Water-restricted mice were trained to lever press for a small water reward (around 10 μl water) while freely moving in an operant condition box containing a single retractable lever and a shock grid floor (Coulbourn). Mice were allowed to retrieve a maximum of 50 rewards per day, and sessions were terminated after all rewards had been retrieved or after 30 min. After each lever press, the lever was retracted for 5 s before extending again. After mice retrieved 50 rewards for at least 3 consecutive days (typically 2–3 weeks of training), they were allowed to proceed with stimulation experiments. On shock days, mice were given a 1-s, 0.1-mA foot shock after 10% of lever presses instead of water. Shocks were delivered in a pseudorandom order on lever-press trials 5, 13, 24, 31 and 44, and the time to the next lever press was measured from the time elapsed for these trials until the subsequent lever press. During stimulation experiments (both baseline and shock days), optical stimulation was delivered throughout the experiment. Water was delivered using a custom set-up consisting of a lick spout (Popper and Sons, stainless steel 18-gauge) and a solenoid (Valcor, SV74P61T1) controlled by a microcontroller (Arduino Uno R3). Licking was monitored using a capacitive sensing board (Arduino Tinker Kit) wired to the lick spout and interfacing with the microcontroller. Shocks were delivered using an 8-pole scrambled shock floor (Coulbourn). Behavioural stimuli—lever presentations and retractions, and shocks—were controlled with Coulbourn Graphic State software. The timing of lever presses and licks was also recorded at 5 kHz using a data-acquisition hardware (National Instruments, NI PCIe-6343-X).
Free full text: Click here
