 |
 |
Dear McLaren, Sorry form the term, just a bad habit... Maybe I misunderstand the use of PM. In my mind, a regressor (such as movement parameters, or conditions modelized in 1st level analyses) modeled a part of the variance of the whole signal. But for PM, it does not work like this, right ? I mean, if I set the condition C1 and a PM of c1 (lets name it PMc1) and if I look at the effect of C1, it will not mean that it's an effect of C1 without the effect of PMc1, right ? Regards, Alexandre Le 25/02/2014 20:29, MCLAREN, Donald a écrit : > The term "regressor of non-interest" is only used by researchers, > there is absolutely no difference between regressors that you want to > study and those that you do not want to study. In SPM and any > statistical program, all regressors are treated the same way. > > The contrast amplitude should not change (although likely will change > slightly) as C1-C2 without the PM in the model is the mean of C1 minus > the mean of C2. If you add a PM, then C1 and C2 are still the mean C1 > and C2 as the mean value of the PM is set to 0 before convolving with > the task and forming the regressor. As both methods have the mean of > C1 and C2, the results will be similar. They won't be identical > because you are estimating a response and some variance of C2 might be > attributed to the PM of C1 and vice versa. > > Either approach is valid. > > > 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 > <http://www.martinos.org/%7Emclaren> > 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, Feb 25, 2014 at 4:32 AM, Alexandre Obert > <[log in to unmask] <mailto:[log in to unmask]>> wrote: > > Dear SPMers, > > I've got some questions about parametric modulations. > We conducted a study in which participants heard two types of > sentences (lets name them c1 and c2) and had to categorized them. > When we built these sentences (during a pretest), we obtained a > characteristic (such as degree of familiarity) and we would like > to see the effect of this characteristic on activations. > My first idea was to set it as a parametric modulation supposing a > linear increase of activations as this characteristic increase > (1st order, right ?). > However, I wonder if this parametric modulation is considered as a > regressor of non-interest in the design ? > Does it mean that if I set the contrast c1 /vs /c2 from the design > I set the PM, it will mean that the resulting activations will be > without the effect of PM ? > > Regards, > > Alexandre > >