Timeline for What statistics can be used to analyze and understand measured outcomes of choices in binary trees?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 28, 2021 at 12:10 | history | bounty ended | CommunityBot | ||
| Apr 27, 2021 at 14:08 | comment | added | Henrik | One problem is that the meaning of a parameter always needs to be fixed across trees. So if you have a parameter $t_A$ that encodes the (conditional) probability to choose A over B, it always needs to mean choosing A over B (and cannot mean choosing A over F in another tree). Thus, this parameter can only appear in a tree that offers this choice. And if all stimuli always have the same background stimulus, this is of course possible (as this is in some sense always the case). I hope this answers your questions. | |
| Apr 27, 2021 at 10:21 | comment | added | S Pr | Thank you very much, especially for the paper detailing TreeBUGS that I'll have to devote some time to. And apologies for my ignorance, I hope I can clarify the answer precisely for myself: do you mean that some stimuli must be common between different sub-trees? For example, if one tree fork includes A&B, and another forks C&D, the modeling can only be performed when e.g. A and C are the same? It cannot proceed without this commonality between the sub-trees, correct? And separately, would something like MPT models work if all stimuli (A, B, C and D) shared a common background stimulus? | |
| Apr 26, 2021 at 16:01 | history | answered | Henrik | CC BY-SA 4.0 |