scripts/BEH_VV.m

%% Calculate Data
clear all; clc
datapath=('Y:\EEG_Data\PDDys\BEH\');
cd(datapath);

SUBJS=[801:811,813:829];  % Only include here if they have BOTH sessions done!

for subno=SUBJS
    for session=1:2  
        
        disp(['VV Beh --- Subno: ',num2str(subno),'  Session: ',num2str(session)]); disp(' ');
        
        % TRAIN
        fileID=dir([num2str(subno),'_S',num2str(session),'*']);
        load(fileID.name);  clear fileID
        
        % ----- Columns are: -----
        % 1 = block
        % 2 = trial
        % 3 =              (cTrialID)
        % 4 = Forced choice
        % 6 = S1
        % 8 = S2
        % 10 = S1 is left
        % 11 = Stim Selected
        % 13 = Reward (1 or 0)
        % 14 = RT
        
        % ----------- Re-make With More Simpleness
        ABchoose=[1,2]; CDchoose=[3,4]; WXmatch=[5,6]; YZmatch=[7,8];
        for ai=1:length(task_struct.trainTrials)
            if     any(task_struct.trainTrials(ai,6)==ABchoose)
                CONDI=1;
            elseif any(task_struct.trainTrials(ai,6)==CDchoose)
                CONDI=2;
            elseif any(task_struct.trainTrials(ai,6)==WXmatch)
                CONDI=3;
            elseif any(task_struct.trainTrials(ai,6)==YZmatch)
                CONDI=4;
            end
            TRAIN(ai,1)=CONDI;      clear CONDI;                 % Condi: 1=ABchoose  2=CDchoose  3=ABmatch   4=CDmatch
            TRAIN(ai,2)=mod(task_struct.trainTrials(ai,11),2);   % Optimal choice (odd num are optimal: 1,3,5,7)
            TRAIN(ai,3)=task_struct.trainTrials(ai,14);  % RT
            TRAIN(ai,4)=task_struct.trainTrials(ai,13);  % was rewarded or not
        end
        
        % ----------- Re-make With More Simpleness
        
        for ai=1:length(task_struct.testTrials)
            ThisSet=task_struct.testTrials(ai,[6,8]);
            ThisSet=str2num(cat(2,num2str(ThisSet(1)),num2str(ThisSet(2))));
            ThisChoice=task_struct.testTrials(ai,11);
            
            if     any(ThisSet==[12,21]),  CONDI='AB';
            elseif any(ThisSet==[13,31]),  CONDI='AC';
            elseif any(ThisSet==[14,41]),  CONDI='AD';
            elseif any(ThisSet==[15,51]),  CONDI='AW';
            elseif any(ThisSet==[16,61]),  CONDI='AX';
            elseif any(ThisSet==[17,71]),  CONDI='AY';
            elseif any(ThisSet==[18,81]),  CONDI='AZ';
            elseif any(ThisSet==[23,32]),  CONDI='BC';
            elseif any(ThisSet==[24,42]),  CONDI='BD';
            elseif any(ThisSet==[25,52]),  CONDI='BW';
            elseif any(ThisSet==[26,62]),  CONDI='BX';
            elseif any(ThisSet==[27,72]),  CONDI='BY';
            elseif any(ThisSet==[28,82]),  CONDI='BZ';
            elseif any(ThisSet==[34,43]),  CONDI='CD';
            elseif any(ThisSet==[35,53]),  CONDI='CW';
            elseif any(ThisSet==[36,63]),  CONDI='CX';
            elseif any(ThisSet==[37,73]),  CONDI='CY';
            elseif any(ThisSet==[38,83]),  CONDI='CZ';
            elseif any(ThisSet==[45,54]),  CONDI='DW';
            elseif any(ThisSet==[46,64]),  CONDI='DX';
            elseif any(ThisSet==[47,74]),  CONDI='DY';
            elseif any(ThisSet==[48,84]),  CONDI='DZ';
            elseif any(ThisSet==[56,65]),  CONDI='WX';
            elseif any(ThisSet==[57,75]),  CONDI='WY';
            elseif any(ThisSet==[58,85]),  CONDI='WZ';
            elseif any(ThisSet==[67,76]),  CONDI='XY';
            elseif any(ThisSet==[68,86]),  CONDI='XZ';
            elseif any(ThisSet==[78,87]),  CONDI='YZ';
            end
            
            TEST(ai).condi=CONDI;                       % Condi
            TEST(ai).choice=ThisChoice;                 % This Choice
            TEST(ai).RT=task_struct.testTrials(ai,14);  % RT
            clear ThisSet ThisChoice CONDI;
        end
        
        save([num2str(subno),'_sess',num2str(session),'_VVbeh.mat'],'TRAIN','TEST');
        
        clear task_struct disp_struct AB* CD* TRAIN TEST
    end
end

load('Y:\EEG_Data\PDDys\ONOFF.mat','ONOFF')

row=0;
for subno=SUBJS
    for session=1:2;
        row=row+1;
        
        if subno~=ONOFF(row,1) || session~=ONOFF(row,2), BOOM; end  % Kills it if mismatch in ON/OFF Matrix!
        
        load([num2str(subno),'_sess',num2str(session),'_VVbeh.mat'],'TRAIN','TEST');
        
        MEGA(row).ID=subno;
        MEGA(row).session=session;
        MEGA(row).ONOFF=ONOFF(row,3);
        MEGA(row).TRN_blocks=size(TRAIN,1)./40;
        MEGA(row).TRN_ACC=mean(TRAIN(:,2));
        MEGA(row).TRN_RT=mean(TRAIN(:,3));
        for bi=1:4 % Condi: 1=ABchoose  2=CDchoose  3=ABmatch   4=CDmatch
            MINISET=TRAIN(TRAIN(:,1)==bi,:);
            MINISET(:,[5,6])=[MINISET(2:end,[2,3]);[NaN,NaN]];
            MINISET(:,6)=MINISET(:,6)-MINISET(:,3);  % RT diff
            % --------
            WINS=MINISET(MINISET(:,4)==1,[2,5,6]);
            LOSSES=MINISET(MINISET(:,4)==0,[2,5,6]);
            % --------
            WinStay(bi)=mean(WINS(:,1)==WINS(:,2));
            LoseSwitch(bi)=mean(LOSSES(:,1)~=LOSSES(:,2));
            WinSpeed(bi)=mean(WINS(WINS(:,1)==WINS(:,2),3));
            % --------
            clear MINISET WINS LOSSES
        end
        
        MEGA(row).WinStay=WinStay;
        MEGA(row).LoseSwitch=LoseSwitch;
        MEGA(row).WinSpeed=WinSpeed;
        
        % ********************************************************
        for ci=1:128
            RT(ci,1)=TEST(ci).RT;
            % ^^^^ General Accuracy     A,B,C,D  ==   W,X,Y,Z           - 4 of each set
            ACC=NaN; PARSE=NaN;
            if     strmatch(TEST(ci).condi,'AB');
               if TEST(ci).choice==1, ACC=1; elseif TEST(ci).choice==2, ACC=0; end; PARSE=1;
            elseif strmatch(TEST(ci).condi,'WX');
               if TEST(ci).choice==5, ACC=1; elseif TEST(ci).choice==6, ACC=0; end; PARSE=2;
            elseif strmatch(TEST(ci).condi,'CD');
               if TEST(ci).choice==3, ACC=1; elseif TEST(ci).choice==4, ACC=0; end; PARSE=3; 
            elseif strmatch(TEST(ci).condi,'YZ');
               if TEST(ci).choice==7, ACC=1; elseif TEST(ci).choice==8, ACC=0; end; PARSE=4; 
            end
            % ^^^^ free vs. forced   A,B,C,D  ==   W,X,Y,Z             - 8 of each set
            BIAS=NaN;
            if     strmatch(TEST(ci).condi,'AW');
               if TEST(ci).choice==1, BIAS=1; elseif TEST(ci).choice==5, BIAS=0; end; PARSE=5;
            elseif strmatch(TEST(ci).condi,'CY');
               if TEST(ci).choice==3, BIAS=1; elseif TEST(ci).choice==7, BIAS=0; end; PARSE=6; 
            elseif strmatch(TEST(ci).condi,'DZ');
               if TEST(ci).choice==4, BIAS=1; elseif TEST(ci).choice==8, BIAS=0; end; PARSE=7;
            elseif strmatch(TEST(ci).condi,'BX');
               if TEST(ci).choice==2, BIAS=1; elseif TEST(ci).choice==6, BIAS=0; end; PARSE=8; 
            end
            % ^^^^ 
            WITHINSET=NaN;
            if     strmatch(TEST(ci).condi,'AC');
               if TEST(ci).choice==1, WITHINSET=1; elseif TEST(ci).choice==3, WITHINSET=0; end; PARSE=9;
            elseif strmatch(TEST(ci).condi,'AD');
               if TEST(ci).choice==1, WITHINSET=1; elseif TEST(ci).choice==4, WITHINSET=0; end; PARSE=10; 
            elseif strmatch(TEST(ci).condi,'BC');
               if TEST(ci).choice==3, WITHINSET=1; elseif TEST(ci).choice==2, WITHINSET=0; end; PARSE=11;
            elseif strmatch(TEST(ci).condi,'BD');
               if TEST(ci).choice==4, WITHINSET=1; elseif TEST(ci).choice==2, WITHINSET=0; end; PARSE=12; 
            elseif strmatch(TEST(ci).condi,'WY');
               if TEST(ci).choice==5, WITHINSET=1; elseif TEST(ci).choice==7, WITHINSET=0; end; PARSE=13;
            elseif strmatch(TEST(ci).condi,'WZ');
               if TEST(ci).choice==5, WITHINSET=1; elseif TEST(ci).choice==8, WITHINSET=0; end; PARSE=14; 
            elseif strmatch(TEST(ci).condi,'XY');
               if TEST(ci).choice==7, WITHINSET=1; elseif TEST(ci).choice==6, WITHINSET=0; end; PARSE=15;
            elseif strmatch(TEST(ci).condi,'XZ');
               if TEST(ci).choice==8, WITHINSET=1; elseif TEST(ci).choice==6, WITHINSET=0; end; PARSE=16; 
            end
            % ^^^^
            EASY=NaN;
            if     strmatch(TEST(ci).condi,'AX');
                if TEST(ci).choice==1, EASY=1; elseif TEST(ci).choice==5, EASY=0; end; PARSE=17;
            elseif strmatch(TEST(ci).condi,'AY');
                if TEST(ci).choice==1, EASY=1; elseif TEST(ci).choice==7, EASY=0; end; PARSE=18;
            elseif strmatch(TEST(ci).condi,'AZ');
                if TEST(ci).choice==1, EASY=1; elseif TEST(ci).choice==8, EASY=0; end; PARSE=19;
            elseif strmatch(TEST(ci).condi,'BW');
                if TEST(ci).choice==5, EASY=1; elseif TEST(ci).choice==2, EASY=0; end; PARSE=20;
            elseif strmatch(TEST(ci).condi,'CW');
                if TEST(ci).choice==5, EASY=1; elseif TEST(ci).choice==3, EASY=0; end; PARSE=21;
            elseif strmatch(TEST(ci).condi,'DW');
                if TEST(ci).choice==5, EASY=1; elseif TEST(ci).choice==4, EASY=0; end; PARSE=22;
            end
            % ^^^^
            MEDIUM=NaN;
            if     strmatch(TEST(ci).condi,'CX');
                if TEST(ci).choice==3, MEDIUM=1; elseif TEST(ci).choice==6, MEDIUM=0; end; PARSE=23;
            elseif strmatch(TEST(ci).condi,'CZ');
                if TEST(ci).choice==3, MEDIUM=1; elseif TEST(ci).choice==8, MEDIUM=0; end; PARSE=24;
            elseif strmatch(TEST(ci).condi,'BY');
                if TEST(ci).choice==7, MEDIUM=1; elseif TEST(ci).choice==2, MEDIUM=0; end; PARSE=25;
            elseif strmatch(TEST(ci).condi,'DY');
                if TEST(ci).choice==7, MEDIUM=1; elseif TEST(ci).choice==4, MEDIUM=0; end; PARSE=26;
            end
            % ^^^^
            HARD=NaN;
            if     strmatch(TEST(ci).condi,'BZ');
                if TEST(ci).choice==8, HARD=1; elseif TEST(ci).choice==2, HARD=0; end; PARSE=27;
            elseif strmatch(TEST(ci).condi,'DX');
                if TEST(ci).choice==4, HARD=1; elseif TEST(ci).choice==6, HARD=0; end; PARSE=28;
            end
            % ^^^^ 
            TST_ACC(ci,1)=ACC;
            TST_BIAS(ci,1)=BIAS;
            TST_WITHINSET(ci,1)=WITHINSET;
            TST_EASY(ci,1)=EASY;
            TST_MEDIUM(ci,1)=MEDIUM;
            TST_HARD(ci,1)=HARD;
            TST_PARSE(ci,1)=PARSE;
            clear ACC BIAS WITHINSET EASY MEDIUM HARD PARSE;
        end
        
        for di=1:4
            ACCURACIES(di)=nanmean(TST_ACC(TST_PARSE==di));
        end
        for di=5:8
            BIASES(di-4)=nanmean(TST_BIAS(TST_PARSE==di));
        end
        for di=9:16
            WITHINSETS(di-8)=nanmean(TST_WITHINSET(TST_PARSE==di));
        end
        for di=17:22
            EASYS(di-16)=nanmean(TST_EASY(TST_PARSE==di));
        end
        for di=23:26
            MEDIUMS(di-22)=nanmean(TST_MEDIUM(TST_PARSE==di));
        end
        for di=27:28
            HARDS(di-26)=nanmean(TST_HARD(TST_PARSE==di));
        end
        
        
        MEGA(row).TST_ACC=ACCURACIES;
        MEGA(row).TST_BIAS=BIASES;
        MEGA(row).TST_WITHINSET=WITHINSETS;
        MEGA(row).TST_EASY=EASYS;
        MEGA(row).TST_MEDIUM=MEDIUMS;
        MEGA(row).TST_HARD=HARDS;
        MEGA(row).TST_RT=mean(RT);
        
        clearvars -except MEGA subjcount subno session RT row ONOFF SUBJS;
    end
end
save('VV_Behavior.mat','MEGA');

clear ONOFF RT row session subno

%%
clear all; clc
datapath=('Y:\EEG_Data\PDDys\BEH\');
cd(datapath);
SUBJS=[801:811,813:829];  % Only include here if they have BOTH sessions done!

load('Y:\EEG_Data\PDDys\PD_Moderators.mat','Mods','Mods_Hdr')

load('VV_Behavior.mat','MEGA');

row=0; 
for subno=SUBJS
    row=row+1;
    for mi=1:size(MEGA,2)
       if MEGA(mi).ID==subno && MEGA(mi).ONOFF==1
           ON.ID(row,:)=MEGA(mi).ID;
           ON.session(row,:)=MEGA(mi).session;
           ON.TRN_ACC(row,:)=MEGA(mi).TRN_ACC;
           ON.TRN_RT(row,:)=MEGA(mi).TRN_RT;
           ON.WinStay(row,:)=MEGA(mi).WinStay;
           ON.LoseSwitch(row,:)=MEGA(mi).LoseSwitch;
           ON.WinSpeed(row,:)=MEGA(mi).WinSpeed;
           ON.TST_ACC(row,:)=MEGA(mi).TST_ACC;
           ON.TST_BIAS(row,:)=MEGA(mi).TST_BIAS;
           ON.TST_WITHINSET(row,:)=MEGA(mi).TST_WITHINSET;
           ON.TST_EASY(row,:)=MEGA(mi).TST_EASY;
           ON.TST_MEDIUM(row,:)=MEGA(mi).TST_MEDIUM;
           ON.TST_HARD(row,:)=MEGA(mi).TST_HARD;
           ON.TST_RT(row,:)=MEGA(mi).TST_RT;
           ON.Blocks(row,:)=MEGA(mi).TRN_blocks;
       elseif MEGA(mi).ID==subno && MEGA(mi).ONOFF==0
           OFF.ID(row,:)=MEGA(mi).ID;
           OFF.session(row,:)=MEGA(mi).session;
           OFF.TRN_ACC(row,:)=MEGA(mi).TRN_ACC;
           OFF.TRN_RT(row,:)=MEGA(mi).TRN_RT;
           OFF.WinStay(row,:)=MEGA(mi).WinStay;
           OFF.LoseSwitch(row,:)=MEGA(mi).LoseSwitch;
           OFF.WinSpeed(row,:)=MEGA(mi).WinSpeed;
           OFF.TST_ACC(row,:)=MEGA(mi).TST_ACC;
           OFF.TST_BIAS(row,:)=MEGA(mi).TST_BIAS;
           OFF.TST_WITHINSET(row,:)=MEGA(mi).TST_WITHINSET;
           OFF.TST_EASY(row,:)=MEGA(mi).TST_EASY;
           OFF.TST_MEDIUM(row,:)=MEGA(mi).TST_MEDIUM;
           OFF.TST_HARD(row,:)=MEGA(mi).TST_HARD;
           OFF.TST_RT(row,:)=MEGA(mi).TST_RT;
           OFF.Blocks(row,:)=MEGA(mi).TRN_blocks;
       end
    end
end

save('BEH_VV_PD','ON','OFF');

BigN=length(SUBJS);
jitter=rand(1,BigN)./2.5; jitter=jitter-mean(jitter);


% % SOMESX=double(repmat((Mods(:,9)>nanmedian(Mods(:,9))),1,4));
% % SOMESX(SOMESX==0)=NaN;
% % ON.TST_BIAS=ON.TST_BIAS.*SOMESX;
% % OFF.TST_BIAS=OFF.TST_BIAS.*SOMESX;


%%
figure;
subplot(1,3,1); hold on
bar(1:4,mean(ON.WinStay),.25,'w');
bar(1.25:4.25,mean(OFF.WinStay),.25,'r');
errorbar(1:4,mean(ON.WinStay),std(ON.WinStay)./sqrt(BigN),'k.');
errorbar(1.25:4.25,mean(OFF.WinStay),std(OFF.WinStay)./sqrt(BigN),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','CD','WY','XZ'},'ylim',[.5 1]);
title('TRN WinStay');
subplot(1,3,2); hold on
bar(1:4,nanmean(ON.LoseSwitch),.25,'w');
bar(1.25:4.25,nanmean(OFF.LoseSwitch),.25,'r');
errorbar(1:4,nanmean(ON.LoseSwitch),nanstd(ON.LoseSwitch)./sqrt(BigN),'k.');
errorbar(1.25:4.25,nanmean(OFF.LoseSwitch),nanstd(OFF.LoseSwitch)./sqrt(BigN),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','CD','WY','XZ'});
title('TRN LoseSwitch');
subplot(1,3,3); hold on
bar(1:4,nanmean(ON.WinSpeed),.25,'w');
bar(1.25:4.25,nanmean(OFF.WinSpeed),.25,'r');
errorbar(1:4,nanmean(ON.WinSpeed),nanstd(ON.WinSpeed)./sqrt(BigN),'k.');
errorbar(1.25:4.25,nanmean(OFF.WinSpeed),nanstd(OFF.WinSpeed)./sqrt(BigN),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','CD','WY','XZ'});
title('TRN WinSpeed');


%%
figure;
subplot(2,5,1); hold on
bar(1,mean(ON.TRN_ACC),'w');
bar(2,mean(OFF.TRN_ACC),'r');
errorbar(1,mean(ON.TRN_ACC),std(ON.TRN_ACC)./sqrt(BigN),'k.');
errorbar(2,mean(OFF.TRN_ACC),std(OFF.TRN_ACC)./sqrt(BigN),'k.');
% plot(1,ON.TRN_ACC,'b.'); plot(2,OFF.TRN_ACC,'b.');
% plot([1 2],[ON.TRN_ACC OFF.TRN_ACC],'b-');
set(gca,'xlim',[0 3],'xtick',[1:1:2],'xticklabel',{'ON','OFF'},'ylim',[.5 1]);
title('TRN Acc');

subplot(2,5,2:3); hold on
bar(1:4,mean(ON.TST_ACC),.25,'w');
bar(1.25:1:4.25,mean(OFF.TST_ACC),.25,'r');
errorbar(1:4,mean(ON.TST_ACC),std(ON.TST_ACC)./sqrt(BigN),'k.');
errorbar(1.25:1:4.25,mean(OFF.TST_ACC),std(OFF.TST_ACC)./sqrt(BigN),'k.');
% plot(1:4,ON.TST_ACC,'b.'); plot(1.25:1:4.25,OFF.TST_ACC,'b.');
% plot([1 1.25],[ON.TST_ACC(:,1) OFF.TST_ACC(:,1)],'b-');
% plot([2 2.25],[ON.TST_ACC(:,2) OFF.TST_ACC(:,2)],'b-');
% plot([3 3.25],[ON.TST_ACC(:,3) OFF.TST_ACC(:,3)],'b-');
% plot([4 4.25],[ON.TST_ACC(:,4) OFF.TST_ACC(:,4)],'b-');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','WX','CD','YZ'});
title('TST Acc');

subplot(2,5,4:5); hold on
bar(1:4,nanmean(ON.TST_BIAS),.25,'w');
bar(1.25:1:4.25,nanmean(OFF.TST_BIAS),.25,'r');
errorbar(1:4,nanmean(ON.TST_BIAS),nanstd(ON.TST_BIAS)./sqrt(BigN),'k.');
errorbar(1.25:1:4.25,nanmean(OFF.TST_BIAS),nanstd(OFF.TST_BIAS)./sqrt(BigN),'k.');
% plot(1:4,ON.TST_BIAS,'b.'); plot(1.25:1:4.25,OFF.TST_BIAS,'b.');
% plot([1 1.25],[ON.TST_BIAS(:,1) OFF.TST_BIAS(:,1)],'b-');
% plot([2 2.25],[ON.TST_BIAS(:,2) OFF.TST_BIAS(:,2)],'b-');
% plot([3 3.25],[ON.TST_BIAS(:,3) OFF.TST_BIAS(:,3)],'b-');
% plot([4 4.25],[ON.TST_BIAS(:,4) OFF.TST_BIAS(:,4)],'b-');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AW','CY','DZ','BX'});
title('TST BIAS');



subplot(2,5,6); hold on
bar(1,mean(ON.TRN_ACC-OFF.TRN_ACC),'b');
errorbar(1,mean(ON.TRN_ACC-OFF.TRN_ACC),std(ON.TRN_ACC-OFF.TRN_ACC)./sqrt(BigN),'k.');
set(gca,'xlim',[0 2],'xtick',[1:1:1]);
title('TRN Acc DIFF');

subplot(2,5,7:8); hold on
bar(1:4,mean(ON.TST_ACC-OFF.TST_ACC),.25,'b');
errorbar(1:4,mean(ON.TST_ACC-OFF.TST_ACC),std(ON.TST_ACC-OFF.TST_ACC)./sqrt(BigN),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','WX','CD','YZ'});
title('TST Acc');


subplot(2,5,9:10); hold on
bar(1:4,nanmean( ON.TST_BIAS-OFF.TST_BIAS ),.25,'w');
errorbar(1:4,nanmean(ON.TST_BIAS-OFF.TST_BIAS ),std(ON.TST_BIAS-OFF.TST_BIAS )./sqrt(BigN),'k.');
% for plotdiffi=1:4
%     plot(plotdiffi+jitter',(ON.TST_BIAS(:,plotdiffi)-OFF.TST_BIAS(:,plotdiffi)),'mo');
% end
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AW','CY','DZ','BX'});
title('TST BIAS');

%%
MODS=9;

BigDiff=ON.TST_BIAS-OFF.TST_BIAS;
for polyi=1:BigN
    POLY{1}(polyi)=corr(ON.TST_BIAS(polyi,:)',(1:4)');
    POLY{2}(polyi)=corr(OFF.TST_BIAS(polyi,:)',(1:4)');
    POLY{3}(polyi)=corr(BigDiff(polyi,:)',(1:4)');
end

dispidx=3;
figure;
subplot(1,2,1); hold on
bar(mean(POLY{dispidx}),'w');
plot(.86:.01:1.13,POLY{dispidx},'bd');
errorbar(mean(POLY{dispidx}),std(POLY{dispidx})./sqrt(BigN),'k.')
set(gca,'xlim',[0 2],'xtick',[1:1:2])
ylabel('Slope in Accuracy Difference');
[H,P,CI,STATS]=ttest(POLY{dispidx})
title(['BIAS t=',num2str(STATS.tstat),' p=',num2str(P)]);
subplot(1,2,2); hold on
scatter( Mods(:,MODS) , POLY{dispidx}' ,'k'); lsline
[rho,p]=corr( Mods(:,MODS) , POLY{dispidx}' ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['POLY & ',Mods_Hdr{MODS}])

FORSPSS=[BigDiff,POLY{3}'];

for slopei=1:4
    if slopei==1,     X=ON.TST_BIAS; 
    elseif slopei==2, X=OFF.TST_BIAS; 
    elseif slopei==3,  X=CTL.TST_BIAS;   
    elseif slopei==4,  X=ON.TST_BIAS-OFF.TST_BIAS;
    end
    for polyi=1:length(X)
        SLOPES{slopei}(polyi)=corr(X(polyi,:)',(1:4)');
    end
end

figure; 
subplot(2,2,1);  hold on
bar(1,mean(SLOPES{1}),'w');
bar(2,mean(SLOPES{2}),'r');
bar(3,mean(SLOPES{3}),'g');
errorbar(1,mean(SLOPES{1}),std(SLOPES{1})./sqrt(BigN),'k.');
errorbar(2,mean(SLOPES{2}),std(SLOPES{2})./sqrt(BigN),'k.');
errorbar(3,mean(SLOPES{3}),std(SLOPES{3})./sqrt(BigN_ctl),'k.');
ylabel('Slope in Accuracy Difference');
subplot(2,2,2);  hold on
scatter( Mods(:,MODS) , SLOPES{4} ,'b'); lsline
[rho,p]=corr( Mods(:,MODS) , SLOPES{4}' ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['ON-OFF & ',Mods_Hdr{MODS}])
subplot(2,2,3);  hold on
scatter( Mods(:,MODS) , SLOPES{1} ,'k'); lsline
[rho,p]=corr( Mods(:,MODS) , SLOPES{1}' ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['ON & ',Mods_Hdr{MODS}])
subplot(2,2,4);  hold on
scatter( Mods(:,MODS) , SLOPES{2} ,'r'); lsline
[rho,p]=corr( Mods(:,MODS) , SLOPES{2}' ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['OFF & ',Mods_Hdr{MODS}])


%%

% LED & win-speed diff
% LED & BX bias diff

MOD_IDX=2;

A1=sum(OFF.TST_BIAS(:,1),2)
A2=sum(ON.TST_BIAS(:,1),2)
A3=A1-A2;

figure; 
subplot(1,3,1); hold on
scatter( Mods(:,MOD_IDX) , A1 ,'k'); lsline
[rho,p]=corr( Mods(:,MOD_IDX) , A1 ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['OFF & ',Mods_Hdr{MOD_IDX}])
subplot(1,3,2); hold on
scatter( Mods(:,MOD_IDX) ,A2 ,'k' ); lsline
[rho,p]=corr( Mods(:,MOD_IDX) , A2 ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['ON & ',Mods_Hdr{MOD_IDX}])
subplot(1,3,3); hold on
scatter( Mods(:,MOD_IDX) , A3 ,'k'  ); lsline
[rho,p]=corr( Mods(:,MOD_IDX) , A3 ,'type','Spearman','rows','complete' );
text(.4,.1,['r=',num2str(rho),' p=',num2str(p)],'sc')
title(['OFF-ON & ',Mods_Hdr{MOD_IDX}])

%%
load('VV_Behavior_CTL.mat','CTL');

BigN=length(SUBJS);
jitter=rand(1,BigN)./2.5; jitter=jitter-mean(jitter);
BigN_ctl=size(CTL.ID,1);
noise_ctl=rand(1,BigN_ctl)./100;

figure;
subplot(2,5,1); hold on
bar(1,mean(ON.TRN_ACC),'w');
bar(2,mean(OFF.TRN_ACC),'r');
bar(3,mean(CTL.TRN_ACC),'g');
errorbar(1,mean(ON.TRN_ACC),std(ON.TRN_ACC)./sqrt(BigN),'k.');
errorbar(2,mean(OFF.TRN_ACC),std(OFF.TRN_ACC)./sqrt(BigN),'k.');
errorbar(3,mean(CTL.TRN_ACC),std(CTL.TRN_ACC)./sqrt(BigN_ctl),'k.');
set(gca,'xlim',[0 4],'xtick',[1:1:3],'xticklabel',{'ON','OFF','CTL'},'ylim',[.5 1]);
title('TRN Acc');

subplot(2,5,2:3); hold on
bar(1-.25:4-.25,mean(ON.TST_ACC),.25,'w');
bar(1:1:4,mean(OFF.TST_ACC),.25,'r');
bar(1.25:1:4.25,mean(CTL.TST_ACC),.25,'g');
errorbar(1-.25:4-.25,mean(ON.TST_ACC),std(ON.TST_ACC)./sqrt(BigN),'k.');
errorbar(1:1:4,mean(OFF.TST_ACC),std(OFF.TST_ACC)./sqrt(BigN),'k.');
errorbar(1.25:1:4.25,mean(CTL.TST_ACC),std(CTL.TST_ACC)./sqrt(BigN_ctl),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','WX','CD','YZ'});
title('TST Acc');

subplot(2,5,4:5); hold on
bar(1-.25:4-.25,mean(ON.TST_BIAS),.25,'w');
bar(1:1:4,mean(OFF.TST_BIAS),.25,'r');
bar(1.25:1:4.25,mean(CTL.TST_BIAS),.25,'g');
errorbar(1-.25:4-.25,mean(ON.TST_BIAS),std(ON.TST_BIAS)./sqrt(BigN),'k.');
errorbar(1:1:4,mean(OFF.TST_BIAS),std(OFF.TST_BIAS)./sqrt(BigN),'k.');
errorbar(1.25:1:4.25,mean(CTL.TST_BIAS),std(CTL.TST_BIAS)./sqrt(BigN_ctl),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AW','CY','DZ','BX'});
title('TST BIAS');

[H,P,CI,STATS]=ttest2(ON.TST_BIAS(:,1),CTL.TST_BIAS(:,1))
[H,P,CI,STATS]=ttest2(OFF.TST_BIAS(:,1),CTL.TST_BIAS(:,1))

subplot(2,5,6); hold on
bar(1,mean(ON.TRN_ACC-OFF.TRN_ACC),'b');
errorbar(1,mean(ON.TRN_ACC-OFF.TRN_ACC),std(ON.TRN_ACC-OFF.TRN_ACC)./sqrt(BigN),'k.');
set(gca,'xlim',[0 2],'xtick',[1:1:1]);
title('TRN Acc DIFF');

subplot(2,5,7:8); hold on
bar(1:4,mean(ON.TST_ACC-OFF.TST_ACC),.25,'b');
errorbar(1:4,mean(ON.TST_ACC-OFF.TST_ACC),std(ON.TST_ACC-OFF.TST_ACC)./sqrt(BigN),'k.');
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AB','WX','CD','YZ'});
title('TST Acc DIFF');

subplot(2,5,9:10); hold on
bar(1:4,mean(ON.TST_BIAS-OFF.TST_BIAS),.25,'w');
errorbar(1:4,mean(ON.TST_BIAS-OFF.TST_BIAS),std(ON.TST_BIAS-OFF.TST_BIAS)./sqrt(BigN),'k.');
% % for plotdiffi=1:4
% %     plot(plotdiffi+jitter',(ON.TST_BIAS(:,plotdiffi)-OFF.TST_BIAS(:,plotdiffi)),'mo');
% % end
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'AW','CY','DZ','BX'});
title('TST BIAS DIFF');

%%

%%


figure; 
subplot(3,2,1:2)
hold on
bar(1-.25:1:8-.25,mean(VV.CTL.TST_WITHINSET),.25,'g');
bar(1:1:8,mean(VV.ON.TST_WITHINSET),.25,'w');
bar(1+.25:1:8+.25,mean(VV.OFF.TST_WITHINSET),.25,'r');
errorbar(1-.25:1:8-.25,mean(VV.CTL.TST_WITHINSET),std(VV.CTL.TST_WITHINSET)./sqrt(BigN_ctl),'k.');
errorbar(1:1:8,mean(VV.ON.TST_WITHINSET),std(VV.ON.TST_WITHINSET)./sqrt(BigN),'k.');
errorbar(1+.25:1:8+.25,mean(VV.OFF.TST_WITHINSET),std(VV.OFF.TST_WITHINSET)./sqrt(BigN),'k.');
plot([0 9],[.5 .5],'b:')
set(gca,'xlim',[0 9],'xtick',[1:1:8],'xticklabel',{'AC','AD','CB','BD','WY','WZ','YX','XZ'},'ytick',[0:.25:1]);
title('TST w/in set Acc');

subplot(3,2,3:4)
hold on
bar(1-.25:1:6-.25,nanmean(VV.CTL.TST_EASY),.25,'g');
bar(1:1:6,nanmean(VV.ON.TST_EASY),.25,'w');
bar(1+.25:1:6+.25,nanmean(VV.OFF.TST_EASY),.25,'r');
errorbar(1-.25:1:6-.25,nanmean(VV.CTL.TST_EASY),nanstd(VV.CTL.TST_EASY)./sqrt(BigN_ctl),'k.');
errorbar(1:1:6,nanmean(VV.ON.TST_EASY),nanstd(VV.ON.TST_EASY)./sqrt(BigN),'k.');
errorbar(1+.25:1:6+.25,nanmean(VV.OFF.TST_EASY),nanstd(VV.OFF.TST_EASY)./sqrt(BigN),'k.');
plot([0 7],[.5 .5],'b:')
set(gca,'xlim',[0 7],'xtick',[1:1:6],'xticklabel',{'AX','AY','AZ','WB','WC','WD'},'ytick',[0:.25:1]);
title('TST EASY Acc');

subplot(3,2,5)
hold on
bar(1-.25:1:4-.25,nanmean(VV.CTL.TST_MEDIUM),.25,'g');
bar(1:1:4,nanmean(VV.ON.TST_MEDIUM),.25,'w');
bar(1+.25:1:4+.25,nanmean(VV.OFF.TST_MEDIUM),.25,'r');
errorbar(1-.25:1:4-.25,nanmean(VV.CTL.TST_MEDIUM),nanstd(VV.CTL.TST_MEDIUM)./sqrt(BigN_ctl),'k.');
errorbar(1:1:4,nanmean(VV.ON.TST_MEDIUM),nanstd(VV.ON.TST_MEDIUM)./sqrt(BigN),'k.');
errorbar(1+.25:1:4+.25,nanmean(VV.OFF.TST_MEDIUM),nanstd(VV.OFF.TST_MEDIUM)./sqrt(BigN),'k.');
plot([0 5],[.5 .5],'b:')
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'CX','CZ','YB','YD'},'ytick',[0:.25:1]);
title('TST MEDIUM Acc');

subplot(3,2,6)
hold on
bar(1-.25:1:2-.25,nanmean(VV.CTL.TST_HARD),.25,'g');
bar(1:1:2,nanmean(VV.ON.TST_HARD),.25,'w');
bar(1+.25:1:2+.25,nanmean(VV.OFF.TST_HARD),.25,'r');
errorbar(1-.25:1:2-.25,nanmean(VV.CTL.TST_HARD),nanstd(VV.CTL.TST_HARD)./sqrt(BigN_ctl),'k.');
errorbar(1:1:2,nanmean(VV.ON.TST_HARD),nanstd(VV.ON.TST_HARD)./sqrt(BigN),'k.');
errorbar(1+.25:1:2+.25,nanmean(VV.OFF.TST_HARD),nanstd(VV.OFF.TST_HARD)./sqrt(BigN),'k.');
plot([0 3],[.5 .5],'b:')
set(gca,'xlim',[0 3],'xtick',[1:1:2],'xticklabel',{'ZB','DX'},'ytick',[0:.25:1]);
title('TST HARD Acc');




% ^^^^^^



figure; 
subplot(3,2,1:2)
hold on
bar(1-.3:1:8-.3,nanmean(VV.ON.TST_WITHINSET(Esx,:)),.15,'w');
bar(1-.1:1:8-.1,nanmean(VV.OFF.TST_WITHINSET(Esx,:)),.15,'r');
bar(1+.1:1:8+.1,nanmean(VV.ON.TST_WITHINSET(Lsx,:)),.15,'w');
bar(1+.3:1:8+.3,nanmean(VV.OFF.TST_WITHINSET(Lsx,:)),.15,'r');
errorbar(1-.3:1:8-.3,mean(VV.ON.TST_WITHINSET(Esx,:)),std(VV.ON.TST_WITHINSET(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1-.1:1:8-.1,mean(VV.OFF.TST_WITHINSET(Esx,:)),std(VV.OFF.TST_WITHINSET(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1+.1:1:8+.1,mean(VV.ON.TST_WITHINSET(Lsx,:)),std(VV.ON.TST_WITHINSET(Lsx,:))./sqrt(LateN),'k.');
errorbar(1+.3:1:8+.3,mean(VV.OFF.TST_WITHINSET(Lsx,:)),std(VV.OFF.TST_WITHINSET(Lsx,:))./sqrt(LateN),'k.');
plot([0 9],[.5 .5],'b:')
set(gca,'xlim',[0 9],'xtick',[1:1:8],'xticklabel',{'AC','AD','CB','BD','WY','WZ','YX','XZ'},'ytick',[0:.25:1]);
title('TST w/in set Acc');

subplot(3,2,3:4)
hold on
bar(1-.3:1:6-.3,nanmean(VV.ON.TST_EASY(Esx,:)),.15,'w');
bar(1-.1:1:6-.1,nanmean(VV.OFF.TST_EASY(Esx,:)),.15,'r');
bar(1+.1:1:6+.1,nanmean(VV.ON.TST_EASY(Lsx,:)),.15,'w');
bar(1+.3:1:6+.3,nanmean(VV.OFF.TST_EASY(Lsx,:)),.15,'r');
errorbar(1-.3:1:6-.3,mean(VV.ON.TST_EASY(Esx,:)),std(VV.ON.TST_EASY(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1-.1:1:6-.1,mean(VV.OFF.TST_EASY(Esx,:)),std(VV.OFF.TST_EASY(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1+.1:1:6+.1,mean(VV.ON.TST_EASY(Lsx,:)),std(VV.ON.TST_EASY(Lsx,:))./sqrt(LateN),'k.');
errorbar(1+.3:1:6+.3,mean(VV.OFF.TST_EASY(Lsx,:)),std(VV.OFF.TST_EASY(Lsx,:))./sqrt(LateN),'k.');
plot([0 7],[.5 .5],'b:')
set(gca,'xlim',[0 7],'xtick',[1:1:6],'xticklabel',{'AX','AY','AZ','WB','WC','WD'},'ytick',[0:.25:1]);
title('TST EASY Acc');

subplot(3,2,5)
hold on
bar(1-.3:1:4-.3,nanmean(VV.ON.TST_MEDIUM(Esx,:)),.15,'w');
bar(1-.1:1:4-.1,nanmean(VV.OFF.TST_MEDIUM(Esx,:)),.15,'r');
bar(1+.1:1:4+.1,nanmean(VV.ON.TST_MEDIUM(Lsx,:)),.15,'w');
bar(1+.3:1:4+.3,nanmean(VV.OFF.TST_MEDIUM(Lsx,:)),.15,'r');
errorbar(1-.3:1:4-.3,mean(VV.ON.TST_MEDIUM(Esx,:)),std(VV.ON.TST_MEDIUM(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1-.1:1:4-.1,mean(VV.OFF.TST_MEDIUM(Esx,:)),std(VV.OFF.TST_MEDIUM(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1+.1:1:4+.1,mean(VV.ON.TST_MEDIUM(Lsx,:)),std(VV.ON.TST_MEDIUM(Lsx,:))./sqrt(LateN),'k.');
errorbar(1+.3:1:4+.3,mean(VV.OFF.TST_MEDIUM(Lsx,:)),std(VV.OFF.TST_MEDIUM(Lsx,:))./sqrt(LateN),'k.');
plot([0 5],[.5 .5],'b:')
set(gca,'xlim',[0 5],'xtick',[1:1:4],'xticklabel',{'CX','CZ','YB','YD'},'ytick',[0:.25:1]);
title('TST MEDIUM Acc');

subplot(3,2,6)
hold on
bar(1-.3:1:2-.3,nanmean(VV.ON.TST_HARD(Esx,:)),.15,'w');
bar(1-.1:1:2-.1,nanmean(VV.OFF.TST_HARD(Esx,:)),.15,'r');
bar(1+.1:1:2+.1,nanmean(VV.ON.TST_HARD(Lsx,:)),.15,'w');
bar(1+.3:1:2+.3,nanmean(VV.OFF.TST_HARD(Lsx,:)),.15,'r');
errorbar(1-.3:1:2-.3,mean(VV.ON.TST_HARD(Esx,:)),std(VV.ON.TST_HARD(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1-.1:1:2-.1,mean(VV.OFF.TST_HARD(Esx,:)),std(VV.OFF.TST_HARD(Esx,:))./sqrt(EarlyN),'k.');
errorbar(1+.1:1:2+.1,mean(VV.ON.TST_HARD(Lsx,:)),std(VV.ON.TST_HARD(Lsx,:))./sqrt(LateN),'k.');
errorbar(1+.3:1:2+.3,mean(VV.OFF.TST_HARD(Lsx,:)),std(VV.OFF.TST_HARD(Lsx,:))./sqrt(LateN),'k.');
plot([0 3],[.5 .5],'b:')
set(gca,'xlim',[0 3],'xtick',[1:1:2],'xticklabel',{'ZB','DX'},'ytick',[0:.25:1]);
title('TST HARD Acc');






%%
% % % clc;
% % % disp([num2str(subno),'_sess',num2str(session),'_VVbeh'])
% % % disp(' ')
% % % disp('Accuracy: >.5 shows that they learned optimal choice')
% % % disp(['  choose: AB (90/10)','   match: WX (90/10)','   choose: CD (70/30)','   match: YZ (70/30)'])
% % % disp(['Test Acc: ',num2str(MEGA(row).TST_ACC)])
% % % disp(' ')
% % % disp('BIAS: >.5 is prefer Choose over Match  (may only happen for first 2)')
% % % disp(['  AW (90/90)','   CY (70/70)','   DZ (30/30)','   BX (30/30)'])
% % % disp(['Test BIAS: ',num2str(MEGA(row).TST_BIAS)])



%%