 |
 |
Hi, thank you both. I am analyzing a reinforcement learning task in which participants are presented with a pair of stimuli and required to choose one. The choice is followed by a probabilistic feedback (loss, reward or neutral, given the specific condition). The EV I am looking at is the estimated prediction error used as parametric modulator of the onset of the feedback. I am using an ROI of the ventral striatum that was independently built for a previous study. What I usually do is to first look at that EV whole brain level, than I select the small volume and only look at the ROI. At the whole brain level statistics look "normal", and a striatal activation is already evident. This problem only arises when using the small volume option, and it happened also using the same ROI with other contrasts. I am not sure if these are exactly the kind of things you needed to know, sorry, this is my first fMRI analysis ever. Are there any other informations that I should give? Many thanks! Sara - - Sara Garofalo, MSc, PhD Student in Cognitive Neuroscience Department of Psychiatry, Cambridge University Forvie Site, Robinson Way Cambridge CB2 0SR (UK) Tel: 01223 336956 (during office hours) Mobile: +44 7475897028 email: [log in to unmask]; [log in to unmask]; [log in to unmask] Profile on ResearchGate <https://www.researchgate.net/profile/Sara_Garofalo> or Academia <https://cambridge.academia.edu/SaraGarofalo> On 9 June 2015 at 14:13, MCLAREN, Donald <[log in to unmask]> wrote: > Marko, > > You are right -- if the correct cluster P-value is larger than the > uncorrected cluster P-value, then this is an issue and shouldn't be > happening. I misinterpreted the email as the next line asks about peak > values. > > > On Tue, Jun 9, 2015 at 9:08 AM, Marko Wilke < > [log in to unmask]> wrote: > >> Hi, >> >> but wasn't the question whether uncorrected *cluster-level* p-values can >> be higher than corrected cluster-level ones? I think the culprit here may >> be in the ROI/SVC approach, which may mess with the smoothness estimation, >> depending on how it is done. Perhaps you, Sara, could provide some more >> details about how you set up your analyses? >> >> Cheers >> Marko >> >> MCLAREN, Donald wrote: >> >>> Yes. The cluster statistics are always weaker than the individual voxel >>> significance. I prefer to look at the cluster significance as I don't >>> believe that the brain would only have differential >>> activity/connectivity/volume/etc. in a single isolated voxel. >>> >>> Best Regards, Donald McLaren >>> ================= >>> D.G. McLaren, Ph.D. >>> Research Fellow, Department of Neurology, Massachusetts General Hospital >>> and >>> Harvard Medical School >>> Postdoctoral Research Fellow, GRECC, Bedford VA >>> Website: http://www.martinos.org/~mclaren >>> Office: (773) 406-2464 >>> ===================== >>> This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED >>> HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is >>> intended only for the use of the individual or entity named above. If the >>> reader of the e-mail is not the intended recipient or the employee or >>> agent >>> responsible for delivering it to the intended recipient, you are hereby >>> notified that you are in possession of confidential and privileged >>> information. Any unauthorized use, disclosure, copying or the taking of >>> any >>> action in reliance on the contents of this information is strictly >>> prohibited and may be unlawful. If you have received this e-mail >>> unintentionally, please immediately notify the sender via telephone at >>> (773) >>> 406-2464 or email. >>> >>> On Tue, Jun 9, 2015 at 7:03 AM, Sara Garofalo <[log in to unmask] >>> <mailto:[log in to unmask]>> wrote: >>> >>> Hi all, >>> >>> I have some (supposedly) strange results regarding an ROI analysis >>> using small volume correction. When looking at cluster-level >>> statistics the _uncorrected_ p-value is _not significant_ (e.g., >>> p=.12), while the FEW _corrected_ p-value is _significant_ (e.g., >>> p=.001). Does that make any sense? >>> >>> Is it more correct to only look at peak-level statistics? >>> >>> Many many thanks, >>> >>> Sara >>> >>> >>> >> -- >> ____________________________________________________ >> PD Dr. med. Marko Wilke >> Facharzt für Kinder- und Jugendmedizin >> Leiter, Experimentelle Pädiatrische Neurobildgebung >> Universitäts-Kinderklinik >> Abt. III (Neuropädiatrie) >> >> Marko Wilke, MD, PhD >> Pediatrician >> Head, Experimental Pediatric Neuroimaging >> University Children's Hospital >> Dept. III (Pediatric Neurology) >> >> Hoppe-Seyler-Str. 1 >> D - 72076 Tübingen, Germany >> Tel. +49 7071 29-83416 >> Fax +49 7071 29-5473 >> [log in to unmask] >> >> http://www.medizin.uni-tuebingen.de/kinder/epn/ >> ____________________________________________________ >> >> >