Simple cells in the primary visual cortex predicted to match key properties of natural images

What is it about?

Natural images represent the natural environment and objects we see every day, such as shadows, flowers, foliage, landscapes, and textures. The primary visual cortex is one of the first areas in our visual system that is sensitive to complex visual stimulus features, such as the orientation of a bar or an edge in the visual field. Evidence also suggests simple cells carry out efficient coding of natural sensory data: re-coding sensory data in a way that reduces redundancy. This step may facilitate visual information processing at later stages of the visual system. This work takes a well-established model of simple cells and adapts it to perform efficient coding, thereby reproducing two key properties of natural images.

Featured Image

Why is it important?

The aim of this work is to better understand the principals of sensory information processing in the human brain. This in turn could lead to breakthroughs in medicine and health. This work may also help to improve algorithms for machine learning and artificial intelligence applications.