The coding of spatial position by the human visual system: effects of spatial scale and retinal eccentricity

Area of Science:

• Neuroscience
• Visual Perception
• Computational Vision

Background:

• The human visual system's ability to encode spatial position is crucial for navigation and interaction.
• Understanding spatial encoding requires investigating how visual information is processed across the retina, considering factors like spatial scale and retinal eccentricity.

Purpose of the Study:

• To investigate the computations underlying spatial position encoding in the human visual system.
• To explore the relationship between alignment accuracy and retinal eccentricity, independent of luminance, contrast, or orientation cues.
• To determine if spatial scale influences the accuracy of spatial localization.

Main Methods:

• Presented stimuli designed to isolate spatial localization from other visual cues.
• Measured alignment accuracy across varying retinal eccentricities and spatial scales.
• Modeled performance limits using computational models of visual processing.

Main Results:

• The relationship between spatial localization accuracy and retinal eccentricity is invariant with spatial scale when normalized by stimulus envelope size.
• Photoreceptor limitations do not dictate performance; the limit is post-receptoral.
• Performance is constrained by positional uncertainty within early visual filters before response extraction.

Conclusions:

• Spatial encoding of position in the human visual system exhibits scale-invariant properties concerning retinal eccentricity.
• Positional uncertainty in early visual processing, rather than peripheral receptor limitations, is the primary factor limiting spatial localization accuracy.
• This uncertainty is modulated by retinal eccentricity in a consistent manner across different spatial arrays.