tylers_similarity_backup.m

function outString = tyler_similarity(fixMaps, removedAt, t)
    % Initialization
    outString = '';
    fixMapsSize = size(fixMaps);

    % For each participant...
    for i=1:fixMapsSize(3)
        % If it was one we skipped due to missing data, just write N/A.
        if (ismember(i,removedAt))
            outString = [outString 'N/A,'];
        else
            % Pre allocate just to be safe
            summate = zeros(fixMapsSize(1),fixMapsSize(2));
            % Add everything together
            for j=1:fixMapsSize(3)
                % Leave our subject out of the averaging if it's their turn
                if i == j
                    continue; 
                end
                summate = summate + fixMaps(:,:,j);
            end
            % Scale the summation to become the mean
            meanMatrix = summate./(fixMapsSize(3) - 1);
            % Reshape 2D data to 1D samples and meld together
            % Done with magic matlab syntax
            linearMeanMatrix = meanMatrix(:);
            linearMeanMatrix = linearMeanMatrix';
            % linearMeanMatrix = linearMeanMatrix'; % Just flipping again - trust me
            linearIndivMatrix = fixMaps(:,:,i);
            linearIndivMatrix = linearIndivMatrix(:);
            linearIndivMatrix = linearIndivMatrix';
            statsMatrix = cat(2,linearIndivMatrix',linearMeanMatrix');
            % Do the stats!
            cMat=corr(statsMatrix);
            cMat=cMat(find(triu(ones(size(cMat)),1)));
            r=tanh(nanmean(atanh(cMat)));
            % Save to output string
            outString = [outString num2str(r) ','];
        end  
    end
end