Neuronal, affective, and sensory correlates of targeted helping behavior in male and female Sprague Dawley rats

Abstract

Introduction: Empathic behaviors are driven by the ability to understand the emotional states of others along with the motivation to improve it. Evidence points towards forms of empathy, like targeted helping, in many species including rats. There are several variables that may modulate targeted helping, including sex, sensory modalities, and activity of multiple neural substrates.

Methods: Using a model of social contact-independent targeted helping, we first tested whether sex differences exist in helping behavior. Next, we explored sex differences in sensory and affective signaling, including direct visualization and an analysis of ultrasonic vocalizations made between animal pairs. Finally, we examined the neural activity in males and females of multiple regions of interest across time. Here, we aim to examine any behavioral differences in our lab's social contact independent targeted helping task between males and females.

Results and discussion: These findings are the first to intimate that, like other prosocial behaviors, males and females may exhibit similar social-independent targeted helping behavior, but the underlying sensory communication in males and females may differ. In addition, this is the first set of experiments that explore the neural correlates of social-independent targeted helping in both males and females. These results lay the groundwork for future studies to explore the similarities and differences that drive targeted helping in both sexes.

Keywords: c-fos; empathy; perception action model; sex differences; targeted helping; ultrasonic vocalizations.

Figure 1

Schematic of variables tested and experimental timeline. A total of 33 pairs of rats/sex were used in Experiments 1–5. (A) Size-matched, same-sex males and freely cycling females were used. Social contact-dependent (Exp. 1) or -independent (Exp. 2a) targeted helping was compared between males and females. Further, the impact of direct visualization between Target and Observer (Exp. 2b) and the differences in ultrasonic vocalizations (USV, Exp. 2c) were studied during social contact-independent targeted helping. Finally, subsets of animals were sacrificed, and neural activity within regions of interest was evaluated via immunohistochemistry of the immediate early gene c-fos (Exp. 2d). (B) Timeline for all behavioral evaluations. One rat in a cage pair was randomly assigned to be the Target or Observer. During acquisition, Observers were placed in the dry side of the chamber and given the opportunity to release the Target from the water chamber. Data were collected during three timepoints: early acquisition (EA, orange), as an indication of an initial helping response; and late acquisition (LA, green) to examine the effect of habituation to aiding a familiar conspecific.

Figure 2

Males and females readily release a conspecific when social interaction is possible. (A) Performance of male and female (n = 4 pairs/sex) rats during the helping task where social contact is possible. (B) Chain pull latencies for males and females during acquisition did not differ; latencies decreased over time, with days 4–10 significantly faster compared to day 1. (C,D) Unpaired t-tests comparing chain pull latencies during early (EA) (C) and late (LA) (D) acquisition did not show a difference between males and females. Error bars represent ± SEM. *Significant difference from day 1, p < 0.05.

Figure 3

Elucidation of sex differences during targeted helping. (A) Performance of male and female (n = 21 pairs/sex) rats during the social contact-independent targeted helping task revealed the Observers’ latency to release a distressed partner decreased over 10 days. Significantly shorter latencies occurred on days 3–10 compared to day 1. (B) A main effect of sex was also seen in acquisition, with male latencies being faster compared to females. (C,D) Unpaired t-tests comparing chain pull latencies during early (EA) (C) and late (LA) (D) acquisition did not show a difference between males and females. Error bars represent ± SEM.*Significant difference from day 1, p < 0.05. #Significant difference between males and females, p < 0.05.

Figure 4

Sex differences in helping behavior when visualization of conspecific is prevented. (A) Performance of male and female (n = 8 pairs/sex) rats during the social contact-independent targeted helping task when sight of the Target was obstructed. (B) Male and female chain pull latencies decreased over time, with days 4–10 significantly faster compared to day 1. There was a strong trend toward a main effect of sex, but it did not reach significance (p = 0.0589). (C,D) Unpaired t-tests were performed comparing the average latencies for early (EA) and late (LA) acquisition between males and females. (B) Female latencies were significantly faster compared to males in EA. (D) During LA, however, there was no sex difference in chain pull latency. Error bars represent ± SEM. *Significant difference from day 1, p < 0.05. #Significant difference between males and females, p < 0.05.

Figure 5

Comparison of USV frequencies during targeted helping between males and females. (A) Experimental timeline for helping behavior. Ultrasonic vocalizations (USV) were recorded during early (EA) and late (LA) acquisition, as well as reversal (Rev) as a proxy for the rats’ affective state during the task. USV were recorded and categorized based on frequency into two groups, “distress” (18–35 kHz) or “prosocial” (>35 kHz). (B) An example of a distress and prosocial call shown in DeepSqueak software. (C–E) USV analysis for early acquisition. (C) Frequency of distribution graph of all calls during EA indicates a bimodal distribution of call frequencies, roughly corresponding to the “distress” and “prosocial” ranges, in all groups. (D,E) An analysis of calls broken into distress or prosocial range for each group as measured by a percentage of total calls made. (D) Female Targets make a significantly greater percentage of distress calls in EA compared to all other groups. (E) Correspondingly, calls in the prosocial range make up a significantly smaller percentage of female Targets’ total calls compared to the other groups. (F–H) USV analysis during late acquisition. (F) Bimodal distribution of calls is seen in frequency of distribution graph of all calls during LA, however, fewer calls are made in the distress range. (G,H) No differences were seen between groups in % of calls made in distress (G) or prosocial (H) ranges. Error bars represent ± SEM. *p < 0.05.

Figure 6

Correlation between release latency and Fos activity during early acquisition. In order to determine a relationship between helping behavior and a neurobiological endpoint, correlational analysis was performed between the last latency and average Fos + cell counts for male (data points in black) and female (red) Observers during early acquisition in the following regions of interest: anterior insula (A), prelimbic (B), infralimbic (C), anterior cingulate (D), and orbitofrontal (E) cortices, paraventricular nucleus of the thalamus (PVT) (F), central amygdala (G), basolateral amygdala (H), and lateral habenula (I). Significant negative correlations were found between final latency and mean Fos count in the five cortical regions analyzed. In contrast, a positive correlation was observed in the PVT. *Significant correlation, p < 0.05.

Figure 7

Correlation between release latency and Fos activity during late acquisition. In order to determine a relationship between helping behavior and a neurobiological endpoint, correlational analysis was performed between the last latency and average Fos + cell counts for male (data points in black) and female (red) Observers during late acquisition in the following regions of interest: anterior insula (A), prelimbic (B), infralimbic (C), anterior cingulate (D), and orbitofrontal (E) cortices, paraventricular nucleus of the thalamus (PVT) (F), central amygdala (G), basolateral amygdala (H), and lateral habenula (I). Significant positive correlations were found in the anterior cingulate and orbitofrontal cortices, with strong trends also seen in the anterior insula and prelimbic cortex. *Significant correlation, p < 0.05.

Figure 8

Differences in total Fos + cells across neural substrates of interest. Total Fos + cells were quantified in male and female rats that performed the targeted helping task (Behaving, BEH) or were left in their homecage on test day (HCC). Rats (4–6 groups/sex) were sacrificed at two different timepoints; the second day of acquisition (EA), or the final day of acquisition (LA). Fos + counts were compared in each timepoint. (A) This figure depicts total Fos + cell counts for male and female HCC and Beh rats during EA. (B) Representative heat map depicting Fos + cell means in each region of interest for each group. (C,D) Fos activity in neural substrates during LA (C), with the heat map depicting mean activity in each region across the four groups (D). Error bars represent ± SEM. *Significant difference between males and females, p < 0.05. +Significant difference between BEH and HCC within the same sex, p < 0.05.