About the option of computing PLV across trials, keeping time information

Discussions
1

Hi all,

I have 200 trial source data files, and want to computing PLV across trials, which keep time information.
I tried the computing PLV under the condition (1×N, Time window (1000ms), select one scout(22 vertices), scout function (Mean,after), check the ‘keep time information…’). But I could not do that. Then error report was presented as described blow. How should I do that?

Thank you in advance,
Best,
Jun

Line 460: Out of memory.
Please unload some datasets (or restart Matlab) an try again.

Call stack:
>bst_connectivity.m at 460
>process_plv1.m>Run at 93
>macro_methodcall.m at 30
>process_plv1.m at 24
>bst_process.m>Run at 196
>bst_call.m at 26
>macro_methodcall.m at 37
>bst_process.m at 31
>panel_process1.m>RunProcess at 125
>bst_call.m at 28
>macro_methodcall.m at 39
>panel_process1.m at 27
>gui_brainstorm.m>CreateWindow/ProcessRun_Callback at 686 >bst_call.m at 28 >gui_brainstorm.m>@(h,ev)bst_call(@ProcessRun_Callback) at 240

2

Hi Jun,

What you are running needs a lot of memory. There is no solution except for decreasing the amount of memory (less vertices on the cortex surface, less time samples, …) or using a computer with more memory.

Note that the PLV across trials is a very noisy measure. This should be done with an integration over short time windows instead of time point by time point, as done in [Lauchaux, 1999]: https://www.ma.utexas.edu/users/davis/reu/ch3/cwt/lachaux.pdf
There is no implementation of this in Brainstorm yet.

Francois