get eyetracking back up

[GoogleCodeExporter]

ross is looking into this

Original issue reported on code.google.com by erik.fli...@gmail.com on 7 Jun 2012 at 7:11

[GoogleCodeExporter]

Original comment by erik.fli...@gmail.com on 7 Jun 2012 at 7:16

• Added labels: Type-Task
• Removed labels: Type-Enhancement

[GoogleCodeExporter]

look for papers on infrared rodent eye tracking and calibration:
http://www.frontiersin.org/decision_neuroscience/10.3389/fnins.2010.00193%20/abs
tract

picture of our calibration mechanism:
http://www-biology.ucsd.edu/labs/reinagel/SriramMeierPoster2011.pdf

picture of our eyetracks:
http://www-biology.ucsd.edu/labs/reinagel/RodentBehavior2.html

the eyetracker i use.  (was the best as of 5 years ago, but should do
a quick check to make sure that's still true):
http://www.sr-research.com/

come up with a proposal for:

-which eyetracking system and why
 - 1-2kHz (so we can see microsaccades)
 - successfully used w/rodents
   - what optics mods necessary?
   - how calibrate?
 - can be integrated realtime w/matlab
  - ok if via c

-is cox realtime?
-how is cox different from stahl?  how did it get published?

-which camera should we go with?
 -compare prosilica gc660, iscan, eyelink, and any others
   -frame rate
   -communication protocol
   -cost
   -any optics mods we have to make?
 -look for commercial infra red eye tracking systems (esp w/rodent experience)

-software
  -matlab in general
  -psychtoolbox - matlab stuff for graphics
     -includes eyelink toolbox - talks to eyelink
  -ratrix (http://code.google.com/p/ratrix/source/browse/#git%2Fdocumentation)

-has anyone studied microsaccades in rodents?

cris brought up the idea today that in a lot of what we're doing,
infrared (from the 2p laser) will be coming out of the pupils.  this
could screw up the systems we're looking at.  it also might be
something we can use to our advantage, but i wouldn't want to require
that we've added LEDs etc in order to track...

Original comment by erik.fli...@gmail.com on 9 Oct 2012 at 4:20