For each subject in my dataset, I have a data matrix that contains control energy values for certain state-to-state-transitions. Rows denote the transition that was made, columns denote the nodes in my network. Now for each node, I would like to have an estimate of how strong it is associated with certain states. A very rough approach would be to calculate multiple t-tests within each node. For example, one could compute a t-test for all state-to-state-transitions that are associated with the 2back task (e.g. fear-2back, 2back-neut, rnd-2back, etc.) and test this group against all other state-to-state-transitions. This could be a rough estimate for how strong this node is associated with the 2back state. Then one would repeat this for all other states and finally, the whole procedure would also have to be done for all nodes (376 in my case). This however is a lot of t-tests and I wonder if there could be a more elegant way to do this?

enter image description here

  • $\begingroup$ Please clarify your question and the purpose of your investigation. Are all nodes considered the same, or is there any fundamental difference? Is there any way to simplify the transitions (e.g divide them to some groups or define each transition as a coupe of states)? Moreover the definition "how strong is a node associated with certain states" is unclear to me. $\endgroup$
    – Spätzle
    Oct 18 at 13:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.