Auto-advance and feedback for keypress

[[email protected]]

Dear FF team,
For the keypress type, could it be auto-advanced? That is, whenever participants make a response, the current trial will end and advance to the next trial?
A second question, could correct responses/targets be set for keypress type so participants can get a feedback about right or wrong after each trial?
Thanks!
Haoyun

[Ting]

Hi Haoyun,

Let me give you the answers first and then some explanations (which you should feel free to skip over).

For the first question, yes, only if (1) you set the "multiple" parameter to false like below, and (2) there are no other responses on the same trial.

Code:

"multiple": false

For the second question, what you described is a bit unusual. Correct vs incorrect only makes sense in the context of a forced-choice task, and the keypress response isn't really meant to achieve that. If you are trying to avoid the buttons rendered by the choice response (https://help.findingfive.com/api/responses.html#choice), there's a "keyboard_only" mode which lets participants respond by pressing keys instead. With the choice response, you can easily record the targets and provide feedback to the participants, as they are standard features. Would that be useful for implementing your experiment?

--------- A bit explanation on the keypress response ------------

The keypress response is first and foremost a background response, meaning that its expected behavior is to collect participant responses passively. The prototypical use case is to capture the random keystrokes produced by participants as a metric of "impatience". So, for your first question, using the keypress to advance trials works under a limited situation, where the trial "has nothing else to do" (i.e., its whole purpose is just waiting for a single key press). For your second question, what you want the keypress response to do somewhat violates the design purpose (which isn't necessarily a bad thing), because the fact that a keypress has collected data, in general, should not affect the normal progression of trials, letting alone providing feedback.

Hope this is helpful! Happy to discuss further.

[[email protected]]

Hi, Ting,

Thanks for your quick reply. A choice type might work. What I basically want is participants see a word, and they make a button press as soon as possible. Is there a way I can omit the choices showing up on the screen? I want participants to remember which keys to press. Thanks!

[Ting]

Ah I see. Unfortunately right now there's no way to hide the choices, since it's a somewhat unusual design. Sorry!

[[email protected]]

No problem. Thanks for the reply.