Individual Differences in Conditioned Fear and Extinction in Female Rats

Abstract

The inability to extinguish a traumatic memory is a key aspect of post-traumatic stress disorder (PTSD). While PTSD affects 10-20% of individuals who experience a trauma, women are particularly susceptible to developing the disorder. Despite this notable female vulnerability, few studies have investigated this particular resistance to fear extinction observed in females. Similar to humans, rodent models of Pavlovian fear learning and extinction show a wide range of individual differences in fear learning and extinction, although female rodents are considerably understudied. Therefore, the present study examined individual differences in fear responses, including freezing behavior and ultrasonic vocalizations (USVs), of female Long-Evans rats during acquisition of fear conditioning and cued fear extinction. Similar to prior studies in males, female rats displayed individual variation in freezing during cued fear extinction and were divided into extinction competent (EC) and extinction resistant (ER) phenotypes. Differences in freezing between ER and EC females were accompanied by shifts in rearing during extinction, but no darting was seen in any trial. Freezing behavior during fear learning did not differ between the EC and ER females. Vocalizations emitted in the 22 and 50 kHz ranges during fear learning and extinction were also examined. Unlike vocalizations seen in previous studies in males, very few 22 kHz distress vocalizations were emitted by female rats during fear acquisition and extinction, with no difference between ER and EC groups. Interestingly, all female rats produced significant levels of 50 kHz USVs, and EC females emitted significantly more 50 kHz USVs than ER rats. This difference in 50 kHz USVs was most apparent during initial exposure to the testing environment. These results suggest that like males, female rodents show individual differences in both freezing and USVs during fear extinction, although females appear to vocalize more in the 50 kHz range, especially during initial periods of exposure to the testing environment, and emit very few of the 22 kHz distress calls that are typically observed in males during fear learning or extinction paradigms. Overall, these findings show that female rodents display fear behavior repertoires divergent from males.

Keywords: fear extinction; females; freezing; individual differences; ultrasonic vocalization.

FIGURE 1

Testing protocol is shown in panel (A). Freezing (B) and rearing (C) behaviors in extinction competent (EC) and extinction resistant (ER) rats are shown during fear acquisition, contextual recall, extinction learning, extinction recall, and generalization trials. Shaded area represents the unconditioned period before the tone–shock pairings (fear learning) or the first conditioned/novel tone (extinction learning, extinction recall, and generalization). Freezing behavior is expressed as percent freezing in 1-min bins while rearing is the number of rears during each minute. EC and ER rats showed no differences in freezing (A) during acquisition or context recall, but ER rats displayed higher freezing in response to the CS+ (tone) than EC rats during extinction learning and extinction recall. A time by phenotype interaction in freezing was seen during generalization since ER rats exhibited higher freezing than EC rats, but only during the first novel tone presentation during this trial. Bouts of rearing (C) decreased across time during all trials as demonstrated by main effects of time, since rearing was high during periods before the tone presentation (shaded areas) and decreased as freezing behavior increased. Overall differences in rearing bouts between ER and EC groups were only seen during contextual fear extinction, and generalization. Points represent mean ± SEM, with N = 7 ER and N = 7 EC rats per group. ^#P < 0.05 for main ER–EC difference; *P < 0.05 significant differences at each time point. Tones are denoted by the notes.

FIGURE 2

Summary of behaviors in extinction competent (EC; black bars) and extinction resistant (ER, open bars) female rats during fear learning (A), the last 10 tones during extinction learning (B), and the first 5 tones of extinction recall (C). The left Y axis is for the percent of time spent freezing, grooming, or burying, while the right Y axis is for the number of rearing bouts or microphone interactions. Percent freezing is from automated analysis with FreezeScan, while other behaviors were hand-scored using Observer^® XT. During fear learning (A), both the 3 min prior to conditioning (shaded areas) and the 3 min during tone–shock pairings are shown. Both ER and EC rats display similar increases in freezing alongside a decrease in rearing during the tone–shock pairings phase of fear learning. During both extinction learning (B) and extinction recall (C) ER females froze significantly more than EC rats, and ER rats displayed fewer rears than EC females during this same period. Bars represent mean ± SEM, with N = 7 ER and N = 7 EC rats per group. ^∗P < 0.05.

FIGURE 3

Female rats emitted very few 22 kHz ultrasonic vocalizations (USVs) during fear learning, extinction learning, extinction recall, and generalization trials, and no 22 kHz USVs were seen during contextual fear recall (not shown). In panel (A), the number of 22 kHz USVs in extinction competent (filled squares, EC) and extinction resistant (open circles, ER) rats in each 1-min time bin is shown. Shaded areas represent the unconditioned vocalizations before tone–shock pairings (fear learning) or the first conditioned/novel tone (extinction learning, extinction recall, and generalization). Very few females showed 22 kHz vocalizations, and most of these were during fear learning (tone–shock pairings). EC and ER rats did not differ significantly in the number of 22 kHz calls during any of the trials. Panels (B–D) show that EC and ER phenotypes did not show any differences in the acoustic parameters of 22 kHz USVs, including the average amplitude of each call (B), average duration of each call (C), or average peak frequency of the calls (D) across all trials. Panel (E) shows a representative spectrogram of 22 kHz USVs; the blue boxes identify the detected 22 kHz USVs using UltraVox software. Points and bars represent mean ± SEM, with N = 7 ER and N = 7 EC rats per group.

FIGURE 4

Female rats emitted significant numbers of 50 kHz ultrasonic vocalizations (USVs) during fear learning, extinction learning, extinction recall, and generalization trials, although fewer 50 kHz USVs were seen during contextual fear recall. Panel (A) shows extinction competent (filled squares, EC) females emitted a significantly greater number of 50 kHz USVs compared to extinction resistant (open circles, ER) rats over time. The shaded areas represent the unconditioned vocalizations before tone–shock pairings (fear learning) or the first conditioned/novel tone (extinction learning, extinction recall, and generalization). During acquisition, EC rats emit significantly more 50 kHz USVs than the ER rats predominantly during the first 3 min of exposure to Context A prior to tone–shock pairings; 50 kHz USVs decreased during fear acquisition and the tone–shock pairings. Similar to acquisition, EC rats emit more 50 kHz USVs than ER rats during extinction learning, extinction recall, and generalization trials, and the number of 50 kHz USVs was greatest during periods before the tone presentations and decreased over time in these trials. Panels (B–D) show that acoustic parameters of 50 kHz USVs were similar in EC and ER phenotypes including the average amplitude of each call (B), average duration of each call (C), or average peak frequency of the calls (D) across all trials, with the exception of an ER–EC difference in frequency during generalization. Panel (E) shows a representative spectrogram of 50 kHz USVs; the green boxes identify the detected USVs using UltraVox software. Points and bars represent mean ± SEM, with N = 7 ER and N = 7 EC rats per group. ^#P < 0.05 main effect of phenotype; *P < 0.05 significant phenotypic differences at 1-min bins).

FIGURE 5

Representative ethograms from extinction resistant (ER) and extinction competent (EC) females comparing behaviors and ultrasonic vocalizations (USVs) prior to fear learning (A) and during extinction learning (B). Ethograms in panel (A) show 2 min prior to the tone–shock pairings demonstrating similar numbers of rearing bouts (pink) in the ER and EC rats, but significantly more 50 kHz USVs emitted by the EC female compared to the ER rat during the initial exposure to Context A. Ethograms in panel (B) show freezing (yellow), grooming (purple), burying (light green), rearing (pink), microphone exploration (orange), and 50 kHz USVs (green) in an ER and EC female during 2 min of tone presentation (shaded bars) during extinction learning; this represents tones number 15 and 16 during the last 5 min of extinction learning. Note that the ER female shows more freezing, but less rearing, burying, and 50 kHz USVs, compared to the EC female during this same period of the trial. Panel (C) demonstrates the number of rearing bouts is negatively correlated with the percent freezing during extinction learning. Panel (D) shows the number of 50 kHz USVs during the 3 min prior to fear learning is negatively correlated with the percent freezing during extinction learning and extinction recall, suggesting more 50 kHz USVs during the initial exposure to Context A (prior to any fear conditioning) is associated with better extinction (less freezing) during extinction in females. N = 7 ER and N = 7 EC rats per group. Closed symbols are EC females and open symbols are ER females.