Characterizing the Relationship Between Eye Movement Parameters and Cognitive Functions in Non-demented Parkinson's Disease Patients with Eye Tracking

Our protocol aims to study the relationship between eye-movement metrics and cognitive functions in a group of non-demented Parkinson's Disease patients that suffer from both occular-motor and cognitive dysfunction. To design the trial images for the visual search task, use the numbers four, six, seven, and nine exclusively. Ensure that the location of the target number is randomized from trial to trial with the rule that each number can not be in the same visual quadrant for more than three successive trials.

Do not use ambiguous letters such as I and O, and set the size of the fixation cross, letters, and numbers at a 0.85 degree visual angle. Set the program to allow a time lapse of 1.5 seconds after Enter is pressed, before the display of the central fixation cross is switched to a trial image to begin a trial. Ensure that the screen will go blank with the fixation cross reappearing as the mouse is clicked or after 10 seconds have elapsed from the beginning of a trial, whichever is earlier.

Then set the program to generate a csv file that contains the time stamps of the beginning and end of each trial. After conducting a clinical diagnostic interview with the subject and if available, their relatives to exclude dementia and to screen for cognitive impairment using the Mini-Mental State Examination and Montreal Cognitive Assessment, evaluate the subject's visual acuity with a Snellen chart. Then, in a quiet room with an adequate light source, set up a screen based eye-tracker with a sampling rate of at least 300 Hertz, a computer, a mouse, and a standard keyboard.

A chin rest placed 60 centimeters in front of the eye checker screen and the appropriate cognitive assessment tools. To set up for a visual search task, position the subject on a chair with their chin on the chin rest and their forehead against a bar to minimize any head movement. Align the subject's eyes to approximately the center of the computer screen and click the start button in the eye tracking program.

Click start to calibrate the eye tracker with the built-in calibration program. And ask the subject to gaze at the red dot moving across the screen with nine fixation points while keeping their head still. View the calibration plot to check the quality of the calibration, making sure that the length of the green lines, which represent the error vectors fall within the gray circles.

Then click Accept to proceed to the visual search task. To initiate a practice trial, instruct the subject to fix their gaze on the central fixation cross and press Enter on the keyboard to begin the trial. The computer screen will display a single number and 79 randomly scattered distractor letters.

Instruct the subject to look as quickly as possible for the number and to simultaneously click on the mouse while stating the number aloud as soon as the number is located. Then cross check whether the stated number is correct and repeat the practice trial four more times. For eye tracking data processing and analysis in the replay section of the computer program, check the sample's percentage of the eyes during the visual search task.

Click play for the recording, to check the quality of the data by eyeballing the visualized scan path video generated. Discard all of the subject's data if it is grossly erroneous. And discard any trials in which the subject clicked the mouse accidentally or prematurely.

In the data export section of the program, select GazePointX and GazePointY Click export data to export the subject's data and save the data as a csv file. In the interface of the visual search analyzer, select the exported data as the input of the eye data and the generated csv file as the input of the action data. Select ST DB Scan as the classification algorithm and click run.

Then click summary to generate a spreadsheet file containing the main saccade amplitude and the mean fixation duration of the subject. To design the visual search analyzer, program the analyzer such that it extracts and analyzes only the data from the beginning to the end of the trial using the generated csv file. To program the analyzer such that it fills in the data lost due to eye blinking, average the x and y coordinates of the gaze point immediately before and after the blinking.

Then use the algorithm based on the ST DB scan to program the analyzer such that it classifies the raw data into either saccade or fixation. Consistent with previous studies, in this analysis, the Parkinson's disease group showed a poorer performance in multiple cognitive tasks, compared to the control group. Using the in-house algorithm for classification of the visual search task data, fixations and saccades can be identified and extracted for calculation and analysis.

Indeed, the disease group had a smaller mean saccadic amplitude compared to the control group, and the mean fixation duration was not significantly different between the groups. In this evaluation, after adjusting for covariates, negative correlations were found between the mean fixation duration, the performance in the verbal recognition memory and discrimination score, in the pattern recognition memory, and in the categorical verbal fluency test within the fruit and vegetable categories. However, there was no significant interactions found in these correlations between the disease and control groups, suggesting that the correlations are not specific to the disease group.

The design of the visual search task is important as the aim is to minimize the interference of the cognitive ability of the distractor eye movement parameters. If this research task is highly tolerable it can be applied to trans-diagnostically to other new degenerative disorder to answer similar research questions.