<div dir="ltr">if you change noise scaling you should change the SNR as it means you have<div>a lot worse SNR. the regularization parameter lambda is a function of the snr</div><div>as you can see in the script.</div><div><br></div><div>dSPM and sLORETA are like t-stat or f-stat. So if you multiply the cov by 1e6</div><div>then you will have dSPM and sLORETA solutions which are just divided by 1e3</div><div>(the scale of the standard deviation).</div><div><br></div><div>HTH</div><div>Alex</div><div><br></div><div><br></div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Feb 14, 2018 at 8:33 PM, Evgenii Kalenkovich <span dir="ltr"><<a href="mailto:e.kalenkovich@gmail.com" target="_blank">e.kalenkovich@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi all,<div><br></div><div>I noticed that if I scale the noise covariance matrix, it does not change the source estimates in any way. In <a href="https://martinos.org/mne/stable/auto_tutorials/plot_mne_dspm_source_localization.html#sphx-glr-auto-tutorials-plot-mne-dspm-source-localization-py" target="_blank">this example</a> from the example gallery if I change the method to "MNE" before the first stc calculation and then do this:</div><div><br></div><div><pre style="font-family:Consolas;font-size:10.8pt">inverse_operator = make_inverse_operator(info, fwd, noise_cov,<br> <span style="color:rgb(102,0,153)">loose</span>=<span style="color:rgb(0,0,255)">0.2</span>, <span style="color:rgb(102,0,153)">depth</span>=<span style="color:rgb(0,0,255)">0.8</span>)<br>stc = apply_inverse(evoked, inverse_operator, lambda2,<br> <span style="color:rgb(102,0,153)">method</span>=method, <span style="color:rgb(102,0,153)">pick_ori</span>=<span style="color:rgb(0,0,128);font-weight:bold">None</span>)</pre><pre style="font-family:Consolas;font-size:10.8pt"><span style="color:rgb(0,0,128);font-weight:bold">from </span>copy <span style="color:rgb(0,0,128);font-weight:bold">import </span>deepcopy<br>noise_cov_scaled = deepcopy(noise_cov)<br>noise_cov_scaled[<span style="color:rgb(0,128,128);font-weight:bold">'data'</span>] *= <span style="color:rgb(0,0,255)">1000000<br></span>inverse_operator_scaled = make_inverse_operator(info, fwd, noise_cov_scaled,<br> <span style="color:rgb(102,0,153)">loose</span>=<span style="color:rgb(0,0,255)">0.2</span>, <span style="color:rgb(102,0,153)">depth</span>=<span style="color:rgb(0,0,255)">0.8</span>)<br>stc_scaled = apply_inverse(evoked, inverse_operator_scaled, lambda2,<br> <span style="color:rgb(102,0,153)">method</span>=method, <span style="color:rgb(102,0,153)">pick_ori</span>=<span style="color:rgb(0,0,128);font-weight:bold">None</span>)</pre></div><div><br></div><div>Then stc_scaled contains exactly the same data as stc. With "dSMP" and "sLORETA" the result do differ. Why doesn't "MNE" care about the scale of the covariance matrix?<span class="HOEnZb"><font color="#888888"><br></font></span></div><span class="HOEnZb"><font color="#888888"><div><br></div><div>Evgenii</div></font></span></div>
<br>______________________________<wbr>_________________<br>
Mne_analysis mailing list<br>
<a href="mailto:Mne_analysis@nmr.mgh.harvard.edu">Mne_analysis@nmr.mgh.harvard.<wbr>edu</a><br>
<a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis" rel="noreferrer" target="_blank">https://mail.nmr.mgh.harvard.<wbr>edu/mailman/listinfo/mne_<wbr>analysis</a><br>
<br>
<br>
The information in this e-mail is intended only for the person to whom it is<br>
addressed. If you believe this e-mail was sent to you in error and the e-mail<br>
contains patient information, please contact the Partners Compliance HelpLine at<br>
<a href="http://www.partners.org/complianceline" rel="noreferrer" target="_blank">http://www.partners.org/<wbr>complianceline</a> . If the e-mail was sent to you in error<br>
but does not contain patient information, please contact the sender and properly<br>
dispose of the e-mail.<br>
<br></blockquote></div><br></div>