-
Notifications
You must be signed in to change notification settings - Fork 6
/
seq_feargen_information.m
174 lines (171 loc) · 5.5 KB
/
seq_feargen_information.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
function [out]=seq_feargen_information(s,varargin)
%[out]=seq_feargen_information(s,varargin)
%
% Will spit out interesting information on the sequences S. Interesting
% stuff are also shown as a figure. Assumes the last and last -1 are UCS
% and oddballs, respectively. OUT contains all the information. Use
% VARARGIN flag to turn off visualizations or verbose. [0 0]
% no figure, no verbose, [1 0], only figure, [0 1] only vervose. Default
% [1 1].
%
%
verbose = 1;
fig = 1;
if nargin > 1
fig = varargin{1}(1);
verbose = varargin{1}(2);
end
%%set some constants
nummods = 10;%number of time bin in the FIR matrix
odd_id = 9;
ucs_id = 10;
null_id = 0;
%%
s = s(:);
tcond = length(unique(s));%with the null event
tstimcond = max(unique(s));%stimulus showing conditions i.e. without the null
conds = 0:tstimcond;
out.ttrial = length(s);
if verbose
fprintf('\n\n\n\n\n');
fprintf('Total trial: %g\n',out.ttrial);
fprintf('%g and %g are considered as UCS and oddball, respectively\n',ucs_id,odd_id);
fprintf('Probability of different conditions:\n');
end
count = zeros(1,tstimcond);
for c = 1:tcond
count(c) = sum(s == conds(c));
if verbose
fprintf('Cond %g: %g (%g)\n',conds(c),count(c),count(c)./out.ttrial);
end
end
ucs = sum(s == ucs_id);
csp = sum(s == 2);
out.rr = ucs./(ucs+csp)*100;
odd = sum(s == odd_id);
out.or = odd./out.ttrial*100;
if verbose
fprintf('=====================================\n');
fprintf('RRei: %g percent.\n',out.rr);
fprintf('=====================================\n');
fprintf('ROdd: %g percent.\n',out.or);
fprintf('=====================================\n');
fprintf('Transition Check\n');
end
n = hist3([s(1:end-1) s(2:end)],{conds conds});
[dummy_eff dummy_det] = calc_meffdet(s, 10 , tstimcond, 3);
[~,~,max_det,max_eff] = tcurve(tstimcond,10,length(s));
tmaxeff = out.ttrial/(2*(tcond)*10);
eff_norm = dummy_eff(1)./max_eff(1);
if verbose
fprintf('=====================================\n');
fprintf('Efficiency: %g\n',dummy_eff(1));
fprintf('Max efficiency: %g\n',tmaxeff);
fprintf('Normalized efficiency: %g percent\n',eff_norm(1)*100 );
fprintf('Power: %g\n',dummy_det(1));
fprintf('Normalized power: %g\n',dummy_det(1)/max_det(1)*100);
fprintf('=====================================\n');
for isis = [1 2.5 3 6]
fprintf('Expected duration: ISI of %g s -> %g minutes..\n',isis,isis*out.ttrial/60);
end
end
%% get the FIR matrix and efficiencies
[X,CX] = seq_seq2fir(s,nummods);
[effmat] = calc_meffdet(s,nummods,tstimcond,eye(out.ttrial));
[~,~,~,tmaxeff] = tcurve(tstimcond,nummods,out.ttrial,0);
[~, v] = eig(CX(1:(tstimcond-2)*nummods,1:(tstimcond-2)*nummods));
%% compute also the entropies
out.ent_order = 1:5;
for order = 1:5
[out.ent(order),out.ent0,out.entmax] = calcent(s,order);
end
%% viz stuff.
if fig
figure(1000);
clf;
nr = 3;
nc = 4;
%
subplot(nr,nc,1:3);
plot(s,'.-','markersize',10);hold on;
plot(find(s == tstimcond-1),s(s == tstimcond-1),'+r','markersize',10);
plot(find(s == tstimcond),s(s == tstimcond),'sg','markersize',10);
xlabel('trials')
set(gca,'ytick',0:tstimcond,'yticklabel',{'N' '1' '2' '3' '4' '5' '6' '7' '8' 'UCS' 'ODD'})
grid on;
title(sprintf('the sequence (length: %g, duration: %3.3g m (ISI:3s))',out.ttrial,3*out.ttrial/60))
box off
hold off;
ylim([0 tstimcond+1])
%
subplot(nr,nc,4);
imagesc(n);
thincolorbar('vertical');
title('2nd order Transition Matrix');
set(gca,'ytick',1:tstimcond+1,'yticklabel',{'N' '1' '2' '3' '4' '5' '6' '7' '8' 'UCS' 'ODD'})
set(gca,'xtick',1:tstimcond+1,'xticklabel',{'N' '1' '2' '3' '4' '5' '6' '7' '8' 'U' 'O'})
%
subplot(nr,nc,5)
imagesc(X);colormap('jet');
title('FIR matrix')
set(gca,'xtick',[],'ytick',[])
xlabel('conditions');ylabel('trials')
%
subplot(nr,nc,6)
bar([1./mean(effmat(2:tstimcond)) ;1./effmat(2:tstimcond+1)],'k')
hold on
plot(xlim,[tmaxeff tmaxeff],'r--')
ylim([0 tmaxeff]*1.5);
xlim([0 tstimcond+1+1])
box off
xlabel('conditions');
ylabel('efficiency');
set(gca,'xticklabel',{'All' '1' '2' '3' '4' '5' '6' '7' '8' 'U' 'O'})
title(sprintf('Eff. of estimation(%2.4g %% ) ',1./mean(effmat(2:tstimcond))/tmaxeff*100))
hold off
%
subplot(nr,nc,7)
bar(1./effmat(tstimcond+2:end),'k')
hold on;
plot(xlim,[tmaxeff tmaxeff],'r--');
hold off
title('Eff. of contrasts')
axis tight;
ylim(ylim*1.5);
xlim([0 length(effmat)-tstimcond] );
box off;
%
subplot(nr,nc,8)
bar([0 out.ent_order]+1, [ out.ent0 out.ent],'k')
hold on
xlabel('n^{th} order')
ylabel('entropy')
box off
axis tight;
plot(xlim,[out.entmax out.entmax],'r--')
ylim(ylim*1.2);
title('entropy')
grid on;
%
subplot(nr,nc,9)
imagesc(cov(X));
thincolorbar('vertical');
title(sprintf('covariance (rank: %g (%g))',rank(cov(X)),size(X,2)));axis off;
%
subplot(nr,nc,10)
plot(diag(v),'ok-')
title('Eigenvalue spectrum')
box off;
%
subplot(nr,nc,11)
hist(diff(find(s ~= 0)));
title('Distribution of ISIs')
box off;
end
%% create output
out.eff_overall = 1./mean(effmat(2:tstimcond));
out.eff_estimate = 1./effmat(2:tstimcond+1);
out.eff_contrasts = 1./effmat(tstimcond+2:end);
out.eff_max = tmaxeff;
out.eff_norm = out.eff_overall./out.eff_max*100;
out.probability = count./out.ttrial;