giving RT-dependent feedback

Hi - In this experiment, I want to give feedback during an initial set of training trials; if the subject responds within a certain RT = correct, outside that range = incorrect. How do I use RT range as a parameter for feedback?

temp estimation.zip (21.8 KB)

This is easily done in the Event Editor’s Feedback tab by selecting the conditions “The reaction time is too slow” or “The reaction time is too fast”. After you select it, you can set what the RT is:

thanks, Hisham. A follow-up question: my subjects are meant to estimate the length of a presented tone. So they hear Tone 1 (of 7 lengths) and then during Tone 2 (which is very long) they hit SPACE when they believe that the length is same as Tone 1.
During training trials, they should get feedback as to wehether their estimate was too long or too short; so, the feedback is dependent on a previous (Tone 1) event.

Can I use a stim list or parameter set or something for RTs? Or do I need to create 7 different Tone 2 feedback trials (one for each Tone 1 length)? I don’t see a way of creating a parameter list for FEEDBACK.

any suggestions?

hi there- I really would be grateful for any suggestions on this feedback issue. Thanks!!

per request, posting task

Inside the task is the superlab program as well as associated stimulus files.

Temporal Estimation.zip (350 KB)

The experiment is doable in SuperLab 5. Two things need to be done:

  • Determine whether the response is correct using rules at the trial level instead of feedback at the event level

  • Create a parameter of type “Timer”, then start it and pause it. We will use this timer to measure RT.

In the attached experiment, I focused on running trial “TrainingTrial” only as a way of illustrating how this can be done. I created a parameter “RT timer”. This timer is being reset to zero in event “ITI-initial”, started in event “ToneMATCHtext”, and paused in event “estimate-feedback500”. Please note that these are the events used in trial “TrainingTrial”. You may need to start/pause it in other events as well.

The steps above help us capture and store the RT for event “estimate-feedback500” in a parameter. At the trial level, I have created two rules that evaluate how long it took the participant to respond to the 500ms and 1100ms tones.

You’ll need to create similar rules for the remaining five tones.

Other changes I’ve made:

  • Tagged the items in stimulus list “TwoTonesPractice”

  • Deleted parameter “ RTlength” since it was no longer used

  • I created two events that present “Correct” or “Incorrect” messages, though only the Correct one is being used in the attached experiment.

I hope this helps.

TemporalEstimation rev B.sl5 (16.8 KB)

One more thing: parameters of type “Timer” were added in SuperLab 5.0.3 (in case you are using an earlier version).

You can always obtain the most recent version of SuperLab 5 (currently at 5.0.5) from http://cedrus.com/superlab/update/v5.

Thank you so much, Hisham – your suggestions completely addressed the problem and I was able to program the experiment exactly as I wanted!! I really appreciate your help.