Simulated prosthetic vision confirms checkerboard as an effective raster pattern for epiretinal implants

Area of Science:

• Neuroscience
• Biomedical Engineering
• Ophthalmology

Background:

• High-density retinal implants require spatial scheduling of electrode activation (rastering) for safe operation.
• The perceptual impact of different raster patterns on simulated prosthetic vision (SPV) is not well understood.
• Optimizing raster patterns is crucial for enhancing functional vision in retinal prostheses.

Purpose of the Study:

• To systematically evaluate the effect of various raster patterns on performance and perceived difficulty in SPV.
• To identify optimal spatial arrangements of electrode activation for improved visual prosthetic usability.

Main Methods:

• Sighted participants performed letter recognition and motion discrimination tasks using an immersive SPV system.
• Four raster patterns (horizontal, vertical, checkerboard, random) were simulated, emulating epiretinal implant perception.
• Psychophysically validated models accounted for electrode activation, phosphene appearance, and visual processing dynamics.

Main Results:

• The checkerboard raster pattern achieved significantly higher accuracy and lower perceived difficulty compared to other patterns.
• Horizontal and vertical patterns induced motion artifacts, while checkerboard minimized these biases.
• Random patterns yielded the lowest performance; checkerboard matched performance without rastering under safety constraints.

Conclusions:

• This study provides the first quantitative, task-based assessment of raster patterns in SPV.
• Checkerboard scheduling enhances perceptual clarity and usability in retinal prostheses with minimal computational overhead.
• This finding offers a clinically relevant strategy for improving next-generation retinal implant functionality.