 |
 |
Dear SPMers, We are trying to analyse a fMRI data set in SPM5 and have a few queries. The data is from two groups of subjects (n=15 in each group) who performed a task in a block design with two alternating conditions - a 'Task condition' and a 'baseline condition'. Single subject level analysis was performed to derive the Task-Baseline contrast images and entered into a one sample t-test with covariates for each of the two groups to generate SPMs which were FDR corrected. We find regions of activations which appear to clearly differ in the two groups. However we fail to see any significant differences in a two-sample t-test in which the same Task-Baseline contrast images were entered for between-group comparisons. How would one explain such a scenario where the regional activation patterns as seen in the within group one-sample t-tests are not reflected in the between group t-test comparisons? We would be grateful for any comments / suggestions. With regards Harsha ------------------------ Harsha Halahalli Research Associate Department of Psychiatry National Institute of Mental Health and Neurosciences (NIMHANS) Bangalore, INDIA