demo_experiments

This folder contains a series of simple readme scripts that demonstrate different functionality of Titta. The recorded data is stored in a .mat file, and can be analyzed using the example scripts in the demo_analysis folder of this repository.

Furthermore contained are a simple smooth pursuit display demo that shows how to access the acquired data by plotting it after running a trial, a complete antisaccade protocol and an eye-controlled version of the BreakOut! game implemented using Titta functionality.

Each readme script is configured for use with a Tobii Pro Spectrum at its default sampling rate (600 Hz). If you have a different eye tracker, or your Spectrum does not support 600 Hz, the script will not run with the default setup, providing you clear error messages telling you what is wrong. To adapt Titta's settings to your setup, as needed:

• Change the line settings = Titta.getDefaults('Tobii Pro Spectrum'); in the scripts to name a different eye tracker (see the main readme for naming of the eye trackers, or just run the demo as is and it will tell you what eye tracker it did find)
• Add a line settings.freq = 300; to change the sampling frequency to, e.g., 300 Hz.

In general the logic is that you should never change any of the code in the Titta distribution to adapt to your setup (if that is necessary, that is a bug--please let me know by opening an issue), but by changing the settings in your script, like the readme scripts demonstrate. Note that Titta's debug mode is switched on in the demos (settings.debugMode = true;) to provide more verbosity about its operation, to provide some insight about what is going on behind the scenes (mostly during calibration and validation). You can skip this in your own scripts without changing functionality.

The following readme scripts are available:

• readmeMinimal.m: Base readme script, showing "default" operation and providing the best starting point for developing your own experiment.
• readme.m: Expanded version of the readmeMinimal.m base script adding some less-used options (such as bi-monocular calibration and more configurability of the calibration display).
• readmeChangeColors.m: Version of readme.m showing how to change the background screen color of the calibration display, as well as the colors of various other elements on the setup and calibration screens.
• readmeImageCal.m: Version of readme.m showing how to use the ImageCalibrationDisplay class included with Titta to use a set of images as calibration/validation point. Includes support for animated gifs.
• readmeVideoCal.m: Version of readme.m showing how to use the VideoCalibrationDisplay class included with Titta to use a video or set of videos as calibration/validation point.
• readmeProLabIntegration.m: Version of readme.m showing how to use the TalkToProLab class for allowing experiments created and run from MATLAB/Octave to be visualized and analyzed in Tobii Pro Lab.
• readmeTwoScreens.m: Version of readme.m showing use of the dual monitor mode of Titta that provides separate participant and operator screens.
• readmeChangeColorsTwoScreens.m: Version of readmeChangeColors.m showing use of the dual monitor mode of Titta that provides separate participant and operator screens.

Besides the default operation mode with a simple interface for calibration and validation that is suitable for participants who can follow instructions to fixate the calibration targets, Titta also includes a separate more advanced calibration interface that gives the operator more control over what is shown to participants and automated calibration/validation procedures. This interface may be suitable for calibrating participants who cannot follow instructions, such as non-human primates and infants. The advanced calibration interface and the automated calibration/validation procedures are described in more detail in the paper:

Niehorster, D. C., Whitham, W., Lake, B., Schapiro, S. J., Andolina, I. M., & Yorzinski, J. L. (in prep). Extending the Titta toolbox to enable calibrating and validating Tobii eye trackers for nonhuman primates and other subjects unable to follow instructions.

In addition to the main Titta paper, please cite this paper when using Titta's advanced calibration interface.

This advanced functionality is demoed in the following readme scripts:

• readmeAdvancedCalibration.m: Version of readme.m showing use of the advanced calibration interface implemented with the Titta.calibrateAdvanced() function.
• readmeAdvancedCalibration_auto.m: Version of readmeAdvancedCalibration.m that shows usage of a calibration controller (MultiStepCalController) to automatedly run a calibration procedure consisting of multiple steps, and provide rewards when subjects look at the calibration points.
• readmeAdvancedCalibration_nonhuman_primate.m: Version of readmeAdvancedCalibration.m that shows usage of a calibration controller (NonHumanPrimateCalController) to automatedly run calibration and validation procedures driven by of gaze-contingent data collection using small videos to capture attention. Before the collection of calibration data starts, subjects are first trained by means of rewards to look at the screen and then at increasingly small videos.