Detection of visual borders is an essential function for transforming the visual scene into perceivable boundaries. Avian predators, for example, are able to recognize complex textured borders of conspicuous prey. It has been suggested that the process of border perception begins as early as in the retina through a class of ganglion cell, the local edge detector (LED), which responds selectively to luminance edges. But what neuronal circuitry is used to accomplish this, and can this circuitry also be used to detect complex textured borders? Here we use patch-clamp electrophysiology to show that selectivity to luminance edges in the LED is accomplished by surround-originated feedback inhibition that suppresses excitation via GABAa and GABAc receptors. Furthermore, we find that excitatory circuitry has several characteristics that facilitate independent responses of individual bipolar cells to features as small as one-eighth the size of the LED's receptive field. This enables the LED to respond to areas of texture even if they contain no net luminance change. We observed that feedback inhibition is similarly activated by small features, which in turn causes excitation not only to respond selectively to luminance edges, but also to boundaries of luminance-neutral texture in synthetic and natural scenes, as well as a noise-suppression function suggested by modeling. We also characterized direct feedforward inhibition to the LED, and found it to be cospatial with excitation, glycinergic, and sensitive to small features as well. These characteristics enable the suppression of spikes during rapid luminance shifts, encoding an "edge in time". Our results suggest mechanisms by which three kinds of edges can be encoded by the retina and transmitted through a specialized channel to higher visual areas in the brain.

One of the visual system’s many tasks is to be able to distinguish objects from the background. The ability to do this is limited and affected by the relationship between the object (or stimulus) of interest and the background. Adaptation in retinal neurons is the process of changing the cell’s response to a stimulus according to that stimulus’s background. When the stimulus is hard to discern from the background, the retina adapts by improving its sensitivity to low contrast. The large response range maintained by adaptation comes at a cost, however. Adaptation complicates neural coding by making the brain interpret identical stimuli as different based on differences in background. In order to further our understanding of adaptation, this study modified the background to be in terms of time and space rather than light intensity as is the norm. By changing the interval between two circular stimuli (inter-stimuli interval; IsI) of the same diameter, and by changing the diameter over a common IsI, we measured a ganglion cell’s output for one stimulus relative to another stimulus. the results show saturation (loss of output to the 2nd stimulus) of stimuli at lower IsIs. Also, the degree of saturation for a given IsI depends on the diameter of the stimulus. These combinations of results illustrate the temporal and spatial dependence of adaptation on ganglion cells. A larger-diameter stimulus involves multiple neurons surrounding the ganglion cell being recorded so various pathways most likely influence that cell’s ultimate output. Rapid stimuli (low IsI) can be defined as having large mean luminosity that directly affects ganglion cell output.

We studied the interactions between excitation and inhibition in morphologically-identified amacrine cells in the light-adapted rabbit retinal slice and wholemount using the patch clamp technique with pharmacology.

We found that the majority of ON amacrine cells received glycinergic OFF inhibition. About half of the OFF amacrine cells receive glycinergic ON inhibition. By mapping the receptive field of inhibition, we found that this glycinergic inhibition was typically smaller than the receptive field of excitation. In retinal slice, a minority of ON, OFF, ON-OFF amacrine cells received both glycinergic ON and GABAergic OFF inhibition. In retinal wholemount, ON-OFF inhibition was found to be glyceringic. These interactions were found in cells with diverse morphologies (wide and narrow cells, with either monostratified or diffuse processes). Most ON-OFF amacrine cells received no inhibition and had monostratified processes confined to the middle of the inner plexiform layer. Glycinergic inhibition was the dominant type of inhibition we measured in amacrine cells. The most common interaction between amacrine cells that we measured is "crossover inhibition," where OFF inhibits ON, and ON inhibits OFF.

After we recorded from amacrine cells, we identified their neurotransmitter content using immunocytochemical techniques, and correlated this with their morphological and physiological characteristics. We found that narrow cells, or cells that conveyed local information, usually contained glycine. Wide cells, or cells that conveyed global information, usually contained GABA. Some narrow field cells that received wide-field inhibition contained GABA. Some cells had receptive fields smaller than their dendritic fields, and appeared to be able to generate action potentials.

Data from electrophysiology, pharmacology, morphology and immunocytochemistry revealed that amacrine cells interact with one another using primarily glycine. GABAergic amacrine cells play a larger role in feedback and feedfoward inhibition to bipolar and ganglion cells than in amacrine cells. Narrow-field, glycinergic amacrine cells are primarily involved in crossover inhibition, which is also a common circuit topology in electronics. These interactions between amacrine cells forms the basis for the diverse inhibitory inputs to ganglion cells. With a deeper understanding of retinal circuits, we hope to design more effective retinal prostheses for the treatment of retinal disease.