): When a visual stimulus is presented repeatedly, a ganglion cell fires spikes at nearly identical times, often within 1 ms. This remarkable precision may support a timing code, which compared to a rate code can carry more information per spike. Previous research used extracellular recording to characterize the temporal precision of the ganglion cell's spike output. I propose to use whole-cell and intracellular recordings to identify mechanisms (microcircuits and ion channels) that set the timing of the ganglion cell's synaptic inputs. This effort has three parts: (1) Identify mechanisms that cause precise timing of the ganglion cell's input. Preliminary studies show ganglion cells with sequences of spontaneous postsynaptic currents (PSC) whose times are strongly correlated, often within 2 ms. Thus, precise timing of synaptic input might contribute to precise timing at the spike output. I will determine the mechanism(s) responsible for setting the timing of PSC inputs and determine how altering these mechanisms affects PSC timing. For example, preliminary studies show that inhibiting L-type calcium currents decorrelates PSCs. (2) Test the hypothesis that distinct microcircuits cause distinct patterns of input timing. Ganglion cells exhibit various cross-correlation patterns between glutamatergic (bipolar) PSCs and autocorrelations between GABA/Glycinergic (amacrine) PSCs. I will identify these patterns for specific morphological types of ganglion cell. I will then examine their synaptic input with the electron microscope to determine which circuit features, such as feedforward or feedback synapses, correspond to different correlation patterns. (3) Determine how efficiently the ganglion cell encodes its input as a spike train output. This will be accomplished in the intact retina by intracellularly injecting random amplitude currents (simulated postsynaptic currents) and measuring the ganglion cell's intrinsic temporal precision and information coding capacity. Then, by stimulating with random light intensities, I will measure the actual rate of information transmission. To transmit information requires metabolic energy, thus there may be selective pressure to match the ganglion cell's intrinsic information capacity to the amount of information that it actually transmits. These studies will advance a basic understanding of how discrete neural events (vesicular synaptic release and spike generation) might implement a timing code in the mammalian retina.
Showing the most recent 10 out of 16 publications
