gamification of paired-associates learning task

46 views
Skip to first unread message

Levan Bokeria

unread,
Jul 10, 2020, 9:59:36 AM7/10/20
to Online Experiments
Hello,

I have a paired-associates task using 16 different birds and 3 Christmas toy stimuli. 3 of the birds are associated with the Christmas toys, and the rest are not. The birds are very similar, so they are perceptually tough to discriminate.

I use a trial-and-error learning approach, where participants just guess the correct association, get feedback, and move to the next trial. I need the participants to learn the associations very well, and piloting shows that people get bored very quickly or sometimes even loose temper.

What general strategies would you recommend to make the task more engaging? To "gamify" it in some way? 

Some preliminary thoughts:
  • Playing positive/negative sounds during feedback.
  • Keep an ongoing performance score for each paired-associated displayed on the screen.
  • Encouraging remarks during the breaks.

Thank you!

Levan Bokeria

Josh de Leeuw

unread,
Jul 10, 2020, 10:57:34 AM7/10/20
to Levan Bokeria, Online Experiments
Hi Levan,

Is the single stimulus trial-and-error approach crucial for the design of the experiment? One thing that might be more engaging is to present an array of stimuli and have participants guess which bird has the toy. This might feel a little bit more like a puzzle to solve and less like random guessing.

Josh

--
You received this message because you are subscribed to the Google Groups "Online Experiments" group.
To unsubscribe from this group and stop receiving emails from it, send an email to online-experime...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/online-experiments/f75a73db-fbf7-4f97-81b2-bdcf8d0637afn%40googlegroups.com.

Levan Bokeria

unread,
Jul 10, 2020, 11:11:44 AM7/10/20
to Josh de Leeuw, Online Experiments
Thank you Josh!
We'd like to avoid showing all the birds on the screen at the same time. Right now, I show a toy and then 2 birds, and ask the participants which bird is associated with the toy. Feedback tells them if they were correct, and what toy the other bird was associated with as well. 

I guess we could show them 3 birds at a time. 

Best,
Levan

Becky Gilbert

unread,
Jul 15, 2020, 12:43:31 PM7/15/20
to Online Experiments
Bumping this up! 
I know this post is asking about a specific task, but I think a lot of researchers would benefit from advice and examples of how to make a standard/boring psych task into something more engaging and fun. I hoping that others will chime in with some suggestions. 
And here are two websites with examples of gamified tasks:

Joshua Hartshorne

unread,
Jul 16, 2020, 1:23:55 PM7/16/20
to Online Experiments
Hi Levan,

If people are saying the task is too boring, you probably want to think of a different task that gets at the same thing you want.

I like the idea of thinking about how to make the task a puzzle-solving task. Could you make the birds a clue to something (like where the toy is going to show up)? So I'm thinking more of an SRT task, where the goal is to press a key indicating which toy it is. But you see the bird in advance. Maybe that's not the right kind of learning for you, but you get the idea. 

A surprisingly effective method is to play a lot of video games and see if you can find any that involve the behavior you are interested in. And then you get to play video games and claim you are working!

Joshua K Hartshorne
Assistant Professor
Boston College

Levan Bokeria

unread,
Jul 22, 2020, 6:21:35 AM7/22/20
to Joshua Hartshorne, Online Experiments
Thank you Becky and Joshua!

I see the tasks in the Games with Words and the Music Lab involve telling the participants something about themselves. That seems to be a good intrinsic motivation. I'll try to think of something along those lines.
-----
PhD Student
MRC Cognition and Brain Sciences Unit
University of Cambridge
(+44)-7-943669454


--
You received this message because you are subscribed to a topic in the Google Groups "Online Experiments" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/online-experiments/z8G38n94AIo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to online-experime...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/online-experiments/405dfedc-b0c9-417b-9753-3c526ffb449do%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages