The olfactometer consisted of an operant chamber connected to an odorant-delivery unit controlled by a computer. The odorants were presented in the sampling port via an odor port. A mouse was placed in the operant chamber prior to the begining of the test.

When the mouse puts its nose in the sampling port it breaks a photo beam that triggers a 2 s presentation of either an odorant used as the rewarded stimulus (S+) or a different odorant used as the unrewarded stimulus (S-).The presentation of the odorants is regulated by a final valve and the mouse has to keep its head in the sampling port for ~0.1 s before the odorant is presented. The final valve relaxes directly after the 0.1 s, if the mouse successfully stayed in the sampling port, presenting the odorant. A correct response to the S+ stimulus consists of licking at a waterspout in the sampling port, which opens a reinforcement valve for 0.04 s presenting a water droplet of 2.5 µl through the waterspout. If the mouse licks at the waterspout when the S- is presented the trial is regarded as incorrect and no reinforcement will be presented. 

All mice underwent 3 days of a Begin program and then 10 days of a D2 program.

The Begin program is divided in two stages;

The first stage is shaping, in which the mouse learns to put its head into the sampling port and lick at a waterspout (licking will be rewarded every time).

In the second stage the mouse learns to build an association between the presentation of an odorant and licking at the waterspout. Stage two is subdivided into 6 blocks each containing 20 trials. In the first block the odor is immediately presented when the mouse puts its head in the sampling port (makes a nose poke), while in the following five blocks, the mouse has to keep its head in the sampling port for 0.3, 0.6, 0.9, 1.0 and 1.2 s before the odour is presented. At the same time the reinforcement volume was step-wise increased so that the mice learned to continously lick at the waterspout when presented with a S+ stimuli.

The D2 program introduces the mouse to a two-odorant discrimination task in which two odorants are presented in pseudo randomized order. In this program the mouse learns to lick in response to a rewarded odorant (S+) and not to lick in response to a non-rewarded odorant (S-).