RTs to Facial Morphs

[jarvis24]

[edited from support email]

I'm trying to do a specific task in Medialab and I'm having problems. A fellow student in my lab, Pontus Leander, has told me you have been very helpful in getting things working, so I'd like to see if you could help me. What I need to do is run a face-morphing file. The task that subjects
will perform is stopping the morph at the point they see the expression change, and what I need to measure is the time at which they do so.

The morphs have been created with a program called Popims Animator (available for download at http://www.popims.com/en/animator.htm), and they have a file extension of .pim. I will need the morph to stop when the subject presses the response key. This is a paradigm that was done years ago on a program called Morph 2.5 that is now so convoluted and ancient it's hard to decipher. I (and probably some other researchers) would love to be able to do this in Medialab.

I appreciate your help.

[jarvis24]

Let me ask you two questions--first--can you save the pim files as avi's or animated gif's? Second, does Popims give you an easy way to embed the .pim files in a web/html page?
-Blair

[jarvis24]

[edited from followup support email]

Thanks for your help. We have successfully saved the files in .avi format, and they play in Medialab as a movie item type. We haven't figured out how to do exactly what we need, however. What we would like to do is allow subjects to stop the movie. They'll be seeing a face morph from one emotion to another, and our DV is how long it takes them to see a change in emotion. We were hoping to record an RT for when they stop the movie. Is that possible? My colleagues only have Medialab at this point - could we do what we need with DirectRT?

[jarvis24]

If you include it as a backVideo on an instruction item, then they can press space to continue and you can get the RT. Will that do it?

[jarvis24]

[edited from followup support email]

It looks like that will be perfect. I knew there must be a simple way to do this. Thanks for your time and attention.

[bhatch83]

This is a nice way to do this. I am setting up a paradigm much like this. I need the video to pause and the image to stay on the screen while participants respond to multiple responses. Any suggestions?

[jarvis24]

One way you could approach this is to embed the video with an html page and then use that HTML page as a defined region in the background. To give you the pause and re-start control, you would just need to embed the video in a way that gives you this control. I expect there is a lot of sample HTML code out there to guide you in doing this--would love to see a sample posted if you can work it out. You could also run a search on the forums here on the term "embed html"--you might turn up something useful. Finally, another quick way to do this might be to upload your video to Youtube or other similar resource for playing videos. Then you could just use the Youtube URL as the name of the HTML background region. The trick would be having only displayed what you want and not all the surrounding links and other distracting stuff. Maybe another streaming video host offers a plainer option yet still offers the user the video control you are looking for.

[bhatch83]

Thank you, I am not sure if this will give me the rt data i need. I will clarify experiment for you. It is as follows. I would like a video to be presented on the screen. I would then like participants to press a button (e.g., spacebar) when they think they recognize what the video shows. I would like the reaction time for this to be recorded. I would then like the stopped video image to stay on the screen and have a multiple choice item be provided below. I am also considering using Direct RT.

Thanks :-)

[jarvis24]

bhatch83--This sounds very much like a DirectRT task with exception that you can't freeze on the final frame. For that you would need some kind of HTML in which to embed a video player or a custom item in MediaLab that would allow you to do it this way. Everything but keeping the frame on screen at the point of the RT would be most easily and (perhaps more importantly for your purpose) more accurately in terms of display and RT timing.