Homecage two-bottle choice test

In 14 homecage test sessions, we gave water restricted rats 45 minutes homecage access to two bottles; one containing SQ or SUCRA and the other containing purified water. We presented the solutions in order of ascending concentration. We reduced location bias by giving each concentration twice, in two consecutive daily sessions that reversed the position of the bittersweet solution. We calculated individual preference scores by dividing the amount of bittersweet solution consumed by the total volume of fluid consumed during the test session (mL bittersweet solution/ (mL bittersweet solution + mL water)). To have a single preference score for each concentration, we averaged preference scores from the two consecutive daily sessions at each concentration. We determined preferring and avoiding groups by averaging preference scores at the 5^th and 6^th concentrations for each solution (SQ 0.5 mM and 1 mM, SUCRA 2.5 mM and 5 mM), which were based on prior studies examining bittersweet preference [11]. An individual rat was defined as preferring or avoiding independently for SQ and SUCRA. Preferring composite preference scores ranged from 0.60 to 1, while avoiding preference scores ranged from 0 to 0.40. Intermediates had a preference between 0.41 to 0.59.

The experimental timeline is presented in Fig 1. We gave two rounds of homecage preference testing, counterbalanced for order of bittersweet solution exposure. In round 1 of homecage preference testing, we gave half of the rats (n = 18) two bottle choice of water and ascending concentrations of SQ, while we gave the other half of the rats (n = 18) two bottle choice of water and ascending concentrations of SUCRA. In round 2 of homecage preference testing, we gave rats two bottle choice of water and ascending concentrations of the alternate solution that they had not experienced in round 1. After these two rounds, we gave rats two retest sessions to determine whether their preference for the round 1 solution remained stable. During these two retest sessions, we gave rats two bottle choice of water and C5 and C6 (counterbalanced) of the round 1 solution. The retest preference scores were calculated in the same manner as the preference score during initial testing. In addition, we retested rats to determine whether their preference for the round 2 solution remained stable. Both retest rounds occurred between 24–44 days after the last exposure to that tastant. Rats received five sessions of Pavlovian lever autoshaping as previously reported [12] (between the 1^st and 2^nd retest, but data from this phase is not presented here. At the end of the experiment rats were deeply anesthetized with isoflurane (~90 s) and perfused transcardially with 100 ml of 0.1 M PBS with 1.25% sodium nitrite, followed by 400 ml of 4% para-formaldehyde in 0.1 M sodium phosphate, pH 7.4.

We gave two rounds (Rd 1 and Rd 2) of homecage preference testing, counterbalanced for order of bittersweet solution exposure. In round 1 of homecage preference testing, we gave half of the rats (n = 18) two bottle choice of water and ascending concentrations of SQ, while we gave the other half of the rats (n = 18) two bottle choice of water and ascending concentrations of SUCRA. In round 2, we gave rats two bottle choice of water and ascending concentrations of the alternate solution that they had not experienced in round 1. After these two rounds, we gave rats two retest sessions to determine whether their preference for the round 1 and round 2 solutions remained stable. For the Retest 1 sessions, we gave rats two bottle choice of water and C5 and C6 (counterbalanced) of the round 1 solutions, and for Retest 2, we gave rats two bottle choice of water and C5 and C6 (counterbalanced) of the round 2 solutions. Rats were tested in Pavlovian lever autoshaping between Retest 1 and Retest 2 (data not presented).