[Eeglablist] more std_stats questions

Hamish INNES-BROWN HINNES-BROWN at bionicsinstitute.org
Sun Jun 19 23:15:12 PDT 2011


Sorry for the onslaught of stats questions, I think I'm a bit more advanced than where I was thismonring, but a few problems remain.

So, I have ERPs from 2 conditions which I would like to test for differences and plot. 60 electrodes, 500 time points, and 12 subjects:

ERP_c1 = [60 x 500 x 12]
ERP_c2 = [60 x 500 x 12]

% put them together in the format I THINK is needed
erpdata = {ERP_c1 ERP_c2}


% Firstly calculate the stats:

paired = 'on' ; mode = 'param' ; naccu = 200;	%%%  0.01 = 200, .001 = 2000
[stats, df, pvals, surrog] = statcond( erpdata, 'paired', paired, 'mode', mode, 'naccu', naccu );

This seems to work OK and provides 
1) t-values in a [60 x 500] and 
2) p values in a [60 x 500] (later I will change this to permutation test).

Now for plotting:

locs = eeg_mergelocs(ALLEEG.chanlocs);
plotsubjects = 'off';
times = EEG(1).times;

std_plotcurve(times, erpdata, 'condstats', {pvals}, 'chanlocs', locs, 'plotsubjects', plotsubjects, ...
		'condnames', cnames, 'plottopo', 'off');



No matter how I try I can't seem to get the input sizes right for this command. I keeping errors such as:
??? Error using ==> plotcurve at 87
Size of time input and array input does not match

Error in ==> std_plotcurve at 371
                    plotcurve( allx, tmpdata, 'colors', tmpcol, plotopt{2:end},
                    'title', opt.titles{c,g});

Which suggests that something is wrong with the dimensions of the inputs.  As far as I can see all the relevant inputs are [60 x 500], apart from times, which is just [1 x 500]. If I change that to [60 x 500} (with the x values repeated 60 times), I get the same thing.





Hamish Innes-Brown
Research Assistant
The Bionics Institute of Australia
 
t: +61 3 9667 7529
f: +61 3 9667 7518
e: hinnes-brown at bionicsinstitute.org
w: http://www.bionicsinstitute.org/profiles/HamishInnes-Brown.html
w2: http://musicalbionics.wordpress.com/

2nd International Medical Bionics Conference, Phillip Island, 20-23 Nov 2011
http://www.iceaustralia.com/medicalbionics2011/ 






More information about the eeglablist mailing list