Assessing visual performance during intense luminance changes in virtual reality

Area of Science:

• Visual neuroscience
• Human-computer interaction
• Ophthalmology

Background:

• Human vision struggles with rapid luminance changes during transitions (e.g., indoor-outdoor).
• Existing photoreceptor limitations pose risks of overlooking low-contrast hazards.
• Adaptation processes are crucial for restoring optimal vision after intense light shifts.

Purpose of the Study:

• To develop and validate a virtual reality (VR) simulation platform for studying human visual performance under steep luminance changes.
• To investigate the impact of simulated self-tinting lenses on visual detection times following luminance transitions.
• To assess the efficacy of adaptive lens technologies in mitigating visual impairment during dynamic lighting conditions.

Main Methods:

• Linearization of VR headset luminance output for accurate stimulus presentation.
• Recruitment of ten participants to record visual detection times.
• Presentation of small, low-contrast stimuli in peripheral visual fields (±10°) after controlled luminance changes (1s or 3s duration).
• Comparison of detection times between simulated self-tinting lenses and fixed transmission lenses.

Main Results:

• Significantly faster detection times were observed with simulated self-tinting lenses compared to fixed lenses following luminance decreases.
• No significant difference in detection times was found for luminance increases across lens conditions.
• The VR platform successfully simulated steep luminance changes and measured visual performance.

Conclusions:

• The developed VR simulation platform is effective for studying visual perception during and after rapid luminance fluctuations.
• Simulated self-tinting lenses demonstrate a potential benefit in improving visual detection following decreases in ambient light.
• Findings support the development of adaptive eyewear, such as self-tinting lenses, to enhance visual safety and performance in dynamic lighting environments.