How Much the Eye Tells the Brain

Abstract

In the classic "What the frog's eye tells the frog's brain," Lettvin and colleagues showed that different types of retinal ganglion cell send specific kinds of information. For example, one type responds best to a dark, convex form moving centripetally (a fly). Here we consider a complementary question: how much information does the retina send and how is it apportioned among different cell types? Recording from guinea pig retina on a multi-electrode array and presenting various types of motion in natural scenes, we measured information rates for seven types of ganglion cell. Mean rates varied across cell types (6-13 bits . s(-1)) more than across stimuli. Sluggish cells transmitted information at lower rates than brisk cells, but because of trade-offs between noise and temporal correlation, all types had the same coding efficiency. Calculating the proportions of each cell type from receptive field size and coverage factor, we conclude (assuming independence) that the approximately 10(5) ganglion cells transmit on the order of 875,000 bits . s(-1). Because sluggish cells are equally efficient but more numerous, they account for most of the information. With approximately 10(6) ganglion cells, the human retina would transmit data at roughly the rate of an Ethernet connection.

Retina is structured to process an excess of darkness in natural scenes

“…It may also be metabolically efficient, because both types respond sensitively to small variations while maintaining low firing rates. Maintaining low rates is important, because information rates increase sublinearly with the spike rate (39) and energy cost and axonal volume increase supralinearly with firing rate (39)(40)(41). Thus, space and energy efficiency (bits/μm 3 ; bits/ATP) improve when contrast signals are rectified into lower rate ON and OFF channels.…”

Probability distributions of the logarithm of inter-spike intervals yield accurate entropy estimates from small datasets

“…Only a few years after Shannon's hallmark information papers (Shannon & Weaver, 1949), physiologists had adapted the theory to analyze neuronal activity (MacKay & McCulloch, 1952). However, despite major advances in entropy estimation techniques (e.g., Treves and Panzeri, 1995;Strong et al, 1998;Paninski, 2003) and substantial insights information theory has helped illuminate (e.g., Reinagel et al, 1999;Koch et al, 2006), its application to neuroscience remains a niche field. While there may be many justifiable reasons for neuroscientists to avoid information theory, we believe the primary reason that entropy measures are not more widely used is a perception that information theoretic results are less intuitive and more difficult to interpret than results from simpler variability statistics, including coefficients of variation (CV) or coherences, and ad hoc measures such as burst indexes.…”

Microcircuitry for Two Types of Achromatic Ganglion Cell in Primate Fovea

“…Also, it receives a nearly threefold higher proportion of amacrine contacts, which are likely to be inhibitory. In other mammals, low information rates and strong inhibitory input are characteristic of "local-edge" cells (Koch et al, 2004(Koch et al, , 2006van Wyk et al, 2006). Furthermore, because in other mammals local-edge cells are the smallest cells, they are also among the most numerous and are thought to convey spatial acuity at low temporal frequencies (van Wyk et al, 2006).…”

“…Local-edge cells were rarely recorded in other mammals in vivo because the cells are small and the axons are fine (for review, see Troy and Shou, 2002). Only now that they can be easily recorded by patch electrodes in vitro (Koch et al, 2004;Xu et al, 2005;van Wyk et al, 2006) and by multielectrode arrays (DeVries and Baylor, 1997;DeVries, 1999;Koch et al, 2006) are they getting the attention warranted by their relatively high densities.…”