MasterWrapper.m~

% Walkthrough of the steps that led to the results in the paper
% 'Differential temporal dynamics of visual imagery and perception' by N.
% Dijkstra, P. Mostert, F. de Lange, S.E. Bosch & M.A.J. van Gerven
%
% For clarity, only from after preprocessing and preparation of source
% info, for full details see paper

%% Some parameters and settings

addpath('/vol/ccnlab-scratch1/naddij/shareIMAMEG/Analyses')
root     = '/vol/ccnlab-scratch1/naddij/shareIMAMEG/Data';
analyses = '/vol/ccnlab-scratch1/naddij/shareIMAMEG/Analyses';
subjects = {'S01','S03','S04','S06','S07','S08','S09','S11','S14','S15','S16','S17','S18','S19','S20','S21','S22','S23','S24','S25','S26','S27','S28','S29','S30'};
nSubjects = length(subjects);

for sub = 1:nSubjects
    
    % get the subject info
    subjectID              = subjects{sub}';  
    subjectDir             = fullfile(root,subjectID);
    dataDir                = fullfile(subjectDir,'CleanData');
    
    %% Multivariate analyses   
    addpath(fullfile(analyses,'MultivariateAnalyses'))
    
    cfg                    = [];
    cfg.root               = root;
    cfg.nFold              = 5; % number of folds for cross-validation
    cfg.channels           = 'MEG';
    
    cfg.decfg.gamma        = 0.05;
    cfg.decfg.samplemethod = 'downsample'; % to balance trial numbers
    cfg.decfg.nMeanS       = 9; % average over x samples, for smoothing
    
    % calculate the temporal generalization matrices
    cfg.outputDir          = fullfile(subjectDir,'DecodingTG');
    
    % for perception
    cfg.dataFiles{1}       = fullfile(dataDir,'dataFP'); % training data
    cfg.dataFiles{2}       = fullfile(dataDir,'dataFP'); % test data
    
    cfg.conIdx{1,1}        = [101:108,201:208]; % the markers for each class
    cfg.conIdx{1,2}        = [109:116,209:216];
    cfg.conIdx{2,1}        = cfg.conIdx{1,1};
    cfg.conIdx{2,2}        = cfg.conIdx{1,2};
    
    DecodingCross(cfg)    
    
    % for imagery
    cfg.dataFiles{1}       = fullfile(dataDir,'dataIM');
    cfg.dataFiles{2}       = fullfile(dataDir,'dataIM'); % test data
    
    cfg.conIdx{1,1}        = [121:128,221:228];
    cfg.conIdx{1,2}        = [129:136,229:236];
    cfg.conIdx{2,1}        = cfg.conIdx{1,1};
    cfg.conIdx{2,2}        = cfg.conIdx{1,2};
    
    DecodingCross(cfg)
    
    
    % calculate the cross-decoding matrices
    cfg.outputDir          = fullfile(subjectDir,'DecodingCROSS');
    
    % form perception to imagery    
    cfg.dataFiles{1}       = fullfile(dataDir,'dataFP'); % training data
    cfg.dataFiles{2}       = fullfile(dataDir,'dataIM'); % test data
    
    cfg.conIdx{1,1}        = [101:108,201:208];
    cfg.conIdx{1,2}        = [109:116,209:216];
    cfg.conIdx{2,1}        = [121:128,221:228];
    cfg.conIdx{2,2}        = [129:136,229:236];
    
    DecodingCross(cfg)
    
    % from imagery to perception 
    cfg.dataFiles{1}       = fullfile(dataDir,'dataIM'); % training data
    cfg.dataFiles{2}       = fullfile(dataDir,'dataFP'); % test data
    
    cfg.conIdx{1,1}        = [121:128,221:228];
    cfg.conIdx{1,2}        = [129:136,229:236];
    cfg.conIdx{2,1}        = [101:108,201:208];
    cfg.conIdx{2,2}        = [109:116,209:216];
    
    DecodingCross(cfg)    
    
    %% Vividness correlation
    % takes the output from the classifier and correlates it with vividness
    cfg          = [];
    cfg.dataDir  = subjectDir;
    cfg.dataName = '/DecodingCROSS/dataFP to dataIM.mat';
    DecoderVividnessOutput(cfg)
    
    %% Source reconstruction
    addpath(fullfile(analyses,'SourceReconstruction'))    
    
    % calculate source activity difference ERF
    cfg             = [];
    cfg.movie       = 'no';
    cfg.dataDir     = dataDir;
    cfg.output      = fullfile(subjectDir,'SourceReconstruction');
    cfg.subjectID   = subjectID;
    cfg.time        = [];%[0 1];
    cfg.lpfilter    = 'yes';
    cfg.dataName    = 'dataIM';
    cfg.conIdx{1}   = [101:108,121:128,221:228]; cfg.conIdx{2} = [109:116,129:136,229:236];
    SourcerecConDiff(cfg)
    
    cfg.dataName    = 'dataFP';
    cfg.time        = [];
    SourcerecConDiff(cfg)
    
    % seperate grid points into atlas regions for later averaging
    cfg             = [];
    cfg.plot        = 'no';
    cfg.diff        = 'yes'; % look at the difference or average over conditions
    cfg.dataSet     = 'dataIM';
    cfg.subjectID   = subjectID;
    cfg.sourceRecDir = fullfile(subjectDir,'SourceReconstruction');
    cfg.root        = root;
    SourcerecAtlas(cfg)
    
    cfg.dataSet     = 'dataFP';    
    SourcerecAtlas(cfg)       
    
end

%% Group analyses decoding results

% perception temporal generalization
cfg.dataName = 'dataFP';
cfg.dirName  = 'DecodingTG';
cfg.root     = root;
cfg.nPerm    = 1000; % number of permutations
cfg.alpha    = 0.05;
cfg.subjects = subjects;
load(fullfile(root,subjects{1},cfg.dirName,cfg.dataName),'time') % steal a time variabel
cfg.time     = time;

DecodingGA(cfg); 

% imagery temporal generalization
cfg.dataName = 'dataFP';
load(fullfile(root,subjects{1},cfg.dirName,cfg.dataName),'time') % steal a time variabel
cfg.time     = time;

DecodingGA(cfg);

% cross-decoding from perception to imagery


% cross-decoding from imagery to perception


%% Make figures