Hi Ian,

I have a little experience in this area, and while Python indeed does fit
the bill for cross-platform it does have its own set of drawbacks:

-Time: You mentioned that you're recording response time by key presses: If
you're doing a PhD you'll likely need very accurate information, which you
most likely won't get from Python. There are various timing issues and
inconsistencies in the language (google "Python GIL" or "garbage
collection" for various specifics). Although you probably *can* make it
work, and do lots of tests to prove that your collected data is accurate
and verifyable, perhaps its not the easy route.

There are various domain specific languages set up for this kind of thing,
you mentioned LiveCode : I'm not sure it has the timing critical elements
that you'll find are most likely necessary for your project.

I would advise you to look at languages / enviroments that are designed for
writing psychology experiments:
I'm an open source guy, so will advise starting here :)
http://pebl.sourceforge.net

Perhaps the best "next step" is to get some details on how timing-accurate
your data will need to be, and how you can best collect that data based on
that?

HTH, -Harry

PS: I'm doing low-latency real-time audio programming, which is extremely
timing dependent, and hence I just can't use Python for the audio
processing - its too slow. That said, python is very powerful to use for
writing glue code / code that isn't timing critical : like a menu that
allows you choose which test / experiment to run, etc etc..

-- 


Reply via email to