SCNI_EyeCalib - nimh-nif/SCNI_Toolbar GitHub Wiki

This function presents a fixation marker to the observer, at either the center of the screen or at pseudo-randomized locations within a pre-defined grid. The observer is rewarded for fixating the marker with the required accuracy, while the experimenter is able to update the eye tracking calibration parameters. The settings specific to this experiment can be edited via the GUI that is opened using SCNI_ShowImagesSettings.m (accessible from the SCNI_Toolbar).

In the SCNI, the observer's eye position is measured by the EyeLink II video-based eye tracker system, which is output as 6 channels of analog voltages via the analog output card: left eye X, Y and pupil diameter; right eye X, Y and pupil diameter. These analog voltages (+/-5V range) are routed to both the TDT RZ2 and the DataPixx2 via the SCNI Interface Box. Since the EyeLink II software (unlike EyeLink 1000) does not allow the user to calibrate the analog output, adjustments must be made on the DataPixx side instead.

1. Central fixation

With the central fixation tickbox selected in the SCNI_EyeCalibSettings.m GUI, the fixation marker will only be presented at the center of the screen. The experimenter should observe the subject's eye movements to discern whether the subject is in fact attempting to fixate the central target. Look for consistent location of the gaze marker on the experimenter's display. To set the correct offset voltage, either:

  • press the C key while the observer is successfully fixating
  • move the mouse cursor to the point on the experimenter's display where the gaze marker appears when the observer is successfully fixating, and press the M key at any time.

2. Peripheral fixation

With the central fixation tickbox unselected in the SCNI_ShowImagesSettings.m GUI, the fixation marker will appear randomly at locations across the screen. You can determine which locations are used by selecting either a grid or radial layout, and select the total number of marker locations. Radial targets appear at 8 polar angles, while grid targets appear in a NxN grid (where N is greater than or equal to 3). While the observer fixates the targets across the screen, the experimenter must manually adjust the gain settings as follows:

  • press the E key to toggle between left and right eyes (the eye parameter text will change color)
  • press the Right Arrow key to toggle between gain adjustment for X and Y channels (the highlight box will move)
  • press the Up Arrow key to increase the gain (degrees of visual angle per Volt) for the currently selected eye and channel, or the 'Down Arrow' to decrease it.
  • press the I key to invert the polarity of the gain for the currently selected eye and channel

3. Calibration complete

When you are satisfied with the accuracy of the calibration parameters (i.e. the gaze marker consistently hovers over the fixation marker position when the observer is fixation), hit the S key to save the current calibration parameters. These data will be written to the .mat file specified at the start of the session. Any subsequent SCNI Toolbar experiments will load these parameters each run and send them to the TDT. This information is critical for converting eye position voltages into meaningful coordinates (pixels, degrees, centimeters, etc.)!

You can re-run the eye calibration procedure at any point during a session (e.g. in between runs of the main experiment). The updated calibration values will be saved to subsequent .mat files, and the values will be sent to the TDT as either event codes (via digital input) or text strings (if UDP is available).