Schematic of the olfactometer configuration and stimulation protocols for conditioning (A) and testing (B-D).

<p><b>A</b>. For Pavlovian conditioning of all moths, the odor cartridge was connected to the normally closed output line (in gray) and the normally open output line (in red) was not used. The nozzle was placed 10 cm from the moth to provide enough distance to create an odor dispersion field (inset gray triangle) wide enough to cover most of the antennae. During each conditioning trial, odor (conditioning stimulus) was presented for 4 s; 3 s into odor presentation, the moth was presented with 0.75 m sucrose solution (unconditioned stimulus). <b>B</b>. In experiments 1-4 and 7 conditioned moths were tested with pulsed odor that was interleaved with clean air. This was achieved by connecting the normally open output line (red) with to the output of the normally closed line (gray) after the odor cartridge. In this configuration air is constantly flowing out the nozzle (inset gray and red triangle), independent of odor. <b>C</b>. In order to create oscillations in air flow velocity the normally open output line was not used (as in A). In both B and C the odor delivery nozzle was positioned approximately 2-3 mm in front of a single antenna, which was held in position at its base. This insured that the temporal structure of the pulse train was preserved and that further dilution of the odor with the surrounding air was minimized. <b>D</b>. Stimulation protocols used during testing. Shown are the first 200 ms (4 pulses) of the 4 s 20 Hz pulse trains for both duty cycles and stimulation protocols (interleaved and puffed).