Search the Legacy Support Archive (up to v2020)
  1. Read-only archive (2005–2024). Posting is closed. Need help with MediaLab or DirectRT 2027? Email service@empirisoft.com.

    All words · “exact phrase” · -leave out

    Advanced search

randomize bgr and jump wgr

  1. #1

    randomize bgr and jump wgr

    dear helping-team

    I want to do the following:

    In a recognition task I ask my participants: “Do you recognize this sentence: XXX ?” – they choose yes or no. If they choose no I jump to the next sentence asking: “Do you recognize this sentence: XXY ?” If they coose yes, I ask two further questions, lets say qu_A and qu_B, and so on.
    The sentences XXX, XXY, .. should be presented randomly, but I need to know afterwards in which order they appeared. And of course I need every response. Therefor I chose randomization via bgr (and not via $recog.txt as stim).

    here is my problem:
    I used the jump-command, (e.g. RT:1>26,9>28 – 1 woud be yes and trial 26 the further question qu_A, 9 would be no and trial 28 the next sentence XXYZ) to react to the yes-or-no response of a participant. Now, when randomly the last group was selected by directRT, my jump-command causes to jump to the next trial outside the randomized groups without that every sentence/group was presented.

    Can you help me?

    Thanks in advance,
    Julia

  2. #2
    jarvis24 Blair Jarvis, Administrator
    Not sure I totally understand--could you post your input file so I can take a look? Preferably a reduced or simplified version that has just what it takes to illustrate the problem.

  3. #3
    dear jarvis, here is a part of my input file.
    I want to randomize between the groups. Though INSIDE a group the first question determines if the other two questions should follow or if the next group should be started.
    In this solution, if directRT picks the last group this ends all the questions even if not each group was presented yet.
    If you see another approach to get what I want I would be happy as well.

  4. #4
    now I put all 3 questions in one trial and skip mid-trial to the next if the response of the participant determines that the following questions should not be presented.
    So I can randomize via $ and not via BGR.

    This should work.
    But thanks for thinking.

  5. #5
    jarvis24 Blair Jarvis, Administrator
    Glad you found a solution that worked, thanks for the update Julia.