Feedback inhibition is generally believed to be a ubiquitous feature o
f brain circuitry, but few specific instances have been documented. An
example in cats is the supposed feedback circuit involving relay cell
s of the lateral geniculate nucleus and cells of the perigeniculate nu
cleus (a part of the thalamic reticular nucleus). geniculate relay cel
ls innervate the perigeniculate nucleus, which, in turn, provides an i
nhibitory, GABAergic projection back to the lateral geniculate nucleus
. However, feedback inhibition at the single-cell level requires that
a given perigeniculate cell project back onto the same geniculate rela
y cell that innervates it. We probed for this in an in vitro slice pre
paration of the cat's lateral geniculate nucleus. We evoked a single a
ction potential in a geniculate cell via a brief, depolarizing pulse d
elivered through an intracellular recording electrode and looked for a
ny evoked hyperpolarizations. For 6 of the 36 geniculate cells tested,
we observed a long-lasting hyperpolarization after the action potenti
al, and much of this was eliminated by application of bicuculline, sug
gesting synaptically activated inhibitory postsynaptic potentials. We
interpreted this to be dear evidence that a given neuron may inhibit i
tself via circuitry mediating feedback inhibition in the cat's lateral
geniculate nucleus.