Color vision at low light intensity, dark adaptation, Purkinje shift, critical flicker frequency and the deterioration of vision at low illumination. Neurophysiology at the nanometer range of neural structure.

The discovery that color vision extends to low illumination reported in this communication eliminates the duplicity theory as an explanation of vision differing at high and low illumination. Instead, an explanation of the difference was found when analyzing synaptic interaction between retinal neurons, made possible by revealing the synaptic connections between the neurons through three-dimensional reconstruction of the outer plexiform layer and by applying information communicated by published recordings of the potential of retinal neurons. The synaptic connections revealed the existence of a large horizontal cell network and of cone networks. The networks contribute continuous information regarding average light intensity over an area of the retina. The opposite sign of the network input maintains bipolar cell threshold constant when illumination varies. When at low illumination the network potential approaches a minimum the consequent extensive increase of transmitter release at network synapses eliminates fine tuning of synaptic transmission at these synapses. This accounts for the deterioration of vision at low illumination by eliminating spatial brightness contrast enhancement and also accounts for the difference in critical flicker frequency at high and low illumination. Network interference accounts for the two phases of dark adaptation and for the Purkinje shift. The analysis revealed conditions for particularly fast synaptic transmission leading to cascade like transmission at sequences of synapses. The overall design of the neural circuits establishes conditions for fast processing of information. This is the consequence of the neurons responding with graded changes of the membrane potential and conducting potentials electrotonically. Such neurons are therefore particularly suitable for processing of information.