A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss

Macular degeneration, a leading cause of vision impairment, lacks standardized therapies. It causes central vision loss, forcing reliance on peripheral vision for daily tasks. Our research seeks to understand visual changes after vision loss and develop effective training interventions by establishing a standardized framework for studying central vision loss.

For the experiment, we have faced a variety of challenges such as synchronization errors between monitors due to the need for speedy screen refresh rates for our gaze-contingent display, making sure that the code is consistent across sites on our data collection computers, creating a smooth, short latency movement of the scotoma and accurately measuring the ocular motor performance of participants. By training and assessing portions of the participant's peripheral vision, we're able to look at changes across low, mid, and high levels of visual processing. This will allow us to better understand how these different parts of the visual processing change with perceptual learning following central vision loss.

Our work impacts both scientists'fundamental understanding of learning and its clinical implications, particularly in individuals with vision loss. Our approach allows us to examine what strategies improve different aspects of vision. We can also use these methods to understand what aspects of brain function are changed when we learn to use vision in these various ways.

Our next steps focus on two direction. First, we're using these tools to understand how the brain changes when different aspects of visual processing improve. And second, we're working on implementation of these assessments using tools like VR headsets that patients can use easily at home.

This extends the reach of the work to other populations that can't come into the lab three times a week. To begin, provide the participant with audio-visual instructions for the session tasks. Include dedicated video instructions with screen captures from the actual task for each activity.

Then, explain the instructions verbally to ensure the participant fully understands what to expect during the task. Prior to each task, perform calibration and validation of eye movements. Then, provide the participant with practice trials before beginning each main task.

Clarify any participant questions about the task during the trials. Next, have the subject conduct the fixation task during the initial visit before gaze-contingent tasks. Train participant to position their simulated scotoma within a white central box on the screen for varying durations, while gradually increasing spatial tolerance across trials.

Display a gaze-contingent computer interface with an opaque disc at the center of their gaze, simulating a scotoma that moves with their eye movements. Instruct the participant to move the simulated scotoma close to the opaque disc to reveal the target. To perform free-eye movement tasks, present an instructional video and scripted verbal instructions to the participant.

Then, calibrate and validate the eye tracking system between tasks. Instruct the participant on free-viewing tasks. Conduct tasks that require the participant to use their gaze for varied actions rather than fixating on one area.

Perform scotoma-based tasks by placing the scotoma near a cue to trigger a stimulus. Provide the participant with onscreen instructions at the beginning of tasks. Follow the practice trials to familiarize participants with the task requirements.

Ensure that the participant shows proficiency in using visual layouts, required oculomotor actions, and task responses before measuring performance. After each task, provide auditory feedback to the participant indicating the accuracy of their response. Introduce short breaks of up to one minute during tasks to prevent fatigue.

During fixation-constrained tasks, ask the participant to maintain their head position in the chinrest throughout the task duration. Ensure calibration accuracy remains consistent with the original position during the task. Provide the participant with onscreen instructions, followed by practice trials before the task.

After completing the practice trials, give a reminder set of onscreen instructions before starting the main task. Ask the participant to focus their eyes on the center of the screen using fixation aids, while responding to stimuli that appear in the peripheral vision to either side of the fixation box. Next, ask the participant to respond to stimuli using their right index finger on a five-button response box located to the right.

After each trial, provide auditory feedback to the participant indicating the accuracy of their response. As expected, participant one showed a significant queuing effect at the left location in the exogenous attention task based on cue congruency. However, no significant effect was found at the right location.

For participant two, there was no significant effect of cueing observed at either location. As expected, MNRead task results showed an increase in reading time as font size decreased, with participant two requiring more time than participant one at smaller font sizes. As expected, in the trail making task, both participants took significantly longer to complete part B than part A, with participant two requiring more time in both parts.

Fixation stability was greater for participant two, as indicated by a smaller bivariate contour ellipse area of 51 square degrees, compared to 61 square degrees for participant one. The probability density analysis using kernel density estimation showed distinct preferred retinal locus for both participants.