Figure 3.
Female adjustment to reversals in the early phase was more affected by vlOFC inhibition than in males, whereas BLA’s role in probabilistic reversal learning was not sex-dependent. Plotted is the accuracy for the first 150 trials with a sliding window of 10 trials. Learning of deterministic (100/0) (A) and probabilistic (90/10) (B) reversals as measured by probability correct (P(Correct)). Drug order was counterbalanced such that on R2 and R4 (not shown) animals received VEH if they were administered CNO first on R1 and R3 and vice versa. Chemogenetic inhibition of vlOFC lowered P(Correct) in early first deterministic R1 and first probabilistic R3 in females but not males, whereas BLA inhibition attenuated probabilistic reversal learning in both males and females. Bonferroni-corrected post hoc comparisons upon GLM result of sex × drug order interaction resulted in an effect of drug order only in vlOFC females, not in eGFP. The impairing effect of BLA inhibition on learning was not sex-dependent. There were no significant sex differences and no effect of CNO on learning in the eGFP group. **p = 0.01, *p < 0.05. The RL model fit to choice behavior also indicated that chemogenetic inhibition of vlOFC increased the decay rate (decreased memory) of the unchosen option (γd) in female rats (Fig. 5).