounter = 0; %reset
  200. y = z; %if you hit a boundary need to start looking again for a candidate epoch onset...
  201. break
  202. end
  203. end
  204. end
  206. y = y + 1;
  208. end
  210. % STEP 12: Segment data into fixed length epochs
  211. %task_event_markers = {'NE'}; %new epoch events, created in step 11.5 (Entrain)
  213. EEG = eeg_checkset(EEG);
  214. EEG = pop_epoch(EEG, {'NE'}, [0 EpochLength], 'epochinfo', 'yes');
  215. EEG = eeg_checkset( EEG );
  216. EEG = pop_rmbase( EEG, [EEG.times(1) EEG.times(end)]); %baseline correct to the entire epoch
  217. EEG = eeg_checkset( EEG );
  219. EEG.setname = [SubjectName '_epoched'];
  220. EEG = pop_saveset( EEG, 'filename', [SubjectName '_epoched'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/epochs/']); %"real" subjects (not
  223. end
  228. %% LOAD PARAMETERS (recommended by Ana) and RUN FFT
  230. % Output frequencies to keep
  231. foutput = [0.6 10];
  233. % Range of frequnecies to use to compute SNR.
  234. % It is better if it includes the foutput and if the lower limit is bigger than the high-pass filter used in the data
  235. fband = [0.2 15];
  237. % target frequencies if indeicated won't be used to compute the SNR.
  238. % Not really necessary, if left empty should give more or less the same
  239. ftarget = []; % [1.11 3.33] | []
  241. % Bins around each frequnecy bin to compute SNR
  242. binsSNR = [-5 -4 -3 -2 -1 1 2 3 4 5];
  244. % SNR method for the power
  245. % based on the linear fit of log(Power) vs log(freq) of the adjacent
  246. % freuency bins. it seems the most sensitive
  247. typeSNRPWR = 'linear';
  249. % SNR method for the PLV
  250. % In principle it shouldn't be necessary but by computing it sensitivity
  251. % may slightly improve. I put 'none' but you can try 'linear' or 'mean'
  252. typeSNRPLV = 'none';
  254. % Z-score or not the final power.
  255. % Not really necessary, only to center at zero the final data
  256. zscorePWR = 0;
  258. % Z-score or not the final PLV.
  259. % Not really necessary, only to center at zero the final data
  260. zscorePLV = 0;
  263. cd([CurrentDirectory 'FXS Project/infant pipeline/FFT files/']) %so that files go in proper place
  265. DataQualityMtx = [];
  267. for subjecti= NewSubjectIndices %1:length(datafile_names)
  269. clear EEGfrq
  271. SubjectName = datafile_names{subjecti}(1:4);
  273. %load dataset
  274. %load preprocessed dataset
  275. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  276. EEG = pop_loadset( 'filename', [SubjectName '_epoched.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/epochs/']);
  277. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  279. EEGfrq = frqa_plvpwr(EEG, 'foutput',foutput,'fband',fband,'binsSNR',binsSNR,...
  280. 'typeSNRPWR',typeSNRPWR,'typeSNRPLV',typeSNRPLV,...
  281. 'zscorePWR',zscorePWR,'zscorePLV',zscorePLV);
  283. filename = char(strcat(SubjectName, '_Structured_EEGfreq'));
  284. save(filename, 'EEGfrq'); %how to save when filename is a variable set beforehand
  286. end
  291. %% Individual FFT plots
  293. FFTfile_location = [CurrentDirectory 'FXS Project/infant pipeline/FFT files'];
  295. cd(FFTfile_location)
  297. %FFTfiles = dir('*Structured_EEGfreq.mat'); %this only considers the files that were succesfully preprocessed and not those tagged with 'no usable data, all bad epochs.'
  298. %FFTfile_names={FFTfiles.name};
  300. plotchan= 1:104; % we removed the outer ring of electrodes
  302. locsfile = [CurrentDirectory 'FXS Project/channel location file/EGI104InnerChannelsNoCz.locs']; %exported this .locs file one time from EEG dataset. writelocs( chanstruct, filename ); writelocs( EEG.chanlocs, 'testlocs' );
  304. for subji = NewSubjectIndices %1:length(FFTfile_names)
  306. cd([CurrentDirectory 'FXS Project/infant pipeline/FFT files'])
  308. %Load struct_ersp and rand_ersp for each subject
  309. clear Struct_PLV EEGfrq*
  311. SubjectName = datafile_names{subji}(1:4);
  312. %SubjectNumber = extractBetween(FFTfile_names{subji}, 1, 4);
  314. load([SubjectName '_Structured_EEGfreq.mat'])
  316. %load(FFTfile_names{subji})
  317. Struct_PLV = EEGfrq.plv(:,:);
  319. freqvals = EEGfrq.freqs; %freqs based on last file loaded
  321. freqvalindicesunder5Hz = find(freqvals <= 5);
  322. [~, WordFreq] = min(abs(freqvals-1.1111));
  323. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  325. %extract neighbouring bins
  326. NeighbouringWordFreqBins = [WordFreq-3, WordFreq-2, WordFreq-1, WordFreq+1, WordFreq+2, WordFreq+3];
  327. NeighbouringSyllableFreqBins = [SyllableFreq-3, SyllableFreq-2, SyllableFreq-1, SyllableFreq+1, SyllableFreq+2, SyllableFreq+3];
  329. %compute actual and normalized ITC values for topoplot
  331. WordITC = Struct_PLV(plotchan, WordFreq);
  332. SyllableITC = Struct_PLV(plotchan, SyllableFreq);
  334. WordNeighboursITC = squeeze(mean(Struct_PLV(plotchan, NeighbouringWordFreqBins ),2));
  335. SyllableNeighboursITC = squeeze(mean(Struct_PLV(plotchan, NeighbouringSyllableFreqBins ),2));
  337. WordITCNormalized = WordITC - WordNeighboursITC;
  338. SyllableITCNormalized = SyllableITC - SyllableNeighboursITC;
  340. cd([CurrentDirectory 'FXS Project/infant pipeline/individual plots'])
  342. figure; plot(freqvals(freqvalindicesunder5Hz), mean(Struct_PLV(plotchan,freqvalindicesunder5Hz),1), 'r');
  343. hold on
  344. xline(1.111, '--')
  345. hold on
  346. xline(2.222, '--')
  347. hold on
  348. xline(3.333, '--')
  349. hold on
  350. legend([SubjectName ' structured'])
  351. figurename = char(strcat(SubjectName, '_Structured'));
  352. print('-dpdf', figurename); %save as postscript in order to merge two files
  354. %topoplot
  356. WordFreqScaleLimits = [-0.25 .25];
  357. SyllableFreqScaleLimits = [-0.6 0.6];
  359. figure;
  360. subplot(2,2,1); topoplot(WordITC, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits , 'shading','interp'); colorbar;
  361. title([SubjectName 'ITC Word Frequency']);
  362. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  363. subplot(2,2,2); topoplot(SyllableITC, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits , 'shading','interp'); colorbar;
  364. title('ITC Syllable Frequency');
  365. subplot(2,2,3); topoplot(WordITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits , 'shading','interp'); colorbar;
  366. title('Normalized ITC Word Frequency');
  367. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  368. subplot(2,2,4); topoplot(SyllableITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits , 'shading','interp'); colorbar;
  369. title('Normalized ITC Syllable Frequency');
  370. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  371. filename = char(strcat(SubjectName, ' ITC Topoplot'));
  372. print ('-dtiff', filename);
  375. end
  380. %% GRAND AVERAGE - FFT -- ALL PARTICIPANTS - OVERALL EXPOSURE
  382. clear SubjectIDs ITCVars
  384. clear FFTfile* Structured_PLVMtx
  386. clear Structured_PLVMtx
  388. plotchan= 1:104; %SigWordElectrodes; %1:105; % we removed the outer ring of electrodes
  390. locsfile = [CurrentDirectory 'FXS Project/channel location file/EGI104InnerChannelsNoCz.locs']; %exported this .locs file one time from EEG dataset. writelocs( chanstruct, filename ); writelocs( EEG.chanlocs, 'testlocs' );
  393. FFTfile_location = [CurrentDirectory 'FXS Project/infant pipeline/FFT files'];
  395. cd(FFTfile_location)
  397. FFTfiles = dir('*Structured_EEGfreq.mat'); %this only considers the files that were succesfully preprocessed and not those tagged with 'no usable data, all bad epochs.'
  399. FFTfile_names={FFTfiles.name};
  402. SubjectList = [];
  403. GroupName = ASD; %FXS; %TDC; %FXS; %ASD %TDC; %ASD; %FXS; %TDC; %FXS; %TDC; AllSubjects; %
  405. %get subject list
  406. for x = 1:length(FFTfile_names)
  408. SubjectNumber = cell2mat(extractBetween(FFTfile_names{x}, 1, 4));
  410. if ismember(SubjectNumber, GroupName)
  411. SubjectList = [SubjectList, x];
  412. end
  414. end
  417. % -- STRUCTURED --
  418. clear PLVMtx Struct* Rand* Grand* PLV*
  420. for i = SubjectList
  422. display(i)
  423. %Load struct_ersp and rand_ersp for each subject
  424. %load([SubjectID{i} '_EEGfrq.mat'])
  425. load(FFTfile_names{i})
  427. disp(FFTfile_names{i})
  429. PLV = EEGfrq.plv(plotchan,:); %only extracts relevant channels
  431. Structured_PLVMtx(i,:,:) = PLV; %rows without corresponding subjects (e.g., 20) just contains 0s. This is taken into account below by selecting SubjectsToInclude as a dimension.
  433. % -- to extract individual level variables --
  434. freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  436. [~, WordFreq] = min(abs(freqvals-1.1111));
  437. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  439. %extract neighbouring bins
  440. NeighbouringWordFreqBins = [WordFreq-3, WordFreq-2, WordFreq-1, WordFreq+1, WordFreq+2, WordFreq+3];
  441. NeighbouringSyllableFreqBins = [SyllableFreq-3, SyllableFreq-2, SyllableFreq-1, SyllableFreq+1, SyllableFreq+2, SyllableFreq+3];
  443. %compute individual actual and normalized ITC values to store for analysis
  444. WordITC = PLV(:, WordFreq);
  445. SyllableITC = PLV(:, SyllableFreq);
  447. WordNeighboursITC = squeeze(mean(PLV(:, NeighbouringWordFreqBins ),2));
  448. SyllableNeighboursITC = squeeze(mean(PLV(:, NeighbouringSyllableFreqBins ),2));
  450. WordITCNormalized = WordITC - WordNeighboursITC;
  451. SyllableITCNormalized = SyllableITC - SyllableNeighboursITC;
  453. %average across channels
  454. MeanWordITC = mean(WordITC,1);
  455. MeanSyllableITC = mean(SyllableITC,1);
  456. MeanWordITCNormalized = mean(WordITCNormalized,1);
  457. MeanSyllableITCNormalized = mean(SyllableITCNormalized,1);
  459. ITCVars(i,:,:,:) = [MeanWordITC, MeanSyllableITC, MeanWordITCNormalized, MeanSyllableITCNormalized];
  461. %store subject and group information
  463. SubjectNumber = cell2mat(extractBetween(FFTfile_names{i}, 1, 4));
  465. if ismember(SubjectNumber, TDC)
  466. GroupNum = 1;
  467. elseif ismember(SubjectNumber, FXS)
  468. GroupNum = 2;
  469. elseif ismember(SubjectNumber, ASD)
  470. GroupNum = 3;
  471. end
  473. SubjectIDs{i,1} = SubjectNumber;
  474. SubjectIDs{i,2} = GroupNum; %also store group membership
  476. %CongruentSubjectIDs{i,:} = SubjectID{i};
  478. end
  480. %freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  482. %average across channels
  483. GrandAveragePLV_Structured = squeeze(mean(Structured_PLVMtx(SubjectList,:,:),1));
  484. GrandAveragePLV_avacrosschans_Structured = mean(GrandAveragePLV_Structured(:,:),1);
  486. %Standard Error
  487. PLV_averageacrosschans_Structured = squeeze(mean(Structured_PLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  488. STD_Structured = std(PLV_averageacrosschans_Structured(:,:)); %nfreqs
  490. SE_Structured = STD_Structured/sqrt(numel(SubjectList));
  492. NSubjects = numel(SubjectList);
  494. %---- Plots ------
  495. cd([CurrentDirectory 'FXS Project/infant pipeline/grand averages/'])
  498. %truncated
  499. freqvalindicesunder5Hz = find(freqvals <= 5);
  501. figure; plot(freqvals(freqvalindicesunder5Hz), GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz),'r');
  502. hold on
  503. %shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), 'lineprops','r') %old syntax for shadedErrorBar
  504. shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), {'Color', 'r' }, 1);
  505. hold on
  506. xline(1.111, '--')
  507. hold on
  508. xline(2.222, '--')
  509. hold on
  510. xline(3.333, '--')
  511. legend('structured', 'SE')
  512. pdfname = ['ITC FFT All Channels, n = ' num2str(NSubjects)];
  513. %pdfname = ['ITC FFT Auditory ROI'];
  514. print('-dpdf', pdfname); %save as postscript in order to merge two files
  517. % PLOT TOPOS
  519. %figure out freqs of interest
  520. [~, WordFreq] = min(abs(freqvals-1.1111));
  521. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  523. %extract neighbouring bins
  524. NeighbouringWordFreqBins = [WordFreq-3, WordFreq-2, WordFreq-1, WordFreq+1, WordFreq+2, WordFreq+3];
  525. NeighbouringSyllableFreqBins = [SyllableFreq-3, SyllableFreq-2, SyllableFreq-1, SyllableFreq+1, SyllableFreq+2, SyllableFreq+3];
  527. Structured_ITC_WordFreq = squeeze(GrandAveragePLV_Structured(:,WordFreq));
  528. Structured_ITC_SyllableFreq = squeeze(GrandAveragePLV_Structured(:,SyllableFreq));
  530. WordNeighboursITC = squeeze(mean(GrandAveragePLV_Structured(:, NeighbouringWordFreqBins ),2));
  531. SyllableNeighboursITC = squeeze(mean(GrandAveragePLV_Structured(:, NeighbouringSyllableFreqBins ),2));
  533. WordITCNormalized = Structured_ITC_WordFreq - WordNeighboursITC;
  534. SyllableITCNormalized = Structured_ITC_SyllableFreq - SyllableNeighboursITC;
  537. WordFreqScaleLimits = [-0.2 .2];
  538. SyllableFreqScaleLimits = [-0.5 0.5];
  540. NormalizedWordFreqScaleLimits = [-0.05 .05];
  541. NormalizedSyllableFreqScaleLimits = [-0.4 0.4];
  543. PValueLimits = [0 0.1];
  546. figure;
  547. %regular raw values
  548. subplot(2,2,1); topoplot(Structured_ITC_WordFreq, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits , 'shading','interp'); colorbar;
  549. title('Structured ITC Word Frequency');
  550. subplot(2,2,2); topoplot(Structured_ITC_SyllableFreq, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits , 'shading','interp'); colorbar;
  551. title('Structured ITC Syllable Frequency');
  552. %normalized
  553. subplot(2,2,3); topoplot(WordITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', NormalizedWordFreqScaleLimits , 'shading','interp'); colorbar;
  554. title('ITC Word Normalized');
  555. subplot(2,2,4); topoplot(SyllableITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', NormalizedSyllableFreqScaleLimits , 'shading','interp'); colorbar;
  556. title('ITC Syllable Normalized');
  557. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  558. filename = ['Condition x Frequency Topo Plots n = ' num2str(NSubjects)];
  559. print ('-dtiff', filename);
  561. %figure out which electrodes are significant
  562. NormalizedWordITC_SubjectByChannel = Structured_PLVMtx(SubjectList,:,WordFreq) - mean(Structured_PLVMtx(SubjectList,:,NeighbouringWordFreqBins),3); %NSubject x NChan
  563. NormalizedSyllableITC_SubjectByChannel = Structured_PLVMtx(SubjectList,:,SyllableFreq) - mean(Structured_PLVMtx(SubjectList,:,NeighbouringSyllableFreqBins),3);
  565. [SigWordElectrodesArray, PWord] = ttest(NormalizedWordITC_SubjectByChannel, 0, 'tail', 'right'); %one directional t-test, mean is greater than 0
  566. [SigSyllableElectrodesArray, PSyllable] = ttest(NormalizedSyllableITC_SubjectByChannel, 0, 'tail', 'right'); %one directional t-test, mean is greater than 0
  568. SigWordElectrodes = find(SigWordElectrodesArray ==1);
  569. SigSyllableElectrodes = find(SigSyllableElectrodesArray ==1);
  571. NumSigWordElectrodes = numel(SigWordElectrodes);
  573. %plot significant electrodes
  575. figure;
  576. subplot(1,2,1)
  577. topoplot(WordITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', NormalizedWordFreqScaleLimits,'shading','interp', 'emarker2', {SigWordElectrodes ,'p','k', 7, 1} ); colorbar;
  578. title(['Significant Word ITC Electrodes: ' num2str(NumSigWordElectrodes)]);
  579. subplot(1,2,2)
  580. topoplot(SyllableITCNormalized, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', NormalizedSyllableFreqScaleLimits,'shading','interp', 'emarker2', {SigSyllableElectrodes ,'p','k', 7, 1} ); colorbar;
  581. title(['Significant Syllable ITC Electrodes']);
  582. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  583. filename = ['Sig Electrodes Topo Plots n = ' num2str(NSubjects)];
  584. print ('-dtiff', filename);
  586. %if you want you could run it from the top one more time to plot the ITC
  587. %graph with only the significant word electrodes
  589. % %create a table
  590. % ITCVarsCell = num2cell(ITCVars);
  591. % ITCVarsWithSubjectIDs = horzcat(SubjectIDs, ITCVarsCell);
  592. %
  593. % ColumnNames = {'Subject', 'Group', 'WordITC', 'SyllableITC', 'WordITCNormalized', 'SyllableITCNormalized'};
  594. % IndividualLevelVariablesTableForAnalysis = cell2table(ITCVarsWithSubjectIDs(1:end,:), 'VariableNames', ColumnNames);
  595. %
  596. % LocationForCSV = [CurrentDirectory 'FXS Project/infant pipeline/grand averages/stats for analysis']; %on the server
  597. % cd(LocationForCSV)
  598. %
  599. % writetable(IndividualLevelVariablesTableForAnalysis,'IndividualLevelVariablesTableForAnalysis.csv')
  603. %% CREATE 100 SURROGATE DATASETS PER PARTICIPANT
  605. %load the same FFT parameters as above.
  607. EpochLength = 9;
  609. cd([CurrentDirectory 'FXS Project/infant pipeline/shuffled FFT files/']) %so that files go in proper place
  612. NumIterations = 100;
  614. for subjecti = 1:length(datafile_names)
  616. clear EEGfrq_ByIteration
  618. disp(subjecti)
  620. %load dataset
  621. SubjectName = datafile_names{subjecti}(1:4);
  622. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  623. EEG = pop_loadset( 'filename', [SubjectName '_epoched.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/epochs/']);
  624. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  627. %for each 200 "actual onset" event code, get the "urevent" - in EGI, this
  628. %is stored as EEG.event.description
  630. %double check to make sure each epoch has the correct number of events
  631. OnsetCodeStrings = {'NE'};
  632. EventDescriptionMtx = [];
  633. for x = 1:size(EEG.event,2) %epoch
  634. if ismember( EEG.event(x).type, OnsetCodeStrings )
  635. EventDescription = str2double(EEG.event(x).description);
  636. EventDescriptionMtx = [EventDescriptionMtx, EventDescription];
  637. end
  638. end
  640. EventDescriptionMtxUnique = unique(EventDescriptionMtx); %remove duplicates
  642. %load preprocessed AB corrected continuous data
  643. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  644. EEG = pop_loadset( 'filename', [SubjectName '_artifact corrected.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/ABCleaned/']);
  645. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  647. ParentEEG = EEG; %parentEEG will now be used for all iterations
  649. %for each iteration
  651. for iter = 1:NumIterations
  653. disp(iter)
  655. %for each UR Event identified above (corresponding to a 200 code), randomly select an onset from -1 to 1 seconds before or
  656. %after to shuffle the timing while preserving the general sleep state
  657. DatapointLength = 1000/EEG.srate;
  658. %plus or minus 900 ms (word rate) from the true event
  659. ShufflingWindowMin = -1*900/(DatapointLength); % -900;
  660. ShufflingWindowMax = 900/(DatapointLength); %+900 datapoints (EEG.srate = 1000).
  662. EventsToAdd = [];
  663. for x = 1:size(ParentEEG.event,2) %number of events
  664. if ismember( str2double(ParentEEG.event(x).description), EventDescriptionMtxUnique)
  665. %randomly select an interval +/- 1.0 seconds
  666. RandomInterval = ShufflingWindowMin + (ShufflingWindowMax-ShufflingWindowMin).*rand(1); %in data points
  667. ShuffledOnsetLatency = ParentEEG.event(x).latency + RandomInterval; %in datapoints
  668. EventsToAdd = [EventsToAdd; 300, ShuffledOnsetLatency];
  669. end
  670. end
  672. %insert new events (300 codes) that are +/- 900 msec prior to the real onsets
  673. EEG = pop_importevent(ParentEEG, 'event', EventsToAdd, 'fields', {'type', 'latency'}, 'append', 'yes', 'optimalign', 'off', 'timeunit', NaN ); %datapoints
  674. EEG = eeg_checkset( EEG );
  676. %epoch to all 300 codes.
  677. EEG = pop_epoch( EEG, { '300'}, [0 EpochLength], 'epochinfo', 'yes'); %epoch all the selected events.
  678. EEG = eeg_checkset( EEG );
  679. EEG = pop_rmbase( EEG, [EEG.times(1) EEG.times(end)]); %baseline correct to the entire epoch (better for running ICA).
  680. EEG = eeg_checkset( EEG );
  682. %double check to make sure each epoch has the correct number of events
  683. for x = 1:size(EEG.epoch,2) %epoch
  684. NumSyllables = 0; %each epoch should have 30 or 31 syllables exactly (31 because the next syllable might just make it into the epoch)
  685. for y = 1:size(EEG.epoch(x).event,2) %event within each epoch
  686. if ismember( EEG.epoch(x).eventtype{y}, SyllableCodeStrings )
  687. NumSyllables = NumSyllables + 1;
  688. end
  689. end
  691. if NumSyllables <30|| NumSyllables >31
  692. disp(['Epoch Error! Epoch to Remove:' num2str(x) ])
  693. disp(['Epoch as only ' num2str(NumSyllables) ' syllable codes, which can happen at the end of a naturally occurring block.'])
  694. else
  695. disp('All epochs contain the expected number of syllable codes!')
  696. end
  697. end
  700. %run FFT
  701. clear EEGfrq
  703. EEGfrq = frqa_plvpwr(EEG, 'foutput',foutput,'fband',fband,'binsSNR',binsSNR,...
  704. 'typeSNRPWR',typeSNRPWR,'typeSNRPLV',typeSNRPLV,...
  705. 'zscorePWR',zscorePWR,'zscorePLV',zscorePLV);
  707. %for the first iteration ONLY, create the big structure
  708. %this will be filled in eventually with all iterations
  709. if iter == 1
  710. EEGfrq_ByIteration.iteration(1:NumIterations) = EEGfrq; %in order to have a bigger structure with the same structure
  711. disp('creating the big structure for all subsequent iterations...')
  712. end
  715. EEGfrq_ByIteration.iteration(iter) = EEGfrq;
  717. end %end final iteration
  719. % save the big structure
  720. filename = char(strcat(SubjectName, '_EEGfreq_ByIteration'));
  721. save(filename, 'EEGfrq_ByIteration'); %how to save when filename is a variable set beforehand
  723. end
  730. %% EXTRACT PLV FROM TRUE DATA AND SURROGATE DATA, AND PLOT
  732. clear SubjectIDs ITCVars
  734. clear FFTfile* Structured_PLVMtx
  736. clear Structured_PLVMtx
  738. plotchan= 1:104; %SigWordElectrodes; %1:105; % we removed the outer ring of electrodes
  740. locsfile = [CurrentDirectory 'FXS Project/channel location file/EGI104InnerChannelsNoCz.locs']; %exported this .locs file one time from EEG dataset. writelocs( chanstruct, filename ); writelocs( EEG.chanlocs, 'testlocs' );
  743. FFTfile_location = [CurrentDirectory 'FXS Project/infant pipeline/FFT files'];
  745. cd(FFTfile_location)
  747. FFTfiles = dir('*Structured_EEGfreq.mat'); %this only considers the files that were succesfully preprocessed and not those tagged with 'no usable data, all bad epochs.'
  749. FFTfile_names={FFTfiles.name};
  752. FFTsurrogatefile_location = [CurrentDirectory 'FXS Project/infant pipeline/shuffled FFT files'];
  755. SubjectList = [];
  756. GroupName = FXS; %ASD %TDC; %ASD; %FXS; %TDC; %FXS; %TDC; AllSubjects; %
  758. %get subject list
  759. for x = 1:length(FFTfile_names)
  761. SubjectNumber = cell2mat(extractBetween(FFTfile_names{x}, 1, 4));
  763. if ismember(SubjectNumber, GroupName)
  764. SubjectList = [SubjectList, x];
  765. end
  767. end
  771. % -- STRUCTURED --
  772. clear PLVMtx Struct* Rand* Grand* PLV*
  774. TruePLVMtx = [];
  775. SurrogatePLVMtx = [];
  776. PLVSurrogate = [];
  778. for i = SubjectList
  779. clear EEGfrq* SurrogateDataFileName SurrogateDatasetsMtx
  780. display(i)
  782. %load "true" data
  783. cd(FFTfile_location)
  784. load(FFTfile_names{i})
  785. disp(FFTfile_names{i})
  787. PLVTrue = EEGfrq.plv(plotchan,:); %only extracts relevant channels
  788. TruePLVMtx(i,:,:) = PLVTrue; %rows without corresponding subjects (e.g., 20) just contains 0s. This is taken into account below by selecting SubjectsToInclude as a dimension.
  791. %load corresponding surrogate data
  792. cd(FFTsurrogatefile_location)
  793. SurrogateDataFileName = [char(extractBetween(FFTfile_names{i}, 1,4)) '_EEGfreq_ByIteration.mat'];
  794. load(SurrogateDataFileName)
  796. for iterationnum = 1:size(EEGfrq_ByIteration.iteration,2)
  797. SurrogateDatasetsMtx(iterationnum,:,:) = EEGfrq_ByIteration.iteration(iterationnum).plv(plotchan,:); %nchan x nfreqs for each surrogate dataset; ndataset x nchan x nfreq
  798. end
  800. PLVSurrogate = squeeze(mean(SurrogateDatasetsMtx,1)); %average across the 5 surrogate datasets for the subject
  802. SurrogatePLVMtx(i,:,:) = PLVSurrogate ; %nchan x nfreqs
  804. end
  806. % -- to extract individual level variables --
  807. freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  809. [~, WordFreq] = min(abs(freqvals-1.1111));
  810. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  812. GrandAverageTruePLV = squeeze(mean(TruePLVMtx(SubjectList,:,:),1));
  813. GrandAverageTruePLV_avacrosschans = mean(GrandAverageTruePLV(:,:),1);
  815. GrandAverageSurrogatePLV = squeeze(mean(SurrogatePLVMtx(SubjectList,:,:),1));
  816. GrandAverageSurrogatePLV_avacrosschans = mean(GrandAverageSurrogatePLV(:,:),1);
  818. %Standard Error
  819. TruePLV_averageacrosschans = squeeze(mean(TruePLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  820. STD_TruePLV = std(TruePLV_averageacrosschans(:,:)); %nfreqs
  821. SE_TruePLV = STD_TruePLV /sqrt(numel(SubjectList));
  823. SurrogatePLV_averageacrosschans = squeeze(mean(SurrogatePLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  824. STD_SurrogatePLV = std(SurrogatePLV_averageacrosschans(:,:)); %nfreqs
  825. SE_SurrogatePLV = STD_SurrogatePLV /sqrt(numel(SubjectList));
  828. NSubjects = numel(SubjectList);
  830. %---- Plots ------
  831. cd([CurrentDirectory 'FXS Project/infant pipeline/grand averages/'])
  833. %truncated
  834. freqvalindicesunder5Hz = find(freqvals <= 5);
  836. figure; plot(freqvals(freqvalindicesunder5Hz), GrandAverageTruePLV_avacrosschans(freqvalindicesunder5Hz),'r');
  837. hold on
  838. %shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), 'lineprops','r') %old syntax for shadedErrorBar
  839. shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAverageTruePLV_avacrosschans(freqvalindicesunder5Hz), SE_TruePLV(freqvalindicesunder5Hz), {'Color', 'r' }, 1);
  840. hold on
  841. plot(freqvals(freqvalindicesunder5Hz), GrandAverageSurrogatePLV_avacrosschans(freqvalindicesunder5Hz),'b');
  842. hold on
  843. shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAverageSurrogatePLV_avacrosschans(freqvalindicesunder5Hz), SE_SurrogatePLV(freqvalindicesunder5Hz), {'Color', 'b' }, 1);
  844. hold on
  845. xline(1.111, '--')
  846. hold on
  847. xline(2.222, '--')
  848. hold on
  849. xline(3.333, '--')
  850. legend('structured', 'SE', 'surrogate', 'SE')
  851. pdfname = ['ITC FFT True data versus Surrogate data, All Channels, n = ' num2str(NSubjects)];
  852. %pdfname = ['ITC FFT Auditory ROI'];
  853. print('-dpdf', pdfname); %save as postscript in order to merge two files
  859. % plot distribution
  861. WordFreqScaleLimits = [-0.25 0.25]; %[-0.15 .15]; %so that 0 is centered on green.
  862. SyllableFreqScaleLimits = [-0.5 .5];
  863. DifferenceEffectScaleLimits = [-0.05 0.05];
  865. figure;
  866. subplot(2,2,1)
  867. topoplot(GrandAverageTruePLV(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  868. title('True Data, Word Frequency');
  869. subplot(2,2,2)
  870. topoplot(GrandAverageTruePLV(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  871. title('True Data, Syllable Frequency');
  872. subplot(2,2,3)
  873. topoplot(GrandAverageSurrogatePLV(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  874. title('Surrogate Data, Word Frequency');
  875. subplot(2,2,4)
  876. topoplot(GrandAverageSurrogatePLV(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  877. title('Surrogate Data, Syllable Frequency');
  878. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  879. filename = ['Topographical Distributions, True versus Surrogate, n = ' num2str(NSubjects)];
  880. print ('-dtiff', filename);
  882. %difference topo
  883. WordFreqDifference = GrandAverageTruePLV(:,WordFreq) - GrandAverageSurrogatePLV(:,WordFreq);
  884. SyllableFreqDifference = GrandAverageTruePLV(:,SyllableFreq) - GrandAverageSurrogatePLV(:,SyllableFreq);
  887. % figure;
  888. % subplot(2,2,1)
  889. % %topoplot(WordFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', DifferenceEffectScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  890. % topoplot(WordFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', DifferenceEffectScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp', 'emarker2', {SignificantElectrodes_Word ,'p','k', 10, 1}); colorbar;
  891. % title('True minus Surrogate, Word Frequency');
  892. % subplot(2,2,2)
  893. % %topoplot(SyllableFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  894. % topoplot(SyllableFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp', 'emarker2', {SignificantElectrodes_Syllable,'p','k', 10, 1}); colorbar;
  895. % title('True minus Surrogate, Syllable Frequency');
  896. % set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  897. % filename = ['Topographical Distributions, True versus Surrogate, Difference, Sig Electrodes'];
  898. % print ('-dtiff', filename);
  901. figure;
  902. subplot(2,2,1)
  903. topoplot(WordFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', DifferenceEffectScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  904. %topoplot(WordFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', DifferenceEffectScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp', 'emarker2', {SignificantElectrodes_Word ,'p','k', 10, 1}); colorbar;
  905. title('True minus Surrogate, Word Frequency');
  906. subplot(2,2,2)
  907. topoplot(SyllableFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  908. %topoplot(SyllableFreqDifference, locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp', 'emarker2', {SignificantElectrodes_Syllable,'p','k', 10, 1}); colorbar;
  909. title('True minus Surrogate, Syllable Frequency');
  910. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  911. title('True minus Surrogate, Syllable Frequency');
  912. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  913. filename = ['Topographical Distributions, True versus Surrogate, Difference, n = ' num2str(NSubjects)];
  914. print ('-dtiff', filename);
  924. %% EXTRACT PLV FROM TRUE DATA AND SURROGATE DATA, AND PLOT AS Z SCORES
  927. clear SubjectIDs ITCVars
  929. clear FFTfile* Structured_PLVMtx
  931. clear Structured_PLVMtx
  933. plotchan= 1:104; %SigWordElectrodes; %1:105; % we removed the outer ring of electrodes
  935. locsfile = [CurrentDirectory 'FXS Project/channel location file/EGI104InnerChannelsNoCz.locs']; %exported this .locs file one time from EEG dataset. writelocs( chanstruct, filename ); writelocs( EEG.chanlocs, 'testlocs' );
  938. FFTfile_location = [CurrentDirectory 'FXS Project/infant pipeline/FFT files'];
  940. cd(FFTfile_location)
  942. FFTfiles = dir('*Structured_EEGfreq.mat'); %this only considers the files that were succesfully preprocessed and not those tagged with 'no usable data, all bad epochs.'
  944. FFTfile_names={FFTfiles.name};
  947. FFTsurrogatefile_location = [CurrentDirectory 'FXS Project/infant pipeline/shuffled FFT files'];
  950. SubjectList = [];
  951. SubjectIDList = []; %for bookkeeping for later analyses
  953. GroupName = ASD; %FXS; %TDC; %FXS; %TDC; AllSubjects; %
  955. %get subject list
  956. for x = 1:length(FFTfile_names)
  958. SubjectNumber = cell2mat(extractBetween(FFTfile_names{x}, 1, 4));
  960. if ismember(SubjectNumber, GroupName)
  961. SubjectList = [SubjectList, x];
  962. SubjectIDList = [SubjectIDList; str2double(SubjectNumber)];
  963. end
  966. end
  969. % -- STRUCTURED --
  970. clear PLVMtx Struct* Rand* Grand* PLV*
  972. TruePLVMtx = [];
  973. PLVSurrogate = [];
  974. SurrogatePLVMtx = [];
  975. zscoreTruePLVMtx = [];
  977. for i = SubjectList
  978. clear EEGfrq* SurrogateDataFileName SurrogateDatasetsMtx zscoreTruePLV
  979. display(i)
  981. %load "true" data
  982. cd(FFTfile_location)
  983. load(FFTfile_names{i})
  984. disp(FFTfile_names{i})
  986. PLVTrue = EEGfrq.plv(plotchan,:); %only extracts relevant channels
  987. TruePLVMtx(i,:,:) = PLVTrue; %rows without corresponding subjects (e.g., 20) just contains 0s. This is taken into account below by selecting SubjectsToInclude as a dimension.
  990. %load corresponding surrogate data
  991. cd(FFTsurrogatefile_location)
  992. SurrogateDataFileName = [char(extractBetween(FFTfile_names{i}, 1,4)) '_EEGfreq_ByIteration.mat'];
  993. load(SurrogateDataFileName)
  995. for iterationnum = 1:size(EEGfrq_ByIteration.iteration,2)
  996. SurrogateDatasetsMtx(iterationnum,:,:) = EEGfrq_ByIteration.iteration(iterationnum).plv(plotchan,:); %nchan x nfreqs for each surrogate dataset; ndataset x nchan x nfreq
  997. end
  999. % Z score the data according to the surrogate distribution
  1000. % each subject serves as their own Z score distribution
  1001. Mean_Surrogate = squeeze(mean(SurrogateDatasetsMtx,1)); %average across the 100 surrogate datasets for the subject
  1002. STD_Surrogate = squeeze(std(SurrogateDatasetsMtx)); %std deviation
  1004. %compute z score
  1005. zscoreTruePLV = (PLVTrue - Mean_Surrogate) ./ STD_Surrogate;
  1007. SurrogatePLVMtx(i,:,:) = Mean_Surrogate; %stores the mean surrogate ITCs for each subject; nregion x nfreqs
  1008. zscoreTruePLVMtx(i,:,:) = zscoreTruePLV; %stores the z scored actual observed data for each subject, nregion x nfreqs
  1010. end
  1012. % -- to extract individual level variables --
  1013. freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  1015. [~, WordFreq] = min(abs(freqvals-1.1111));
  1016. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  1018. GrandAverageTruePLV = squeeze(mean(TruePLVMtx(SubjectList,:,:),1));
  1019. GrandAverageTruePLV_avacrosschans = mean(GrandAverageTruePLV(:,:),1);
  1021. GrandAverageSurrogatePLV = squeeze(mean(SurrogatePLVMtx(SubjectList,:,:),1));
  1022. GrandAverageSurrogatePLV_avacrosschans = mean(GrandAverageSurrogatePLV(:,:),1);
  1024. %Standard Error
  1025. TruePLV_averageacrosschans = squeeze(mean(TruePLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  1026. STD_TruePLV = std(TruePLV_averageacrosschans(:,:)); %nfreqs
  1027. SE_TruePLV = STD_TruePLV /sqrt(numel(SubjectList));
  1029. SurrogatePLV_averageacrosschans = squeeze(mean(SurrogatePLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  1030. STD_SurrogatePLV = std(SurrogatePLV_averageacrosschans(:,:)); %nfreqs
  1031. SE_SurrogatePLV = STD_SurrogatePLV /sqrt(numel(SubjectList));
  1033. %Z scores
  1034. GrandAverageZscore = squeeze(mean(zscoreTruePLVMtx(SubjectList,:,:),1)); %nelectrodes x nfreqs
  1035. GrandAverageZscore_avacrosselectrodes = mean(GrandAverageZscore,1); %across all electrodes
  1036. %standard error
  1037. ZScoreTruePLV_averageacrosschans = squeeze(mean(zscoreTruePLVMtx(SubjectList,:,:),2)); %now nsubj x nfreq
  1038. STD_ZScoreTruePLV = std(ZScoreTruePLV_averageacrosschans(:,:)); %nfreqs
  1039. SE_ZScoreTruePLV = STD_ZScoreTruePLV /sqrt(numel(SubjectList));
  1045. NSubjects = numel(SubjectList);
  1047. %---- Plots ------
  1048. cd([CurrentDirectory 'FXS Project/infant pipeline/grand averages/z scored grand averages/'])
  1050. %truncated
  1051. freqvalindicesunder5Hz = find(freqvals <= 5);
  1053. %this is the same plot as created in the above cell
  1054. % figure; plot(freqvals(freqvalindicesunder5Hz), GrandAverageTruePLV_avacrosschans(freqvalindicesunder5Hz),'r');
  1055. % hold on
  1056. % %shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), 'lineprops','r') %old syntax for shadedErrorBar
  1057. % shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAverageTruePLV_avacrosschans(freqvalindicesunder5Hz), SE_TruePLV(freqvalindicesunder5Hz), {'Color', 'r' }, 1);
  1058. % hold on
  1059. % plot(freqvals(freqvalindicesunder5Hz), GrandAverageSurrogatePLV_avacrosschans(freqvalindicesunder5Hz),'b');
  1060. % hold on
  1061. % shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAverageSurrogatePLV_avacrosschans(freqvalindicesunder5Hz), SE_SurrogatePLV(freqvalindicesunder5Hz), {'Color', 'b' }, 1);
  1062. % hold on
  1063. % xline(1.111, '--')
  1064. % hold on
  1065. % xline(2.222, '--')
  1066. % hold on
  1067. % xline(3.333, '--')
  1068. % legend('structured', 'SE', 'surrogate', 'SE')
  1069. % pdfname = ['ITC True data versus Surrogate data, All Channels, n = ' num2str(NSubjects)];
  1070. % %pdfname = ['ITC FFT Auditory ROI'];
  1071. % print('-dpdf', pdfname); %save as postscript in order to merge two files
  1073. % z score plot as sanity check
  1075. %truncated
  1076. %freqvalindicesunder5Hz = find(freqvals <= 5);
  1078. figure; plot(freqvals(freqvalindicesunder5Hz), GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz),'r');
  1079. hold on
  1080. %shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), 'lineprops','r') %old syntax for shadedErrorBar
  1081. shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz), SE_ZScoreTruePLV(freqvalindicesunder5Hz), {'Color', 'r' }, 1);
  1082. hold on
  1083. xline(1.111, '--')
  1084. hold on
  1085. xline(2.222, '--')
  1086. hold on
  1087. xline(3.333, '--')
  1088. yline(0, 'k-', 'LineWidth', 1)
  1089. ylabel('ZITC')
  1090. xlabel('Frequency (Hz)')
  1091. legend('ZITC', 'SE')
  1092. pdfname = ['Z scored ITC, All Channels, n = ' num2str(NSubjects)];
  1093. print('-dpdf', pdfname); %save as postscript in order to merge two files
  1095. % plot topographical distribution of z scores
  1097. WordFreqScaleLimits = [-1 1]; %[-0.15 .15]; %so that 0 is centered on green.
  1098. SyllableFreqScaleLimits = [-8 8];
  1099. DifferenceEffectScaleLimits = [-0.05 0.05];
  1101. figure;
  1102. subplot(1,2,1)
  1103. topoplot(GrandAverageZscore(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1104. title('ZITC, Word Frequency');
  1105. subplot(1,2,2)
  1106. topoplot(GrandAverageZscore(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1107. title('ZITC, Syllable Frequency');
  1108. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'landscape', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  1109. filename = ['ZITC Topographical Distributions, n = ' num2str(NSubjects)];
  1110. print ('-dtiff', filename);
  1111. print ('-dpdf', filename);
  1113. %plot significaint electrodes that show significant entrainment at each frequency,
  1114. %use a one-sample t-test against 0
  1116. WordZITCByElectrodesandBySubject = zscoreTruePLVMtx(SubjectList, :, WordFreq);
  1117. SyllableZITCByElectrodesandBySubject = zscoreTruePLVMtx(SubjectList, :, SyllableFreq);
  1119. [SigWordElectrodesArray, PWord] = ttest(WordZITCByElectrodesandBySubject, 0, 'tail', 'right'); %one directional t-test, mean is greater than 0
  1120. [SigSyllableElectrodesArray, PSyllable] = ttest(SyllableZITCByElectrodesandBySubject, 0, 'tail', 'right'); %one directional t-test, mean is greater than 0
  1122. SigWordElectrodes = find(SigWordElectrodesArray ==1);
  1123. SigSyllableElectrodes = find(SigSyllableElectrodesArray ==1);
  1125. NumSigWordElectrodes = numel(SigWordElectrodes);
  1126. NumSigSyllableElectrodes = numel(SigSyllableElectrodes);
  1128. figure;
  1129. subplot(1,2,1)
  1130. topoplot(GrandAverageZscore(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits,'shading','interp', 'emarker2', {SigWordElectrodes ,'p','k', 7, 1} ); colorbar;
  1131. title(['Significant Word ITC Electrodes: ' num2str(NumSigWordElectrodes)]);
  1132. subplot(1,2,2)
  1133. topoplot(GrandAverageZscore(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits,'shading','interp', 'emarker2', {SigSyllableElectrodes ,'p','k', 7, 1} ); colorbar;
  1134. title(['Significant Syllable ITC Electrodes: ' num2str(NumSigSyllableElectrodes)]);
  1135. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  1136. filename = ['ZITC Topo, Sig Electrodes, n = ' num2str(NSubjects)];
  1137. print ('-dtiff', filename);
  1141. % extract individual level variables for statistical analysis
  1143. ZScoreITCWordFreqMtx = zscoreTruePLVMtx(SubjectList,:, WordFreq); %nsubj X nchan
  1144. ZScoreITCSyllableFreqMtx = zscoreTruePLVMtx(SubjectList,:, SyllableFreq);
  1146. %ZScoreWLI = (ZScoreITCWordFreqMtx ./ ZScoreITCSyllableFreqMtx); %no longer sure whether the WLI is that informative considering we are now already normalizing by the surrogate shuffled data for each participant and electrode
  1148. %variables based on channel configurations
  1149. allchans= 1:104;
  1151. ZScoreITCWordFreqMtx_allchans = mean(ZScoreITCWordFreqMtx(:,allchans),2);
  1152. ZScoreITCSyllableFreqMtx_allchans = mean(ZScoreITCSyllableFreqMtx(:,allchans),2);
  1153. %ZScoreWLI_allchans = mean(ZScoreWLI(:,allchans),2);
  1156. SPSSToCompare = [ZScoreITCWordFreqMtx_allchans, ZScoreITCSyllableFreqMtx_allchans];
  1158. CorrespondingSubjectIDs = SubjectIDList;
  1160. %can build the array one group at a time
  1163. %% GROUP COMPARISONS - COMPARING TWO GROUPS AT TIME - ZITC GRAND AVERAGE
  1165. Group1Name = TDC;
  1166. Group2Name = FXS;
  1169. % -- GROUP 1 ---
  1170. Group1SubjectList = [];
  1172. %get subject list
  1173. for x = 1:length(FFTfile_names)
  1175. SubjectNumber = cell2mat(extractBetween(FFTfile_names{x}, 1, 4));
  1177. if ismember(SubjectNumber, Group1Name)
  1178. Group1SubjectList = [Group1SubjectList, x];
  1179. end
  1181. end
  1184. % -- GROUP 2 ---
  1186. Group2SubjectList = [];
  1188. %get subject list
  1189. for x = 1:length(FFTfile_names)
  1191. SubjectNumber = cell2mat(extractBetween(FFTfile_names{x}, 1, 4));
  1193. if ismember(SubjectNumber, Group2Name)
  1194. Group2SubjectList = [Group2SubjectList, x];
  1195. end
  1197. end
  1200. % -- GROUP 1 --
  1201. clear PLVMtx Struct* Rand* Grand* PLV*
  1203. Group1TruePLVMtx = [];
  1204. Group1SurrogatePLVMtx = [];
  1205. Group1zscoreTruePLVMtx = [];
  1207. for i = Group1SubjectList
  1208. clear EEGfrq* SurrogateDataFileName SurrogateDatasetsMtx zscoreTruePLV
  1209. display(i)
  1211. %load "true" data
  1212. cd(FFTfile_location)
  1213. load(FFTfile_names{i})
  1214. disp(FFTfile_names{i})
  1216. PLVTrue = EEGfrq.plv(plotchan,:); %only extracts relevant channels
  1217. Group1TruePLVMtx(i,:,:) = PLVTrue; %rows without corresponding subjects (e.g., 20) just contains 0s. This is taken into account below by selecting SubjectsToInclude as a dimension.
  1220. %load corresponding surrogate data
  1221. cd(FFTsurrogatefile_location)
  1222. SurrogateDataFileName = [char(extractBetween(FFTfile_names{i}, 1,4)) '_EEGfreq_ByIteration.mat'];
  1223. load(SurrogateDataFileName)
  1225. for iterationnum = 1:size(EEGfrq_ByIteration.iteration,2)
  1226. SurrogateDatasetsMtx(iterationnum,:,:) = EEGfrq_ByIteration.iteration(iterationnum).plv(plotchan,:); %nchan x nfreqs for each surrogate dataset; ndataset x nchan x nfreq
  1227. end
  1229. % Z score the data according to the surrogate distribution
  1230. % each subject serves as their own Z score distribution
  1231. Mean_Surrogate = squeeze(mean(SurrogateDatasetsMtx,1)); %average across the 100 surrogate datasets for the subject
  1232. STD_Surrogate = squeeze(std(SurrogateDatasetsMtx)); %std deviation
  1234. %compute z score
  1235. zscoreTruePLV = (PLVTrue - Mean_Surrogate) ./ STD_Surrogate;
  1237. Group1SurrogatePLVMtx(i,:,:) = Mean_Surrogate; %stores the mean surrogate ITCs for each subject; nregion x nfreqs
  1238. Group1zscoreTruePLVMtx(i,:,:) = zscoreTruePLV; %stores the z scored actual observed data for each subject, nregion x nfreqs
  1240. end
  1245. % -- GROUP 2 --
  1246. clear PLVMtx Struct* Rand* Grand* PLV*
  1248. Group2TruePLVMtx = [];
  1249. Group2SurrogatePLVMtx = [];
  1250. Group2zscoreTruePLVMtx = [];
  1252. for i = Group2SubjectList
  1253. clear EEGfrq* SurrogateDataFileName SurrogateDatasetsMtx zscoreTruePLV
  1254. display(i)
  1256. %load "true" data
  1257. cd(FFTfile_location)
  1258. load(FFTfile_names{i})
  1259. disp(FFTfile_names{i})
  1261. PLVTrue = EEGfrq.plv(plotchan,:); %only extracts relevant channels
  1262. Group2TruePLVMtx(i,:,:) = PLVTrue; %rows without corresponding subjects (e.g., 20) just contains 0s. This is taken into account below by selecting SubjectsToInclude as a dimension.
  1265. %load corresponding surrogate data
  1266. cd(FFTsurrogatefile_location)
  1267. SurrogateDataFileName = [char(extractBetween(FFTfile_names{i}, 1,4)) '_EEGfreq_ByIteration.mat'];
  1268. load(SurrogateDataFileName)
  1270. for iterationnum = 1:size(EEGfrq_ByIteration.iteration,2)
  1271. SurrogateDatasetsMtx(iterationnum,:,:) = EEGfrq_ByIteration.iteration(iterationnum).plv(plotchan,:); %nchan x nfreqs for each surrogate dataset; ndataset x nchan x nfreq
  1272. end
  1274. % Z score the data according to the surrogate distribution
  1275. % each subject serves as their own Z score distribution
  1276. Mean_Surrogate = squeeze(mean(SurrogateDatasetsMtx,1)); %average across the 100 surrogate datasets for the subject
  1277. STD_Surrogate = squeeze(std(SurrogateDatasetsMtx)); %std deviation
  1279. %compute z score
  1280. zscoreTruePLV = (PLVTrue - Mean_Surrogate) ./ STD_Surrogate;
  1282. Group2SurrogatePLVMtx(i,:,:) = Mean_Surrogate; %stores the mean surrogate ITCs for each subject; nregion x nfreqs
  1283. Group2zscoreTruePLVMtx(i,:,:) = zscoreTruePLV; %stores the z scored actual observed data for each subject, nregion x nfreqs
  1285. end
  1291. % -- to extract individual level variables --
  1292. freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  1294. [~, WordFreq] = min(abs(freqvals-1.1111));
  1295. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  1298. % - Group 1 -
  1299. Group1GrandAverageTruePLV = squeeze(mean(Group1TruePLVMtx(Group1SubjectList,:,:),1));
  1300. Group1GrandAverageTruePLV_avacrosschans = mean(Group1GrandAverageTruePLV(:,:),1);
  1302. Group1GrandAverageSurrogatePLV = squeeze(mean(Group1SurrogatePLVMtx(Group1SubjectList,:,:),1));
  1303. Group1GrandAverageSurrogatePLV_avacrosschans = mean(Group1GrandAverageSurrogatePLV(:,:),1);
  1305. %Standard Error
  1306. Group1TruePLV_averageacrosschans = squeeze(mean(Group1TruePLVMtx(Group1SubjectList,:,:),2)); %now nsubj x nfreq
  1307. Group1STD_TruePLV = std(Group1TruePLV_averageacrosschans(:,:)); %nfreqs
  1308. Group1SE_TruePLV = Group1STD_TruePLV /sqrt(numel(Group1SubjectList));
  1310. Group1SurrogatePLV_averageacrosschans = squeeze(mean(Group1SurrogatePLVMtx(Group1SubjectList,:,:),2)); %now nsubj x nfreq
  1311. Group1STD_SurrogatePLV = std(Group1SurrogatePLV_averageacrosschans(:,:)); %nfreqs
  1312. Group1SE_SurrogatePLV = Group1STD_SurrogatePLV /sqrt(numel(Group1SubjectList));
  1314. %Z scores
  1315. Group1GrandAverageZscore = squeeze(mean(Group1zscoreTruePLVMtx(Group1SubjectList,:,:),1));
  1316. Group1GrandAverageZscore_avacrosselectrodes = mean(Group1GrandAverageZscore,1);
  1317. %standard error
  1318. Group1ZScoreTruePLV_averageacrosschans = squeeze(mean(Group1zscoreTruePLVMtx(Group1SubjectList,:,:),2)); %now nsubj x nfreq
  1319. Group1STD_ZScoreTruePLV = std(Group1ZScoreTruePLV_averageacrosschans(:,:)); %nfreqs
  1320. Group1SE_ZScoreTruePLV = Group1STD_ZScoreTruePLV /sqrt(numel(Group1SubjectList));
  1323. % - Group 2 -
  1325. Group2GrandAverageTruePLV = squeeze(mean(Group2TruePLVMtx(Group2SubjectList,:,:),1));
  1326. Group2GrandAverageTruePLV_avacrosschans = mean(Group2GrandAverageTruePLV(:,:),1);
  1328. Group2GrandAverageSurrogatePLV = squeeze(mean(Group2SurrogatePLVMtx(Group2SubjectList,:,:),1));
  1329. Group2GrandAverageSurrogatePLV_avacrosschans = mean(Group2GrandAverageSurrogatePLV(:,:),1);
  1331. %Standard Error
  1332. Group2TruePLV_averageacrosschans = squeeze(mean(Group2TruePLVMtx(Group2SubjectList,:,:),2)); %now nsubj x nfreq
  1333. Group2STD_TruePLV = std(Group2TruePLV_averageacrosschans(:,:)); %nfreqs
  1334. Group2SE_TruePLV = Group2STD_TruePLV /sqrt(numel(Group2SubjectList));
  1336. Group2SurrogatePLV_averageacrosschans = squeeze(mean(Group2SurrogatePLVMtx(Group2SubjectList,:,:),2)); %now nsubj x nfreq
  1337. Group2STD_SurrogatePLV = std(Group2SurrogatePLV_averageacrosschans(:,:)); %nfreqs
  1338. Group2SE_SurrogatePLV = Group2STD_SurrogatePLV /sqrt(numel(Group2SubjectList));
  1340. %Z scores
  1341. Group2GrandAverageZscore = squeeze(mean(Group2zscoreTruePLVMtx(Group2SubjectList,:,:),1));
  1342. Group2GrandAverageZscore_avacrosselectrodes = mean(Group2GrandAverageZscore,1);
  1343. %standard error
  1344. Group2ZScoreTruePLV_averageacrosschans = squeeze(mean(Group2zscoreTruePLVMtx(Group2SubjectList,:,:),2)); %now nsubj x nfreq
  1345. Group2STD_ZScoreTruePLV = std(Group2ZScoreTruePLV_averageacrosschans(:,:)); %nfreqs
  1346. Group2SE_ZScoreTruePLV = Group2STD_ZScoreTruePLV /sqrt(numel(Group2SubjectList));
  1349. NSubjectsGroup1 = numel(Group1SubjectList);
  1350. NSubjectsGroup2 = numel(Group2SubjectList);
  1352. %---- Plots ------
  1353. cd([CurrentDirectory 'FXS Project/infant pipeline/grand averages/z scored grand averages/'])
  1355. %truncated
  1356. %freqvalindicesunder5Hz = find(freqvals <= 5);
  1357. freqvalindicesunder5Hz = find(freqvals <= 4);
  1359. %z score plot
  1360. figure; h1 = plot(freqvals(freqvalindicesunder5Hz), Group1GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz),'k', 'LineWidth', 2);
  1361. hold on
  1362. %shadedErrorBar(freqvals(freqvalindicesunder5Hz),GrandAveragePLV_avacrosschans_Structured(freqvalindicesunder5Hz), SE_Structured(freqvalindicesunder5Hz), 'lineprops','r') %old syntax for shadedErrorBar
  1363. h = shadedErrorBar(freqvals(freqvalindicesunder5Hz),Group1GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz), Group1SE_ZScoreTruePLV(freqvalindicesunder5Hz), {'Color', 'k' }, 1);
  1364. hold on
  1365. h2 = plot(freqvals(freqvalindicesunder5Hz), Group2GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz),'r', 'LineWidth', 2);
  1366. hold on
  1367. h = shadedErrorBar(freqvals(freqvalindicesunder5Hz),Group2GrandAverageZscore_avacrosselectrodes(freqvalindicesunder5Hz), Group2SE_ZScoreTruePLV(freqvalindicesunder5Hz), {'Color', 'r' }, 1);
  1368. hold on
  1369. xline(1.111, '--')
  1370. hold on
  1371. xline(2.222, '--')
  1372. hold on
  1373. xline(3.333, '--')
  1374. yline(0, 'k-', 'LineWidth', 1)
  1375. ylabel('ZITC')
  1376. xlabel('Frequency (Hz)')
  1377. %legend([h1, h2], {'TDC', 'FXS'})
  1378. set(gca, 'FontSize', 14);
  1379. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'portrait', 'PaperPosition', [0.5 0.5 8.0 10.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  1380. figurename = ['Z scored ITC, Two Group Comparison, All Channels, No Legend, nGroup1 = ' num2str(NSubjectsGroup1), ' nGroup2 = ' num2str(NSubjectsGroup2) ];
  1381. exportgraphics(gcf, [figurename '.pdf'], 'ContentType','vector'); %this should import as a vector now
  1382. %print('-dpdf', figurename); %save as postscript in order to merge two files
  1383. %print('-dtiff', figurename); %save as postscript in order to merge two files
  1386. % plot topographical distribution of z scores
  1388. WordFreqScaleLimits = [-1 1]; %[-0.15 .15]; %so that 0 is centered on green.
  1389. SyllableFreqScaleLimits = [-8 8];
  1390. DifferenceEffectScaleLimits = [-0.05 0.05];
  1392. figure;
  1393. subplot(2,2,1)
  1394. topoplot(Group1GrandAverageZscore(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1395. %title('TDC Word Frequency');
  1396. subplot(2,2,2)
  1397. topoplot(Group1GrandAverageZscore(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1398. %title('TDC Syllable Frequency');
  1399. subplot(2,2,3)
  1400. topoplot(Group2GrandAverageZscore(:,WordFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', WordFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1401. %title('FXS Word Frequency');
  1402. subplot(2,2,4)
  1403. topoplot(Group2GrandAverageZscore(:,SyllableFreq), locsfile, 'style', 'map', 'colormap', jet, 'maplimits', SyllableFreqScaleLimits, 'plotrad',.5,'headrad',.5,'shading','interp'); colorbar;
  1404. %title('FXS Syllable Frequency');
  1405. set(gcf,'PaperPositionMode','manual', 'PaperOrientation', 'landscape', 'PaperPosition', [0.05 0.15 12.0 8.50]) %corresponds to settings on Print Preview. %good measurements for not cutting off title.
  1406. filename = ['ZITC Group Comparison Topographical Distributions, no titles, nGroup1 = ' num2str(NSubjectsGroup1), ' nGroup2 = ' num2str(NSubjectsGroup2)];
  1407. exportgraphics(gcf, filename + ".pdf", 'ContentType','vector');
  1408. %print ('-dtiff', filename);
  1409. %print ('-dpdf', filename);
  1413. %% JACKKNIFE PROCEDURE TO ESTIMATE ITPC RELIABILITY
  1415. %load the same FFT parameters as above. (LOAD PARAMETERS cell)
  1417. ParticipantLevelData = [];
  1418. PseudovalsByParticipantByEpoch = []; %timecourse data
  1420. subjectcounter = 0;
  1422. for subji = AllSubjectIndices %length(datafile_names)
  1424. SubjectName = datafile_names{subji}(1:4); %string
  1425. SubjectNumber = str2double(SubjectName); %number
  1426. disp(subji)
  1427. disp(SubjectName)
  1428. subjectcounter = subjectcounter + 1;
  1430. %figure out GroupID
  1431. clear GroupID
  1432. if any(strcmp(SubjectName, TDC))
  1433. GroupID = 1;
  1434. elseif any(strcmp(SubjectName, FXS))
  1435. GroupID = 2;
  1436. elseif any(strcmp(SubjectName, ASD))
  1437. GroupID = 3;
  1438. else
  1439. GroupID = 0; %unclassified, but at least it won't disrupt the loop
  1440. end
  1443. %load dataset
  1444. clear EEGfrq
  1446. STUDY = []; CURRENTSTUDY = 0; ALLEEG = []; EEG=[]; CURRENTSET=[];
  1447. EEG = pop_loadset( 'filename', [SubjectName '_epoched.set'], 'filepath', [CurrentDirectory 'FXS project/infant pipeline/epochs/']);
  1448. [ALLEEG, EEG, CURRENTSET] = eeg_store( ALLEEG, EEG );
  1450. ParentEEG = EEG;
  1451. NumTrials = EEG.trials;
  1452. FullTrialArray = 1:NumTrials;
  1453. NumberofChannels = size(EEG.chanlocs,2);
  1455. % ------
  1457. %create temporarily subdivided datasets and run the function on each dataset
  1458. %first create the big frq structure
  1459. EEGfrq = frqa_plvpwr(EEG, 'foutput',foutput,'fband',fband,'binsSNR',binsSNR,...
  1460. 'typeSNRPWR',typeSNRPWR,'typeSNRPLV',typeSNRPLV,...
  1461. 'zscorePWR',zscorePWR,'zscorePLV',zscorePLV);
  1463. freqvals = EEGfrq.freqs; %freqs based on last subject loaded
  1465. [~, WordFreq] = min(abs(freqvals-1.1111));
  1466. [~, SyllableFreq] = min(abs(freqvals-3.3333)); %syllable freq = 3.33 Hz
  1468. %for normalization, extract neighbouring bins excluding the word,
  1469. %harmonics, and syllable frequencies.
  1470. [~, Harmonic1Freq] = min(abs(freqvals-2.22222)); %syllable freq = 3.33 Hz
  1471. [~, Harmonic2Freq] = min(abs(freqvals-4.44444)); %syllable freq = 3.33 Hz
  1473. NormalizationFreqBins = setdiff(find(freqvals <= 5), [WordFreq, SyllableFreq, Harmonic1Freq, Harmonic2Freq]);
  1475. %compute ITC
  1476. AllTrialsWordITC = EEGfrq.plv(:,WordFreq);
  1477. AllTrialsSyllableITC = EEGfrq.plv(:,SyllableFreq);
  1479. AllTrialsNormalizedBinsITC = squeeze(mean(EEGfrq.plv(:, NormalizationFreqBins ),2));
  1481. AllTrialsWordITCNormalized = AllTrialsWordITC - AllTrialsNormalizedBinsITC;
  1482. AllTrialsSyllableITCNormalized = AllTrialsSyllableITC - AllTrialsNormalizedBinsITC;
  1484. EEGfrq_jackknifed.replication(1:NumTrials) = EEGfrq; %preallocate - delete extra rows later - in order to have a bigger structure with the same structure
  1486. WordITCJackknifed_ByReplication= [];
  1487. SyllableITCJackknifed_ByReplication = [];
  1488. WordITCNormalizedJackknifed_ByReplication = [];
  1489. SyllableITCNormalizedJackknifed_ByReplication = [];
  1491. for replication = 1:NumTrials
  1493. disp(['Replication Number :' num2str(replication)])
  1495. %compute what ITC would be without that trial
  1496. TrialArrayLessOneTrial = setdiff(FullTrialArray, replication); %removes N trial
  1498. EEGLessOneTrial = pop_select( ParentEEG, 'trial', TrialArrayLessOneTrial);
  1499. %run command on subsampled trials
  1500. EEGfrqLessOneTrial = frqa_plvpwr(EEGLessOneTrial, 'foutput',foutput,'fband',fband,'binsSNR',binsSNR,...
  1501. 'typeSNRPWR',typeSNRPWR,'typeSNRPLV',typeSNRPLV,...
  1502. 'zscorePWR',zscorePWR,'zscorePLV',zscorePLV);
  1504. EEGfrq_jackknifed.replication(replication) = EEGfrqLessOneTrial;
  1506. %compute individual actual and normalized ITC values to store for analysis
  1507. WordITCJackknifed = EEGfrqLessOneTrial.plv(:,WordFreq);
  1508. SyllableITCJackknifed = EEGfrqLessOneTrial.plv(:,SyllableFreq);
  1510. OtherFreqsITCJackknifed = squeeze(mean(EEGfrqLessOneTrial.plv(:, NormalizationFreqBins ),2));
  1512. WordITCNormalizedJackknifed = WordITCJackknifed - OtherFreqsITCJackknifed;
  1513. SyllableITCNormalizedJackknifed = SyllableITCJackknifed - OtherFreqsITCJackknifed;
  1515. %store
  1516. WordITCJackknifed_ByReplication(replication,:) = WordITCJackknifed; %Nreplications x NElectrodes
  1517. SyllableITCJackknifed_ByReplication(replication,:) = SyllableITCJackknifed; %Nreplications x NElectrodes
  1518. WordITCNormalizedJackknifed_ByReplication(replication,:) = WordITCNormalizedJackknifed; %Nreplications x NElectrodes
  1519. SyllableITCNormalizedJackknifed_ByReplication(replication,:) = SyllableITCNormalizedJackknifed; %Nreplications x NElectrodes
  1521. clear EEGfrqLessOneTrial
  1523. end
  1526. %save the participant's EEG freq matrix
  1527. cd([CurrentDirectory 'FXS Project/infant pipeline/jackknifed FFT files/'])
  1528. disp(SubjectNumber)
  1530. %filename = strcat(num2str(SubjectNumber), '_EEGfrq_jackknifed');
  1531. filename = [SubjectName '_EEGfrq_jackknifed'];
  1532. save(filename, 'EEGfrq_jackknifed');
  1534. %average steps to save in participant matrix
  1535. %average across channels
  1536. WordITCJackknifed_ByReplication_meanacrosschans = mean(WordITCJackknifed_ByReplication,2);
  1537. SyllableITCJackknifed_ByReplication_meanacrosschans = mean(SyllableITCJackknifed_ByReplication,2);
  1538. WordITCNormalizedJackknifed_ByReplication_meanacrosschans = mean(WordITCNormalizedJackknifed_ByReplication,2);
  1539. SyllableITCNormalizedJackknifed_ByReplication_meanacrosschans = mean(SyllableITCNormalizedJackknifed_ByReplication,2);
  1541. %average across replications
  1542. MeanWordITCJackknifed_ByReplication_meanacrosschans = mean(WordITCJackknifed_ByReplication_meanacrosschans);
  1543. MeanSyllableITCJackknifed_ByReplication_meanacrosschans = mean(SyllableITCJackknifed_ByReplication_meanacrosschans);
  1544. MeanWordITCNormJackknifed_ByReplication_meanacrosschans = mean( WordITCNormalizedJackknifed_ByReplication_meanacrosschans);
  1545. MeanSyllableITCNormJackknifed_ByReplication_meanacrosschans = mean(SyllableITCNormalizedJackknifed_ByReplication_meanacrosschans);
  1547. %original ITC values as computed across all trials (from above), average across channels
  1548. AllTrialsWordITC_meanacrosschans = mean(AllTrialsWordITC,1);
  1549. AllTrialsSyllableITC_meanacrosschans = mean(AllTrialsSyllableITC,1);
  1550. AllTrialsWordITCNormalized_meanacrosschans = mean(AllTrialsWordITCNormalized,1);
  1551. AllTrialsSyllableITCNormalized_meanacrosschans = mean(AllTrialsSyllableITCNormalized,1);
  1553. %jackknife pseudovals - a pseudo-value is calculated as the difference
  1554. %between the whole sample estimate and the partial estimate. This tells
  1555. %you whether each individual trial contributes to lowering or increasing the ITC and by how much. Perfectly inversely correlated with jackknifed
  1556. %ITC values (rest of the sample ITC). So a higher pseudoval would
  1557. %indicate that the contribution of that specific trial INCREASES the
  1558. %overall ITC (such that when that trial is removed, the remaining trial
  1559. %ITC is lower). Formula from https://www.mattcraddock.com/blog/2020/09/07/jackknife-intertrial-coherence-and-phase/
  1560. WordITCPseudovals_ByReplication_meanacrosschans = NumTrials*AllTrialsWordITC_meanacrosschans - ((NumTrials-1) *WordITCJackknifed_ByReplication_meanacrosschans); %Ntrial
  1561. SyllableITCPseudovals_ByReplication_meanacrosschans = NumTrials*AllTrialsSyllableITC_meanacrosschans - ((NumTrials-1) *SyllableITCJackknifed_ByReplication_meanacrosschans);
  1562. WordITCNormalizedPseudovals_ByReplication_meanacrosschans = NumTrials*AllTrialsWordITCNormalized_meanacrosschans - ((NumTrials-1) *WordITCNormalizedJackknifed_ByReplication_meanacrosschans); %Ntrial
  1563. SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans = NumTrials*AllTrialsSyllableITCNormalized_meanacrosschans - ((NumTrials-1) *SyllableITCNormalizedJackknifed_ByReplication_meanacrosschans);
  1565. %exact values to store in participant level matrices
  1566. %mean of pseudovalues can be viewed as a bias reduced version of the original statistic
  1567. MeanWordITCPseudovals_ByReplication_meanacrosschans = mean(WordITCPseudovals_ByReplication_meanacrosschans);
  1568. MeanSyllableITCPseudovals_ByReplication_meanacrosschans = mean(SyllableITCPseudovals_ByReplication_meanacrosschans);
  1569. MeanWordITCNormalPseudovals_ByReplication_meanacrosschans = mean(WordITCNormalizedPseudovals_ByReplication_meanacrosschans);
  1570. MeanSyllableITCNormalPseudovals_ByReplication_meanacrosschans = mean(SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans);
  1572. %standard deviation of pseudovalues
  1573. StDevWordITCPseudovals_ByReplication_meanacrosschans = std(WordITCPseudovals_ByReplication_meanacrosschans);
  1574. StDevSyllableITPseudovals_ByReplication_meanacrosschans = std(SyllableITCPseudovals_ByReplication_meanacrosschans);
  1575. StDevWordITCNormalPseudovals_ByReplication_meanacrosschans = std(WordITCNormalizedPseudovals_ByReplication_meanacrosschans);
  1576. StDevSyllableITCNormalPseudovals_ByReplication_meanacrosschans = std(SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans);
  1578. %take the highest 5 and 10 pseudovalues:
  1580. %5
  1581. Max5WordITCPseudovals_ByReplication_meanacrosschans = mean(maxk(WordITCPseudovals_ByReplication_meanacrosschans, 5));
  1582. Max5SyllableITPseudovals_ByReplication_meanacrosschans = mean(maxk(SyllableITCPseudovals_ByReplication_meanacrosschans, 5));
  1583. Max5WordITCNormalPseudovals_ByReplication_meanacrosschans = mean(maxk(WordITCNormalizedPseudovals_ByReplication_meanacrosschans, 5));
  1584. Max5SyllableITCNormalPseudovals_ByReplication_meanacrosschans = mean(maxk(SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans, 5));
  1586. %10th
  1587. Max10WordITCPseudovals_ByReplication_meanacrosschans = mean(maxk(WordITCPseudovals_ByReplication_meanacrosschans, 10));
  1588. Max10SyllableITPseudovals_ByReplication_meanacrosschans = mean(maxk(SyllableITCPseudovals_ByReplication_meanacrosschans, 10));
  1589. Max10WordITCNormalPseudovals_ByReplication_meanacrosschans = mean(maxk(WordITCNormalizedPseudovals_ByReplication_meanacrosschans, 10));
  1590. Max10SyllableITCNormalPseudovals_ByReplication_meanacrosschans = mean(maxk(SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans, 10));
  1592. %Trial indices associated with >0 Positive Pseudovalues
  1593. %positive pseudovalues means that the trial was contributing positively to the overall
  1594. %ITC estimate; in other words, when that trial is removed, ITC across
  1595. %the remaining trials DECREASES.
  1596. %A negative pseudovalue means that the trial was disrupting or reducing
  1597. %phase coherence potentially due to noise, phase jitter, or phase-locking at a different angle. In other words, removing that trial INCREASES ITPC
  1598. %this could give insight to the temporal trajecotry of participants' learning.
  1599. PositiveContributingTrialsToWordITC= find(WordITCPseudovals_ByReplication_meanacrosschans >0);
  1600. PositiveContributingTrialsToSyllableITC = find(SyllableITCPseudovals_ByReplication_meanacrosschans >0);
  1601. PositiveContributingTrialsToWordITCNormalized= find(WordITCNormalizedPseudovals_ByReplication_meanacrosschans >0);
  1602. PositiveContributingTrialsToSyllableITCNormalized= find(SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans >0);
  1604. %summarize the total number of positive contributing trials - perhaps
  1605. %this is a proxy of participants' overall time spent parsing or
  1606. %processing the word structures
  1607. NumberPositiveContributingTrialsToWordITC = numel(PositiveContributingTrialsToWordITC);
  1608. NumberPositiveContributingTrialsToSyllableITC = numel(PositiveContributingTrialsToSyllableITC);
  1609. NumberPositiveContributingTrialsToWordITCNormalized = numel(PositiveContributingTrialsToWordITCNormalized);
  1610. NumberPositiveContributingTrialsToSyllableITCNormalized = numel(PositiveContributingTrialsToSyllableITCNormalized);
  1612. ParticipantLevelData(subjectcounter,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:,:) = [SubjectNumber, GroupID, AllTrialsWordITC_meanacrosschans, AllTrialsSyllableITC_meanacrosschans, ...
  1613. AllTrialsWordITCNormalized_meanacrosschans, AllTrialsSyllableITCNormalized_meanacrosschans, MeanWordITCPseudovals_ByReplication_meanacrosschans, MeanSyllableITCPseudovals_ByReplication_meanacrosschans , ...
  1614. MeanWordITCNormalPseudovals_ByReplication_meanacrosschans, MeanSyllableITCNormalPseudovals_ByReplication_meanacrosschans, ...
  1615. Max5WordITCPseudovals_ByReplication_meanacrosschans, Max5SyllableITPseudovals_ByReplication_meanacrosschans, ...
  1616. Max5WordITCNormalPseudovals_ByReplication_meanacrosschans, Max5SyllableITCNormalPseudovals_ByReplication_meanacrosschans, Max10WordITCPseudovals_ByReplication_meanacrosschans, Max10SyllableITPseudovals_ByReplication_meanacrosschans, ...
  1617. Max10WordITCNormalPseudovals_ByReplication_meanacrosschans, Max10SyllableITCNormalPseudovals_ByReplication_meanacrosschans, ...
  1618. NumberPositiveContributingTrialsToWordITC, NumberPositiveContributingTrialsToSyllableITC, NumberPositiveContributingTrialsToWordITCNormalized, ...
  1619. NumberPositiveContributingTrialsToSyllableITCNormalized];
  1621. %now we will save the pseudovals to enable time course analysis:
  1622. TrialArrayToUse = [1:NumTrials]';
  1623. PseudovalArrays = [WordITCPseudovals_ByReplication_meanacrosschans, SyllableITCPseudovals_ByReplication_meanacrosschans, ...
  1624. WordITCNormalizedPseudovals_ByReplication_meanacrosschans, SyllableITCNormalizedPseudovals_ByReplication_meanacrosschans];
  1626. SubjectIDRepArray = repmat(SubjectNumber, size(TrialArrayToUse));
  1628. GroupIDRepArray = repmat(GroupID, size(TrialArrayToUse));
  1630. CurrentSubjectData = [SubjectIDRepArray, GroupIDRepArray, TrialArrayToUse, PseudovalArrays];
  1632. PseudovalsByParticipantByEpoch = vertcat( PseudovalsByParticipantByEpoch , CurrentSubjectData);
  1635. end
  1638. cd([CurrentDirectory 'FXS Project/infant pipeline/jackknifed FFT files/summary tables']) %so that files go in proper place
  1640. %after all participants have run, create and save as a table
  1642. %save as table
  1644. ColumnNames = {'Subject', 'Group', 'AllTrialsWordITC', 'AllTrialsSyllableITC', 'AllTrialsWordITCNormalized', 'AllTrialsSyllableITCNormalized', ...
  1645. 'MeanWordITCPseudovals', 'MeanSyllableITCPseudovals', 'MeanWordITCNormalPseudovals_ByReplication_meanacrosschans', 'MeanSyllableITCNormalPseudovals_ByReplication_meanacrosschans', ...
  1646. 'Max5WordITCPseudovals', 'Max5SyllableITPseudovals', 'Max5WordITCNormPseudovals', 'Max5SyllableITCNormPseudovals', 'Max10WordITCPseudovals', 'Max10SyllableITPseudovals', ...
  1647. 'Max10WordITCNormalPseudovals', 'Max10SyllableITCNormalPseudovals', 'NumberPosContributingTrialsToWordITC', 'NumberPosContributingTrialsToSyllableITC', ...
  1648. 'NumberPosContributingTrialsToWordITCNorm' 'NumberPosContributingTrialsToSyllableITCNorm'};
  1650. ParticipantDataTableForAnalysis = array2table(ParticipantLevelData(1:end,:), 'VariableNames', ColumnNames);
  1652. writetable(ParticipantDataTableForAnalysis,'ParticipantDataTableForAnalysis.csv')
  1655. %Table 2 - timecourse data
  1657. ColumnNames2 = {'Subject', 'Group', 'EpochNum', 'WordITCPseudoval', 'SyllableITCPseudoval', 'WordITCNormPseudoval', 'SyllableITCNormPseudoval'};
  1659. PseudovalTimecourseTableForAnalysis = array2table(PseudovalsByParticipantByEpoch, 'VariableNames', ColumnNames2);
  1660. writetable(PseudovalTimecourseTableForAnalysis,'PseudovalTimecourseTableForAnalysis.csv')

MainITCScript.m, under CC-BY-4.0 · at the source

Overview

Authors: Laura J Batterink1, Yanchen Liu2, Grace Westerkamp2, Jae Citarella2, Peyton Siekierski2,3, Lynxie Voorhees2, Lauren E Ethridge4,5, Elizabeth Smith6, Rana Elmaghraby2,3, Craig A Erickson2,3, Zag ElSayed7, Anubhuti Goel8,9, Steve W Wu10,11, Ernest V Pedapati2,3,10,11
ORCID iDs: Laura J Batterink, Yanchen Liu, Grace Westerkamp, Peyton Siekierski, Lauren E Ethridge, Elizabeth Smith, Zag ElSayed, Steve W Wu, Ernest V Pedapati
  1. Department of Psychology; Western Centre for Brain and Mind, Western Institute for Neuroscience, University of Western Ontario, London, Ontario, Canada
  2. Division of Child and Adolescent Psychiatry, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA
  3. Department of Psychiatry, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA
  4. Department of Pediatrics, Section on Developmental and Behavioral Pediatrics, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, USA
  5. Department of Psychology, University of Oklahoma, Norman, Oklahoma, USA
  6. Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA
  7. School of Information Technology, University of Cincinnati, Ohio, USA
  8. Neuroscience Graduate Program, University of California, Riverside, Riverside, California, USA
  9. Department of Psychology, University of California, Riverside, Riverside, California, USA
  10. Division of Neurology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA
  11. Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA
Institutions: Western University (Canada); Cincinnati Children's Hospital Medical Center (United States); University of Cincinnati (United States); University of Oklahoma Health Sciences Center (United States); University of Oklahoma (United States); University of California, Riverside (United States)
Journal: Autism research : official journal of the International Society for Autism Research, volume 19, issue 8, article e70312
Dates: received 29 October 2025; accepted 3 July 2026; published online 12 July 2026; in print August 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1002/aur.70312 · PMID 42437723 · PMCID PMC13472007 · OpenAlex W4415071967
Open access: hybrid, a free copy (OpenAlex)
Preprint: doi:10.1101/2025.10.09.681399
Status: code verified
Categories: EEG (modality), human (organism), other condition (population), autism (population), cognitive (subfield)
Methods: Spectral & time-frequency, Preprocessing, Statistics, Connectivity, Smoothing, state filtering, decompositions, Physiology & signal measures
Keywords: auditory processing, EEG, electroencephalography, fragile X syndrome, language development, neural entrainment, statistical learning
MeSH: Fragile X Syndrome*, Learning Disabilities*, Speech*, Speech Perception*, Adolescent, Auditory Cortex, Child, Electroencephalography, Female, Humans, Language Development Disorders, Male, Temporal Lobe (* major topic)
Topic: Genetics and Neurodevelopmental Disorders (Genetics, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Funding: Eunice Kennedy Shriver National Institute of Child Health and Human Development (5R01HD108222-03, 5R01HD108222‐03); NICHD NIH HHS (R01 HD108222)
Citations: not cited yet (Europe PMC); 105 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above, with 5 matches between paragraphs and lines of code.

Zenodo 18746668

License: CC-BY-4.0
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Size: 3 files
Software Heritage: not checked
Found in: “Data Availability Statement”
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Statistics and Machine Learning Toolbox (3 files), EEGLAB (1 file), ERPLAB (1 file), shadedErrorBar (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers (HTTP 200)
  • 27 September 2026: the link answers (HTTP 200)
5 files
At the source: doi.org/10.5281/zenodo.18746668

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 5 scripts, each with its path and the digest of its content;
  • 5 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

Datasets cited

Code and data availability statement

The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1002/aur.70312.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

  • Publisher: n/a → Wiley

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 14 authors, 7 keywords, 13 MeSH terms, 2 funders, 104 references.

Cite

This paper

Batterink, L. J., Liu, Y., Westerkamp, G., Citarella, J., Siekierski, P., Voorhees, L., Ethridge, L. E., Smith, E., Elmaghraby, R., Erickson, C. A., ElSayed, Z., Goel, A., Wu, S. W., & Pedapati, E. V. (2026). Aberrant Neural Entrainment to Word-Level Speech Patterns in Fragile X Syndrome: Evidence for a Statistical Learning Deficit. Autism research : official journal of the International Society for Autism Research, 19(8), e70312. https://doi.org/10.1002/aur.70312

BibTeX

@article{batterink2026aberrant,
author = {Batterink, Laura J and Liu, Yanchen and Westerkamp, Grace and Citarella, Jae and Siekierski, Peyton and Voorhees, Lynxie and Ethridge, Lauren E and Smith, Elizabeth and Elmaghraby, Rana and Erickson, Craig A and ElSayed, Zag and Goel, Anubhuti and Wu, Steve W and Pedapati, Ernest V},
title = {{Aberrant Neural Entrainment to Word-Level Speech Patterns in Fragile X Syndrome: Evidence for a Statistical Learning Deficit}},
journal = {Autism research : official journal of the International Society for Autism Research},
year = {2026},
month = jul,
volume = {19},
number = {8},
pages = {e70312},
publisher = {Wiley},
issn = {1939-3792},
doi = {10.1002/aur.70312},
url = {https://doi.org/10.1002/aur.70312},
pmid = {42437723},
pmcid = {PMC13472007}
}

RIS

TY - JOUR
AU - Batterink, Laura J
AU - Liu, Yanchen
AU - Westerkamp, Grace
AU - Citarella, Jae
AU - Siekierski, Peyton
AU - Voorhees, Lynxie
AU - Ethridge, Lauren E
AU - Smith, Elizabeth
AU - Elmaghraby, Rana
AU - Erickson, Craig A
AU - ElSayed, Zag
AU - Goel, Anubhuti
AU - Wu, Steve W
AU - Pedapati, Ernest V
TI - Aberrant Neural Entrainment to Word-Level Speech Patterns in Fragile X Syndrome: Evidence for a Statistical Learning Deficit
T2 - Autism research : official journal of the International Society for Autism Research
J2 - Autism Res
PY - 2026
DA - 2026/07/12
VL - 19
IS - 8
SP - e70312
SN - 1939-3792
PB - Wiley
DO - 10.1002/aur.70312
UR - https://doi.org/10.1002/aur.70312
LA - en
ER -

CSL-JSON

{
"id": "10.1002/aur.70312",
"type": "article-journal",
"title": "Aberrant Neural Entrainment to Word-Level Speech Patterns in Fragile X Syndrome: Evidence for a Statistical Learning Deficit",
"container-title": "Autism research : official journal of the International Society for Autism Research",
"author": [
{
"family": "Batterink",
"given": "Laura J"
},
{
"family": "Liu",
"given": "Yanchen"
},
{
"family": "Westerkamp",
"given": "Grace"
},
{
"family": "Citarella",
"given": "Jae"
},
{
"family": "Siekierski",
"given": "Peyton"
},
{
"family": "Voorhees",
"given": "Lynxie"
},
{
"family": "Ethridge",
"given": "Lauren E"
},
{
"family": "Smith",
"given": "Elizabeth"
},
{
"family": "Elmaghraby",
"given": "Rana"
},
{
"family": "Erickson",
"given": "Craig A"
},
{
"family": "ElSayed",
"given": "Zag"
},
{
"family": "Goel",
"given": "Anubhuti"
},
{
"family": "Wu",
"given": "Steve W"
},
{
"family": "Pedapati",
"given": "Ernest V"
}
],
"container-title-short": "Autism Res",
"volume": "19",
"issue": "8",
"page": "e70312",
"DOI": "10.1002/aur.70312",
"PMID": "42437723",
"PMCID": "PMC13472007",
"ISSN": "1939-3792",
"publisher": "Wiley",
"URL": "https://doi.org/10.1002/aur.70312",
"language": "en",
"issued": {
"date-parts": [
[
2026,
7,
12
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1016/j.isci.2026.116458 [code]
Neural tracking of prosodic and statistical rhythms jointly supports artificial language learning.
Journal: iScience
In common: Statistics and Machine Learning Toolbox, cognitive, 13 references
[2] doi:10.7554/elife.109901
Infants at high and low likelihood for autism show different EEG developmental trajectories in speech tracking and statistical learning.
Journal: eLife
In common: autism, EEG, 12 references
[3] doi:10.1162/opmi.a.358 [code]
Tracking Visual Statistical Learning with Steady-State Visual Evoked Potentials: Effects of Exemplar and Category Information.
Journal: Open mind : discoveries in cognitive science
In common: EEG, cognitive, 11 references
[4] doi:10.1111/desc.70116 [code]
Words and Meters: Neural Evidence for a Connection Between Individual Differences in Statistical Learning and Rhythmic Ability in Infancy.
Journal: Developmental science
In common: EEG, 10 references
[5] doi:10.1162/imag.a.1161
The infant brain rapidly entrains to visual statistical regularities during stimulus exposure.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: EEG, 10 references
[6] doi:10.1016/j.dcn.2026.101767
Neural correlates of learning speed reveal developmental differences in memory.
Journal: Developmental cognitive neuroscience
In common: 8 references
[7] doi:10.1016/j.celrep.2026.117590 [code]
Impaired behavioral inhibition in Fmr1 KO mice is linked to disrupted visual cortex theta oscillations.
Journal: Cell reports
In common: Statistics and Machine Learning Toolbox, autism, EEG, other condition, 5 references
[8] doi:10.1111/cogs.70220 [code]
Shared Neural Computations for Syntactic and Morphological Structures: Evidence From Mandarin Chinese.
Journal: Cognitive science
In common: ERPLAB, EEGLAB, Statistics and Machine Learning Toolbox, EEG, cognitive, 1 reference
[9] doi:10.1038/s41467-026-73878-4 [code]
Neural Response to Familiar Names Predicts Outcome of Comatose ICU Patients: A Prospective Observational Cohort Study.
Journal: Nature communications
In common: ERPLAB, EEGLAB, Statistics and Machine Learning Toolbox, EEG, other condition, 1 reference
[10] doi:10.64898/2026.03.12.710517 [code]
Cortical excitability inversely modulates fMRI connectivity via low-frequency neuronal coupling
Journal: bioRxiv (preprint)
In common: ERPLAB, shadedErrorBar, EEGLAB, 1 other tool

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.