Retinal resistance barriers and electrical lateral inhibition

Abstract

EXTRACELLULAR fields around active neural tissues are usually only a small fraction of the transmembrane voltages generating them, but this is not so in the arthropod compound eye. Light-evoked fields (the electroretinogram, ERG) can range over 10 mV amplitude¹ and grow even larger near the receptor endings at the first synapse, the lamina (Fig. 1). The physical basis for such large fields is not understood although it has been suggested that there may be barriers of high resistance in the extracellular space^2,3 across which current flow may generate large voltages. Direct measurements of extracellular resistance profiles in the insect eye reported here indicate that a barrier does exist. I also show the effect such a barrier must have on the retinal currents produced by light, which is to set up a system of electrical presynaptic inhibition on certain of the receptor terminals, acting through the extracellular space. When this neural circuit is defined in detail it can be shown that it specifically selects for inhibition only those neighbouring receptors with properties dissimilar to the excited elements. Such a system might be of general interest as a mechanism of inhibitory interaction, since it can act in a graded manner, needs no neural wiring to set it up, and no genetic programme to account for the selectivity its inhibition shows. The developmental problem of how a neurone can establish highly specific connections with follower cells^4–6 and yet at the same time form a perhaps equally specific set of laterally-interacting connections with a quite different set of cells, is avoided with this particular mechanism.