Skip to content
Rosemary Le edited this page Sep 10, 2018 · 27 revisions

This wiki describes Rosemary Le's experimental work and analysis in two papers:

The field of view available to the cortical reading circuitry, 2017
Rosemary Le, Nathan Witthoft, Michal Ben-Shachar and Brian Wandell
Journal of Vision, April 2017 Volume 17, Issue 4
  • Reference 2: In progress.

Software dependencies

  • For the localizer experiment, we rely on fLoc
  • For checkerboard retinotopy, we rely on vistadisp
  • For object retinotopy (words, falsefont, etc), we rely on knkutils
  • Psychtoolbox is required to display all stimuli
  • For analyses, we rely on this repository and vistasoft

Experimental Methods

Localizer methods

The localizer was modified from Anthony Stigliani's code package. The steps to modify the code are described in this section.

Stimulus set

Face stimuli: adult_grayBackground directory. TODO: store somewhere that is not github

Word stimuli: TODO: store somewhere that is not github

Software that controlled the presentation

  1. Install the code here: https://github.com/astigliani/fLoc which will set up the directories (data, functions, scripts, and stimuli). Documentation for this code is at: http://vpnl.stanford.edu/fLoc/.
  2. Go into the stimuli directory. Delete all directories except the scrambled.
  3. Next we set up the face stimuli: Copy the adult_grayBackground directory from _____ and place it at the same level as scrambled.
  4. Next we set up the word stimuli: Copy the hebrew directory from ____ and put it at the same level as scrambled.
  5. Now we're ready to run. Open the runMe_ScriptLocalizer_tiled.m script.
  6. Edit subject.name in the second cell.
  7. Run the script. It takes about 30sec for the stimuli to load.

Other notes:

  • Task: The subject hits a button every time a scrambled image appears. This may occur 0 or 1 time within a word and face block or 4x in a row (when the scrambled images are shown in its own block). The subject performance displayed at the end is incorrect ... never got around to fixing that.

  • It is REALLY IMPORTANT that you hit ‘g’ at the end of the run. There will be instructions on the screen that tell you to “hit g to continue" so if you are good at following instructions you should be fine. The order that the categories are displayed (they’re randomized between subjects) will be lost if you forget this step.

Checkerboard pRF methods

Stimulus set

The stimulus is included in the vistadisp repository. Make sure this repository is in your Matlab path.

Software that controlled the presentation

  1. Navigate to coverageReading/retAndLocStimuli/
  2. In the Matlab command window, type:

load paramsForRet_5min.mat

params = ret(params)

The checker retinotopy will then run.

The location of the saved image matrix will be indicated in the params.saveMatrix field.

Bars with words, faces, and objects pRF methods

Stimulus set

TODO The image files are large and github can't handle them ... put them somewhere else.

Software that controlled the presentation

  1. Go to this page: http://kendrickkay.net/analyzePRF/ and follow steps 1-5 under Stimulus presentation code.
  2. In the script call_retinotopyKnk.m (located in coverageReading/retAndLocStimuli/) make the necessary edits in the cell "modify here"
  3. Run it!

Other notes

  • The 'g' button will start the stimulus
  • Once the experiment starts, the 'q' button will let it exit gracefully
  • A params file with all information (stimulus, subject responses) is saved to the same directory where this call_retinotopyKnk.m script is located.

Analyses

Key Figures

Individual FOV (with pRF centers and half max contour)
figScript_coverage_individual.m
E.g. Figure 4 in FOV paper

Group average FOV (with half max contour)
figScript_coverage_contours_generic.m
E.g. Figure 3C in FOV paper

Group average FOV with bootstrapped confidence interval
figScript_coverage_bootstrapOverIndividuals.m
E.g. Figure 8B and D in FOV paper

Variance explained for each subject
summary_individual_varExp.m
E.g. Figure 6A in FOV paper

Time series for a single voxel, multiple data types
tSeries_plotVoxelMultipleRms.m
E.g. Figure 6B in FOV paper

ROI definition on the mesh
meshimg_screenShot.m
E.g. Figure 7A in FOV paper

Distribution of Relative_rmse in an ROI
summary_modelPred_vs_testRetest.m
E.g. Figure 10C in FOV paper

PRF centers, colored by subjects
figScript_coverage_centers_group.m
E.g. Figure 2 in pRF paper

Pairwise comparison of a ret model parameter in a heat map
summary_pairwise_heatMapWithFittedLine.m
E.g. Figure 3 in pRF paper

Overlapping histograms of a ret model parameter
summary_histogram_spline.m
E.g. Figure 6B in pRF paper