[MINC-users] significance question

[Jason Lerch]

Greetings,

there is no one threshold to use - and this applies to fMRI as well.  
You first have to chose the thresholding strategy that you would like  
to use, then compute the appropriate threshold for the degrees of  
freedom and number of voxels in your study. The most commonly used  
method is Keith's random field theory - you can find more information  
about it by starting matlab on one of the SGIs and typing 'help  
stat_threshold'

Cheers,

Jason

On 5-Jan-06, at 10:40 AM, Anil Ashok Kuchinad, Mr wrote:

> Hey all,
>
>
>
> I was wondering what value people use as a threshold for statistical
> significance in a VBM analysis without any a priori hypotheses. I have
> heard a t-value of 4.2 being tossed around, but I was led to believe
> that is for fMRI.
>
>
>
> Cheers,
>
> Anil
>
>
>
> ----------------------------------------------------
>
> Anil Kuchinad
>
> Centre for Research on Pain
>
> McGill University
>
> Strathcona Anatomy and Dentistry Building
>
> 3640 University St., Room M/19
>
> Montreal, Quebec , Canada
>
> H3A 2B2
>
> Tel: (514) 398-1271
>
> Email: anil.kuchinad at mcgill.ca
>
>
>
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users