2.3. Drinking in the Dark (DID) and Frontloading

Approximately 1 week after surgery, mice underwent 1 week of home-cage drinking using the drinking in the dark (DID) paradigm (Figure 1 A) [⁶²]. Mice received access to 20% ethanol (v/v) (Decon Laboratories, Inc.) and no water for 2 hours a day, Monday through Friday. Mice were single-housed in standard shoebox cages in a room with a 12-hour reverse light/dark cycle. Ethanol access was timed to 3 hours into the mice’s dark cycle (noon to 2 PM). Animals had ad libitum access to regular lab chow (LabDiet 5001) [⁶³] at all times when not head-fixed. Ethanol during DID was made available to mice in volumetric sipper tubes (Columbus Instruments, Inc.). These sippers measured the volume of consumed ethanol each minute, which allowed for the analysis of drinking patterns to assess if there was evidence for the presence of frontloading behavior. This approach has been used previously [^63–66].

We used recently introduced techniques to identify alcohol frontloading [⁶⁰]. In general, this approach uses a change point detection algorithm [⁶⁷] to find changes in the drinking rate using the cumulative consumption throughout the session. If these changes in drinking rate indicated that during the first half of the session the animal drank (1) pharmacologically relevant quantities of alcohol at (2) a faster rate than later in the session, the session was categorized as a frontloading session. If the changes in the drinking rate were not well classified using the change point detection algorithm, the drinking pattern was classified as inconclusive. See [⁶⁰] for full details and example software.