 |
 |
Dear SPM community, I want to model a visual working memory task with arrow (point left or right) presented for 200ms, memory array (0, 2 and 4 items) for 500ms, delay for 6000ms, retrieval array for 1500ms and ITI for 800-1400ms. This is a delayed-match-to-sample (DMS) task with 6 conditions (point left/right multiply load 2/4 + 2 no-item conditions). Some studies modeled each trial conditions with separate set of regressors and estimated the overall trial activity summed across WM phases (arrow, memory, delay, retrieval). But I want to distinguish each WM phases activity, how can I design the matrix? （1）I modeled 4 column of covariances for each trial type, for example, [trial type 1 arrow onset list, trial type 1 memory onset list, trial type 1 delay onset list, trial type 1 retrieval onset list, trial type 2 arrow onset list, and so on], and then convolved with a canonical hemodynamic function. Is it correct? （2）If above designed matrix is correct and I want to see the activated voxel for the retriecal phase, can I just set the contrast vector as 1 for the retrieval regressors and 0 for others? Thank you for your help! Ping