Hi Noam,

Higher thresholds are a good idea for other reasons, see the Woo et al paper (Neuroimage, 2014). So yes, 3.1 is probably a good idea.

All the best,

Anderson

On 21 November 2016 at 01:59, Noam Schneck <[log in to unmask]> wrote:

Hello,

What is an appropriate threshold for voxelwise minimum and cluster correction when using the easythresh_conj script? Is it necessary to set a threshold of voxelwise z=3.1 for a conjunction analysis, in accordance with Eklund et al 2016? Or is the old standard of 2.3 still reasonable for a conjunction analysis?

Thank you,
Noam