[Eeglablist] topoplot methods

[Arnaud Delorme]

Hi Josh,

for (1), Stephen already answered it.

for (2), you can build a STUDY and plot ERP, spectrum and ERSP topography, then plot the p-values topography of a 2x2 design as well. I recommend reading the STUDY tutorial. It is pretty easy. If you have a single subject and 4 datasets in your 2x2 design, you can plot ERP scalp topographies and compute statistics based on single trial activity. If you have 4 x 20 datasets (20 subjects), then you can perform statistics across subjects. Note that you can also define a STUDY with a single dataset that would contain 4 types of data epochs for example.

http://sccn.ucsd.edu/wiki/Chapter_02:_STUDY_Creation

Best,

Arno

On 2 Feb 2013, at 16:30, Stephen Politzer-Ahles wrote:

> Hi Josh,
> 
> As for (1), setting the 'shading' parameter to 'interp' makes the plot
> look less pixellated, if that's what you're looking for. Here's an
> example I made: http://i.imgur.com/fMgeXTt.png .
> 
> As for (2), there might be a simpler way to do this... but as a
> nonelegant way, you could just calculate a matrix representing the
> interaction term (in channels*timepoints) and use that as input to
> topoplot(). I haven't tried this by myself, but for instance if your
> conditions at Ia, Ib, IIa, and Ib (and each of those is a
> channels*timepoints matrix), then I guess something like this should
> work:
> 
> ---
> 
> % some simple effects
> effect_of_a_vs_b = Ib - Ia;
> effect_of_I_vs_II = IIa - Ia;
> 
> % expected value of IIb if there were no interaction
> expected = Ia + effect_of_a_vs_b + effect_of_l_vs_ll
> 
> % Interaction
> interaction = IIb - expected
> 
> % vector of values to plot
> range = mean( interaction(:,
> beginning_of_timewindow_of_interest:end_of_timewindow_of_interest),
> 2);
> 
> % plot it (assuming 'locations' has already been loaded)
> topoplot(range,locations, ..... );
> 
> ---
> 
> The matrix "interaction" represents how much bigger (or smaller) the
> signal is in condition IIb than you would expect if the two effects
> were purely additive. But like I said, there might be a more
> straightforward (and/or statistically smarter) way to do this.
> 
> Best,
> Steve
> 
> 
> 
> On Fri, Feb 1, 2013 at 9:15 PM, Joshua Hartshorne
> <jkhartshorne at gmail.com> wrote:
>> 
>> Two questions about topoplot():
>> 
>> 1) Can I change the smoothness/granularity of the topoplot?
>> 
>> 2) Can I plot the topographic distribution of a 2x2 interaction? I can get the pvalues plotted topographically, but not the voltages.
>> 
>> Many thanks,
>> Josh Hartshorne
>> 
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> 
> --
> Stephen Politzer-Ahles
> University of Kansas
> Linguistics Department
> http://people.ku.edu/~sjpa/
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu