1-D Vision: Encoding of Eye Movements by Simple Receptive Fields

Area of Science:

• Neuroscience
• Computational Vision
• Visual Perception

Background:

• Eye movements (eyeM) are crucial for visual perception, enabling scene scanning at different scales.
• The precise trajectory of eye movements is vital for resolving spatial ambiguities caused by retinal image motion.

Purpose of the Study:

• To investigate how the trajectory of eye movements is encoded by retinal receptive fields.
• To determine if this encoding mechanism can explain visual motion illusions.
• To explore the decomposition of 2-D images into 1-D projections through receptive field encoding.

Main Methods:

• Modeling of symmetrical simple retinal receptive fields.
• Analysis of how these receptive fields encode eye movement trajectories.
• Examination of the Ouchi illusion as a case study for motion illusion explanation.

Main Results:

• Symmetrical simple retinal receptive fields can encode eye movement trajectories at high resolution.
• This encoding mechanism successfully accounts for motion illusions like the Ouchi illusion.
• Encoding motion projections via horizontal and vertical receptive fields results in a Cartesian decomposition of 2-D images into two 1-D projections.

Conclusions:

• The trajectory of eye movements is effectively encoded by simple symmetrical retinal receptive fields.
• This encoding provides a potential neural basis for understanding motion perception and illusions.
• The findings suggest a fundamental mechanism for image processing in the visual system, involving dimensional decomposition.