MATLAB utility functions written by Kendrick Kay.zip

- For additional information, see http://kendrickkay.net/analyzePRF/ %%%%%%%%%%%%%%%%%% INSTRUCTIONS FOR PRESENTING STIMULI - quit all unnecessary applications - consider turning off networking - plug in the power cord (if laptop) - change to a directory where you want to save the stimulus presentation .mat files - start an experiment by running the runretinotopy.m script - to quit a run early, press ESC - the run starts when a "t" is detected on the keyboard (this is configurable in the script file) - make sure the stimulus starts at exactly the right time! (the <tfun> input prints out a message to the command window when the trigger is detected.) - after a run completes, a .mat file is written out with timing information and button press information. these are important; keep these files! - make sure to test an entire run (with button presses and so forth) before an actual scan! %%%%%%%%%%%%%%%%%% INFORMATION ON THE STIMULI - there are several types of experiments, and they are numbered as follows: 89. wedges CCW 90. wedges CW 91. rings expand 92. rings contract 93. multidirectional bars 94. wedge/ring combination - each of the five run types is exactly 300 seconds long - there are some rest periods at the beginning and end and in the middle of the runs. the subject should always be fixating the central dot and doing the task. - stimuli are constructed for a resolution of 768 pixels - it is assumed the refresh rate of the display is 60 Hz - either run at 1024 x 768 or a higher resolution (in which case the stimuli won't fill the whole display) %%%%%%%%%%%%%%%%%% INFORMATION ON THE TASK - the task is a fixation task. there is a central dot that can take on different colors. the dot switches color randomly. the time between switches is chosen randomly (uniformly) within a certain range. these properties are all configurable in the runretinotopy.m script. - a suggested task is to press a button whenever the dot color changes. %%%%%%%%%%%%%%%%%% TIMING ISSUES - there are exactly 300 seconds in each run - the refresh rate of the stimulus is 15 Hz, so there are 4500 frames in each run - the stimulus presentation code attempts to compensate for any glitches (missed frames). after the stimulus is complete, make sure the number of glitches reported is low (e.g. 5 or less). - after the presentation of the stimulus is finished, some information is output to the screen regarding timing. important here is the actual empirical time that it took to show the stimulus. this number may not be exactly 300.000 s because the display device may not actually be at 60 Hz (e.g. it might run at 60.2 Hz). make sure to test a complete stimulus run with the laptop computer connected to the display device to see what the precise time is. - I like to set the TR to be exactly the time that is necessary, so that the total duration of the scan is exactly the total duration of the stimulus presentation. - for ease of analysis, it would be nice if the TR was either 1 second or 2 seconds (so that you get 300 or 150 volumes per run). (of course, the precise number might be something like 998.952 ms or 2002.056 ms if you are trying to get it precisely matched to the stimulus computer.)