Undoing of firing rate adaptation enables invariant population codes

Abstract

Neural adaptation supports coding efficiency by tuning responses to prevailing stimulus statistics. However, when information is represented by neural populations, adaptation of individual units could degrade behaviorally relevant signals. Here we investigate how the fly olfactory system implements adaptation in Olfactory Receptor Neurons (ORNs) and the consequences for combinatorial coding in downstream circuits. We show that adaptation of ORN firing rate is compensated at the axon terminal, where calcium transients remain background-invariant through inhibitory presynaptic feedback. Background invariance requires an adaptation strategy that shifts ORN response amplitude rather than sensitivity, diverging from efficient coding principles in single neurons. This property supports contrast encoding in ORN populations necessary for background compensation across the glomeruli. Downstream, the modulation of presynaptic Unc13 proteins maintains postsynaptic projection neurons responses to ON stimuli background invariant. We identify a new coding strategy where olfactory neuronal populations encode asymmetrically contrast information by implementing circuit computations that compensate peripheral firing rate adaptation.