Gabriella Belmarez

October 23, 2015

Decoding the Neural Code for Vision

Researchers at the University of Pennsylvania have calculated that the human retina transmits data at about 10 million bits per second. So what does all of this visual information look like, and how do we interpret it? This summer Gabriella worked on this very question at the National Eye Institute by analyzing the visual signals collected from mice ganglion cells. Visual information is transmitted by neurons in the eye to the brain through electrical pulses or spikes. The visual scene is encoded in a sequence of these electrical pulses, also known as spike trains. Changes in the visual scene cause the increase or decrease of the rate of spikes in these spike trains. However, complex features of vision such as light intensity, motion, color, and object size cannot be conveyed only by an increase in the rate of spikes. Recent statistical analysis suggests that complex temporal patterns within the spike trains, also known as higher order correlation, may provide a more developed model of the neural code for vision.