Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 1.
Published in final edited form as: Ethology. 2024 Jun 4;130(8):e13487. doi: 10.1111/eth.13487

Familiarity and social relationships in degus (Octodon degus)

Amber Thatcher 1, Nathan Insel 1,2,*
PMCID: PMC11412554  NIHMSID: NIHMS1997695  PMID: 39309638

Abstract

Degus (Octodon degus) are a highly gregarious species of caviomorph rodent native to South America. Kinship does not appear to play a role in degu social structure, and alloparenting is often observed between unrelated females. We hypothesize that female degus readily establish new, cooperative peer relationships. Here we examined changes in dyadic behavior as individuals became more familiar, testing the prediction that interactions between female strangers would quickly resemble those of cagemates. Adult degus underwent a several week series of 20 minute “reunion” social exposures, interleaving reunions with initial strangers and, as a control, familiar cagemates. Males showed initially higher levels of interaction with strangers that converged with cagemate levels over experience. Females could be split into two groups: those that consistently interacted more with strangers (SC-HIGH) and those that did not (SC-LOW); however, unlike males, the higher interaction levels observed between strangers did not change with familiarity. Following 10 reunion sessions female strangers were housed together to create “new cagemates”. Even after co-housing, SC-HIGH (but not SC-LOW) females continued to interact more with the relatively unfamiliar peer than their prior cagemate, particularly in face-to-face and rear-sniffing interactions. A final set of reunions with new strangers found that individual differences in female responses to social novelty were preserved. These results reveal sex differences in the rules relating familiarization to social relationships in degus, and that female predispositions toward cooperation may be due to inherent responses to new individuals more than to how they negotiate relationships over time.

Keywords: degu, familiarization, individual differences, novelty, dominance, kinship

Graphical Abstract

graphic file with name nihms-1997695-f0001.jpg

As strangers become more familiar (exposures, x-axis), interaction levels (y-axis) are expected to become more similar to those of long-term cagemates. In males this was the case (triangular, blue shading) with hints of established dominance. Some females interacted more with strangers (SC-HIGH, solid red), and some not (SC-LOW, dashed red), though neither showed differential changes in interaction levels between stranger and cagemates (red shading)—even after co-housing. With new strangers males were more agonistic, females interactive in other ways.

INTRODUCTION

Cooperative behavior can be found across animal taxa. While evolutionary theories of cooperation have often emphasized the importance of genetic relatedness (Hamilton, 1963, 1964), many species show cooperative social structures in the absence of kin association (e.g., flying squirrels: Garroway et al., 2013; golden-crowned sparrows: Arnberg et al., 2015, tufted duck: Liu et al., 2013). A particularly salient example of this phenomenon can be found in degus (Octodon degus), a small, South American caviomorph rodent that lacks kin-based social organization (Davis et al., 2016; Quirici, Faugeron, Hayes, & Ebensperger, 2011) but also cooperates in the form of communal burrowing (Ebensperger & Bozinovic, 2000) and communal nesting and nursing (Ebensperger et al., 2002, 2004, 2012, 2014; Hayes et al., 2009). Kin-independent cooperative social structures are therefore well established, but the basis of this collective, ecological structure in terms of individual psychology remains largely unknown.

There are a number of qualities that may make an individual or group more likely to exhibit cooperative behavior. Animals differ in their general tendency toward anxiety and aggression, which can predispose them against unfamiliar, unrelated individuals (N. S. Lee, Goodwin, Freitas, & Beery, 2019). But even animals that are initially aggressive can become cooperative through relationship formation. Dyads of male mice, for example, typically form dominance relationships over several social exposures that include a series of fighting bouts (W. Lee, Fu, Bouwman, Farago, & Curley, 2019). Prairie voles, although initially aggressive to unknown individuals, will form lasting bonds with experience (N. S. Lee & Beery, 2021). In other words, cooperation (and affiliation) can depend on experience with the other individual through the formation of a relationship, or may be present independent of experience due to innate or learned, generalized behavioral tendencies. These can be dissociated by tracking animals’ behaviors over time, gauging levels of different interaction types as members of a dyad become more familiar with one-another.

Prior work indicates that female degus interact more with strangers than familiar cagemates over short (5 to 30 minute) time periods (Insel, Shambaugh, & Beery, 2020; Lidhar, Thakur, David, Takehara-Nishiuchi, & Insel, 2021; Thatcher & Insel, 2023; Villavicencio, Márquez, Quispe, & Vásquez, 2009). Over longer (3 hour) time periods, they often choose to huddle with strangers as much as they do cagemates, and spend minimal time in non-social contexts (Insel et al., 2020a). We recently reported that female stranger dyads show unexpectedly consistent behavioral patterns over a series of 20 minute reunions (Thatcher & Insel, 2023). This consistency is incongruous with the idea that animals negotiate their relationship roles over multiple exposures, as has been previously described in male mice (W. Lee et al., 2019). The ways in which degu interactions change with progressive familiarity, including after co-housing, remain largely unknown. This information may help disambiguate the reasons why female degus become more cooperative with unrelated individuals.

In the present study, we quantify the progressive changes in specific behaviors taking place across multiple social exposures in both female and male degus to test the hypothesis that interactions in female strangers rapidly become like those of cagemates. We use an experimental design with several key controls: one that involves more extensive familiarization of peers (“new cagemates”) and the other that assesses peer-specificity with “new strangers”. Contrary to our initial prediction, the data reveal that (a) females differ from one-another in how they respond to new individuals, and that (b) responses to initial strangers may persist even after multiple days of co-housing.

Materials and Methods

Subjects

Subjects were fifty-two adult female degus aged seven to twenty-nine months (mean = 11.9 mo, median = 11 mo) and forty-four male degus aged six to twenty months (mean = 10.4, median = 10). Degus were born and reared in a University of Montana vivarium, weaned between 4 to 8 weeks, and housed in same-sex pairs in 50.8 × 40.6 × 21.6 cm plastic cages. Degus were fed a 1:1 mixure of chinchilla and guinea pig “Teklad” feeds (Envigo; Indianapolis, IN). Animals were housed on a 12:12 h light/dark cycle, with all tests occurring during the light (active) cycle. Housing included items for enrichment (e.g., hay, cardboard enclosures, small nylon bones for chewing, and wooden blocks). Cagemate dyads were same-sex individuals paired at the time of weaning; stranger dyads were the same as those for cagemates approximately matched for age. Strangers inhabited the same vivarium but had never been in physical contact.

Apparatus

Video and audio data were recorded using a Logitech HD Pro webcam C920 USB2 camera recorded at 30 frames/s. The recording chamber was 50 × 50 × 50 cm painted and sealed wood (Figure 1A). Following each social exposure the chamber was cleaned using 70% ethanol then dried.

Figure 1. Chamber and protocol.

Figure 1.

Open in a new tab

A) Recordings were performed in a 50 × 50 cm chamber. B) Prior to each 20 minute social exposure with a cagemate (CAG) or stranger (STR), degus were isolated for 24 hours. Following each reunion, degus were reunited for 24 hours with their cagemate. C) The sequence of CAG and STR reunions was pseudorandomized and counterbalanced across dyads. Following 5 CAG and 5 STR exposures, taking roughly 21 days, a subset of female degus were rehoused with their initial strangers to create new cagemates (STR->NC), first for 1 and 2 days (reunions 11 and 12), then for a full 7 and 8 days (reunions 13 and 14). Two final reunions were performed with new strangers (NST).

Testing

Each animal was pre-exposed to the testing chamber for at least 5 minutes each day over five days. Twenty-four hours before the first testing session, cagemates were separated into individual cages. Immediately prior to testing, degus were transported to the testing room and the backside of one member of each pair was marked for identification (Pet Paint; Camarillo, CA). A pair of either cagemates or strangers was then placed in the chamber for 20 minutes of free interaction. When the session was compete, animals were returned to home cages in the vivarium with their previous cagemate for 24 hours (Figure 1B). Re-pairing cagemates after each reunion session allowed the cagemates to maintain relationships over the weeks of the experiment. This cycle of co-housing, isolation, and testing was repeated, with exposures to strangers and cagemates interleaved in a pseudo-randomized order (alternating partners after either one or two reunions with the same partner) for a total of 5 cagemate and 5 stranger sessions (counterbalancing stranger-first and cagemate-first across dyads; Figure 1C).

Following the 10 cagemate/stranger reunion sessions, the female stranger dyads were co-housed to create “new cagemates”. This was not performed with males to avoid aggression-related injury. Of the 26 female stranger dyads, 24 were co-housed for 24 hours and tested after 24 hours of isolation, then again after a second 48 hour cycle (1 day NC; the first two dyads did not receive this condition due to an experimental error). Initial results after 12 dyads were complete suggested 24 hours may be insufficient to fully familiarize the initial strangers; the experiment was therefore updated to add an additional two reunion sessions after 7 and 8 days of co-housing (7 day NC, n = 14 dyads). To prevent sampling bias, all statistical comparisons that included new cagemates omitted animals that did not receive this control condition.

After each of the above reunion sessions, both male and female degus were returned to co-housing with their original cagemtes and then, following 24 hours of isolation, tested with a new stranger. The “new stranger” reunions provided a control condition to assess whether changes over multiple reunions were due to familiarity with the specific individuals, or due to acclimation to the recording chamber and testing protocol.

Behavioral Scoring

Scoring of social interactions was performed using BORIS (Behavioral Observation Interactive Research Software; (Friard & Gamba, 2016)), which allows users to log events during video playback. The majority of sessions were scored by author AT and the remainder by 7 undergraduate students all directly trained by AT (training involved a student completing 1 to 3 pre-scored training videos with feedback given for each; those with lower scoring alignment to AT were given more pre-scored videos and feedback). Efforts were made to pseudo-randomize sessions to ensure that no scorer worked on videos from only a single condition. All raters were blinded to condition, and scored the start and end times of each observed behavior, the type of behavior, and the animal that initiated the behavior. Although the ethogram comprised 17 individual behaviors, for simplicity and consistency with our prior work (Lidhar et al., 2021), analyses collapsed interactive behavior types into the following categories: agonistic (mounting, biting, wrestling, boxing, marking—i.e., a back leg lift or back-leg “glide” over the other individual—rear-push, tail shaking), allogrooming (sniffing of neck or body with small, repetitive movements suggesting use of mouth and teeth), rear-sniffing (anogenital sniffing), face-to-face (nose-to-nose and nose-to-mouth sniffing/contact), and body sniffing (sniffing toward neck and body). No inactive (stationary) behaviors were included among “interactive behaviors”; however, huddling, in which animals maintained physical contact adjacent to or above one another, was analyzed separately. All analyses were custom-written in MATLAB and are freely available on GitHub.

Measures

All measures were computed from ethogram-based scoring of behavior, performed manually in BORIS. Interaction time (time budget analysis) was computed as a proportion of session time, and was considered both as a sum total across behavioral types (agonistic, face-to-face, etc.), and separately between behaviors.

Prior work has demonstrated that establishment of dominance relationships can be measured through changes in the dyadic asymmetry of agonistic and subordinate behaviors (i.e., the degree to which primarily one animal initiates and engage in agonistic behaviors and the other primarily subordinate behaviors; W. Lee et al., 2019). To analyze dominance relationship changes at minute-resolution time scales, we compute the dyadic asymmetry of agonistic interactions using a simple index of asymmetry:

asymmetry=|DeguA−DeguB|/(DeguA+DeguB))

To evaluate changes in asymmetry after peak agonistic interactions, we identified a one-minute window for each dyad during which the agonistic interaction levels of both peers within a dyad reached a maximum. We then took asymmetry scores for all interactions, across reunion sessions, that took place prior to the 1-minute window (“pre-peak”) and after the one minute window (“post-peak”).

Statistics

Statistical evaluation relied on null hypothesis significance testing (NHST), beginning with inspection of distributions. Examination of total interaction time across sessions revealed a tailed distribution that approximately fit a gamma distribution. This distribution also appeared Gaussian after applying a cubed-root transformation (Supplementary Figure 1; discussion of cubed-root transforms can be found in Raudenbush & Bryk, 1987). To test factors influencing interaction levels we used a generalized linear mixed-effects model (GLMM; fixed factors were sex, stranger/cagemate, session number, with subject treated as a random effect). The GLMM can be run fitting data to a gamma distribution; however, gamma distributions are intolerant of zero values, and zeros (“no interaction”) can become common when focusing analyzes on single interaction types. For consistency, therefore, all statistics were run after performing a cubed-root transformation to normalize data. In cases where zero values were rare (e.g., total interaction time) results were validated against gamma distribution models, with no evidence of meaningful differences between these approaches. To guard against the possibility that the chosen statistical models were inappropriate, we repeated analyses using 3 (or 2) factor, mixed effects ANOVA models with session number and stranger/cagemate both treated as repeated measures. For simplicity, results of ANOVA models are omitted from text except when inconsistent with results from GLMM. When only two groups were being compared, standard Student’s t-tests were used. In several cases, multiple comparisons had to be taken into account and corrected for. This includes tests that separately examined each interaction type (5 types tested). Significance values are corrected using the Benjamini and Hochberg false discovery rate procedure (Benjamini & Hochberg, 1995).

Ethical Note

All protocols were approved by the Institutional Animal Care and Use Committee at University of Montana (protocol number 036–18NIPSYC-061918) and all methods were performed in accordance with IACUC, AALAC, and USDA regulation. Degus were housed with extra space and enrichment and no invasive procedures were performed prior to or during the course of the experiment. Original sample sizes were determined by power analysis performed in G*Power (Faul, Erdfelder, Lang, & Buchner, 2007) and were based on preliminary effect sizes obtained from within-dyad behavioral patterns across conditions in a prior experiment (Lidhar et al., 2021). Following the experiment, degus were transferred to other experimental protocols; when retired from the laboratory, animals were transferred either to other labs, adopted locally, or humanely euthanized, with post-mortem tissue often used for other research.

RESULTS

Social novelty increased interactive behavior

Degus interacted more with strangers than cagemates (n = 52 female, 44 male degus across 5 cagemate and 5 stranger sessions; GLMM effect of stranger/cagemate: β = 0.047, SE = 0.018, t = 2.6, p = 9.1 × 10−3; Figure 2A; this result was also reported in Thatcher & Insel, 2023). Male strangers showed particularly high interaction levels relative to other groups (sex × stranger/cagemate: β = 0.067, SE = 0.026, t = 2.6, p = 0.011). Stranger interactions were detectably higher specifically among agonistic, rear-sniffing, and face-to-face interactions (false discovery rate correction for 5 behaviors; stranger/cagemate agonistic: β = 0.10, SE = 0.023, t = 4.35, p = 1.5 × 10−5; rear-sniffing: β = 0.055, SE = 0.021, t = 2.7, p = 7.6 × 10−3; face-to-face: β = 0.083, SE = 0.017, t = 4.9, p = 1.4 × 10−6). These differences tended to hold for both male and female degus, with a few inconsistencies between statistical approaches: in females, the ANOVA but not GLMM model identified stranger-cagemate differences in rear-sniffing (β = 0.021, SE = 0.017, t = 1.2, p = 0.24; 2-way repeated-measures ANOVA: F(1,51)=12.37, p=9.3×10−4); in males, the GLMM but not ANOVA models showed significant stranger-cagemate differences in face-to-face (GLMM: β = 0.083, SE = 0.019, t = 4.5, p = 1.1 × 10–6; ANOVA: F(1,43) 5.8, p = 0.020) and rear-sniffing (GLMM: β = 0.055, SE = 0.022, t = 2.5, p = 0.014; ANOVA: F(1,43) = 2.4, p = 0.056).

Figure 2. Interaction levels across reunions.

Figure 2.

Open in a new tab

A) Total interaction levels across all types in female (left, red) and male (right, blue) degus. Lighter colors are averages in strangers, darker are cagemates. Only female degus were tested in 1 and 7 day “new cagemate” (NC) conditions; both female and male degus were tested in two final “new stranger” (NST) reunions. While male stranger and cagemate interaction levels converged over sessions, female did not. B) Interaction levels broken down by interaction types (AG = agonistic, GR = grooming, RS = rear-sniffing, BS = body-sniffing, FF = face-to-face). In females, agonistic, rear-sniffing, and face-to-face interactions were consistently higher in strangers than cagemates; in males, only agonistic interactions were consistently higher, though levels of face-to-face were higher in strangers on the first day, and a similar trend was observed for rear-sniffing.

Effects of social familiarization differed between males and females

We predicted that the difference in interaction levels between strangers and cagemates would diminish as initial strangers became more familiar. We detected no overall interaction between stranger/cagemate and exposure number (GLMM, β = -4.3 × 10−3, SE = 0.0054, t = -0.79, p = 0.43; Figure 2B); instead, we observed a 3-way interaction between stranger/cagemate, exposure, and sex (β = -0.016, SE = 7.9 × 10−3, t = 2.6, p = 0.041). This was due to the stranger-cagemate differences decreasing in males, but not females (stranger/cagemate × exposure, males: β = 0.012, SE = 5.1 × 10−3, t = 2.4, p = 0.018; females: β = 3.8 × 10−3, SE = 5.5 × 10−3, t = 0.68, p = 0.50). The effect was particularly strong for male face-to-face interactions (stranger/cagemate × exposure number, male: β = -0.019, SE = 5.6 × 10−3, t = -3.4, p = 8.2 × 10−4, female: β = -5.2 × 10−3, SE = 4.3 × 10−3, t = -1.2, p = 0.23). Though rear-sniffing showed a similar trend (males: β = -0.015, SE = 6.8 × 10−3, t = -2.2, p = 0.026, ANOVA: F(4,172)=2.36, p=0.056; females: β = 2.6 × 10–3, SE = 5.2 × 10−3, t = 0.50, p = 0.61).

To further examine the absence of familiarity effects in females, we subdivided female degus into those that had significantly higher levels of interaction with strangers (“SC-HIGH”, 16 of 52 degus, determined using a paired t-test on each animal across 5 stranger, 5 cagemate sessions, p < 0.1) and those that did not (“SC-LOW”, 36 of 52 degus; distribution of variance and difference scores are illustrated in Supplementary Figure 2A). The reasons for individual variation in stranger responses are unclear, though there was likely some form of cohort effect, as the stranger/cagemate, average discrimination score (i.e., (STR-CAG)/(STR+CAG)) varied significantly across litters (1-way ANOVA, F(20,31) = 1.9, p = 0.049), but not based on parents (F(8,43) = 1.3, p = 0.25), weight (Pearson correlation: r = 0.11, p = 0.44) or litter size after limiting the analysis to single individuals within a litter (r = 0.25, p = 0.28). There was no evidence that stranger vs. cagemate interaction levels changed across exposures in either the SC-HIGH animals (GLMM, stranger/cagemate: β = 0.15, SE = 0.036, t = 4.2, p = 4.0 × 10−5, exposure number: β = 9.6 × 10–3, SE = 6.7 × 10–3, t = 1.4, p = 0.16; stranger/cagemate × exposure: β = -0.014, SE = 9.5 × 10–3, t = -1.4, p = 0.15) or the SC-LOW animals (stranger/cagemate: β = 2.8 × 10–4, SE = 0.027, t = 0.011, p = 0.99; exposure number: β = -5.5 × 10–3, SE = 4.4 × 10–3, t = -1.24, p = 0.21; stranger/cagemate × exposure: β = -1.6 × 10–5, SE = 6.2 × 10–3, t = -2.6 × 10–3, p > 0.99; Figure 3A & Supplementary Figure 2B).

Figure 3. Individual differences in female responses to strangers.

Figure 3.

Open in a new tab

A) Female degus could be split into those that interact more with strangers (SC-HIGH, upper plot) and those that do not (SC-LOW, lower). All “D” values refer to Cohen’s D when compared against original cagemates; from left to right: stranger, 1–2 day “new cagemates” (NC; i.e., former strangers after 1–2 days of cohousing, 7–8 day NC), new strangers (NST; i.e., new, unfamiliar peers). No significant changes in the stranger-cagemate differences were observed across exposure numbers. B) Breakdown of stranger vs. cagemate interaction levels among SC-HIGH females. Higher levels of face-to-face persisted in SC-HIGH animals even after co-housing. C) Distribution of relative interaction levels across all SC-HIGH (filled circles) and SC-LOW (open circles) females. Interaction levels with cagemates were highly variable, and therefore never statistically exceeded those of strangers, while approximately 1/3 of females did consistently interact more with strangers. Black arrows denote animals from the two non-sibling cagemate pairs, animals from one pair were classified as SC-HIGH and from the other as SC-LOW animals.

Co-housing female dyads was insufficient to make new relationships resemble original cagemates

A possible explanation for the absence of change in females is that the 20 min reunion sessions provided insufficient time for partners to become fully familiarized. To test this, a subgroup of 40 female degus were co-housed with their stranger partners for 1 day creating “new cagemates”, then tested in two more reunions. A smaller subgroup was further co-housed an additional 7 days and tested twice again. Unexpectedly, SC-HIGH degus continued to interact more with the new cagemates relative to their prior, long-term cagemates after 1 to 2 day (paired t-test, averaging together all cagemate and both new cagemate sessions, t9=5.50, p=3.80×10−4) or 7 to 8 days of co-housing (t7=2.98, p=0.021; “NC” columns in Figure 3A). Specifically, SC-HIGH females had higher rear-sniffing and face-to-face levels after 1 to 2 days of cohousing (RS: t9 = 3.2, p = 0.011; FF: t9 = 5.0, p = 7.2 × 10−4; false-discovery correction); and a trend for higher face-to-face levels (t7 = 3.1, p = 0.017) with comparable effect sizes after 7 to 8 days of co-housing (Figure 3B & Supplementary Figure 2C). Unfortunately, the influence of genetic relatedness in the higher interaction levels of new relationships could not be evaluated statistically because only two of the cagemate dyads were non-siblings; however, it should be noted that both members of one, non-sibling cagemate dyad interacted more with strangers (i.e., they were also SC-HIGH animals) while the members of the other dyad did not (were SC-LOW animals; black arrows in Figure 3C).

Changes in stranger-cagemate differences in males could be attributed to peer familiarization

To control for non-specific effects influencing stranger-cagemate differences over exposures, degus were exposed to a new, unfamiliar conspecific (“new stranger”). In both males and SC-HIGH females, levels of interaction with new strangers were higher than with cagemates (paired t-test between within-animal averages in cagemates and new-strangers, males: t43=5.1, p=6.7 × 10−6, SC-HIGH females: t15=4.2, p= 8.2×10−4) and were not lower than interactions with the original strangers (males: t43=1.8, p=0.086 with new stranger > old stranger; SC-HIGH females: t15=1.1, p=0.28; Figure 2A, right; Figure 3A, top). In particular, SC-HIGH females showed higher levels of face-to-face (t7 = 5.0, p = 9.4 × 10−4) with additional trends for higher agonistic, rear-sniffing, and body sniffing (p < 0.06; Figure 3B). In males, it was the agonistic interactions that were particularly increased with new strangers, exceeding both the original cagemates (t43 = 5.5, p = 1.8 × 10−6) as well as that of original strangers (t43 = 3.3, p = 0.0019). SC-LOW females did not interact more with new strangers than cagemates, or compared with original strangers (p >> 0.1).

Within-session analyses revealed possible sex difference in dominance relationship formation

Further structure in dyadic interactions could be revealed at sub-session time scales. For example, face-to-face interactions tended to be concentrated in the early part of each session, and peak agonistic interactions varied across individuals (Figure 4A & B). We predicted that agonistic asymmetry would increase in males but not females following a period of peak agonistic interaction. As a negative control we first tested changes in cagemates, which likely have long-standing dominance relationships that would not be expected to change during experimental exposures. We found that female cagemates showed more agonistic asymmetry than males, but that there was little change before compared with after a period of peak agonistic activity (2-way ANOVA sex: F(1,25) = 4.7, p = 0.040; pre vs. post: F(1,25) = 0.041, p = 0.84; sex × pre/post: F(1,25) = 0.091, p = 0.77; Figure 4C, top panel; note: dyads without agonistic interactions were removed from analysis). This suggests that degu male cagemates continued to test one-another (e.g., by mounting) in the experimental chamber, while agonistic interactions between females were typically rare and more unilateral. Among strangers, no sex effects were observed, but there was a statistical trend in the predicted direction, with increased agonistic asymmetry (suggesting a strengthening of dominance roles) in males but not females following peak agonistic periods (sex: F(1,31) = ~0.0, p = ~1.0; pre vs. post: F(1,31) = 0.63, p = 0.43; sex × pre/post: F(1,31) = 3.3, p = 0.077; Figure 4C, bottom panel).

Figure 4. Interaction levels over seconds and minutes.

Figure 4.

Open in a new tab

A) Colormaps of interaction types for each degu in 2 s bins across the five exposures. Individual degus (rows) are sorted according to when peak agonistic levels took place (AG, red; other abbreviations: GR = allogrooming, RS = rear-sniffing, BS = body sniffing, FF = face-to-face). Higher concentration of face-to-face at session starts helps demarcate each of the 5 reunion sessions. B) Colormaps showing levels of agonistic activity across each minute, for each degu, sorted according to when peak levels took place (top: female, bottom: male). C) Agonistic asymmetry in male and female cagemate (top panel) and stranger (bottom panel) dyads before (“pre”) compared with after (“post”) the dyad’s 1-minute period of peak agonistic activity. A statistical trend was observed for increased asymmetry in males strangers after agonistic bouts, hinting at the predicted strengthening of dominance roles.

Huddling was highly variable and not clearly different between groups

Huddling, i.e., periods of relative inactivity during which degus remained in direct physical contact, was found to be extremely variables across the 20 minute sessions, with over 50% of sessions involving no huddling (54% among females, 64% among males) and the remaining sessions showing an extremely tailed distribution, where huddling in a few sessions occupied nearly the full 20 minutes (Figure 5A). Consistent with huddling as an inactive alternative to interaction, there was an inverse relationship between huddle time and interaction time—though also notably high variance (Pearson correlation, ignoring all sessions with zero huddling; female: r = -0.36, p = 3.1 × 10−13; male: r = -0.28, p = 2.4 × 10−7; Figure 5B). No differences could be found in levels of huddling between stranger and cagemate groups for either female or male degus (GLMM, p >> 0.05 for stranger/cagemate, exposure number, and stranger/cagemate × exposure, Figure 5C). Nor were stranger-cagemate differences further revealed when subdividing females into SC-HIGH and SC-LOW groups.

Figure 5. Effects of social novelty on huddling behavior.

Figure 5.

Open in a new tab

A) Histogram showing percentage of sessions in each sex and stranger/cagemate category associated with varying amounts of huddling. Many sessions lacked huddling (zero values), with otherwise high variation and no obvious differences between groups. B) Scatterplot of normalized, total interaction time against normalized huddle time. Both male and female sessions showed a negative correlation of huddling with interaction levels, though both also showed a great deal of variance around the slopes. C) Huddle time averaged across animals within each female (left) and male (right) groups, formatted as in Figure 2A. No statistical differences were observed in either condition or sex.

DISCUSSION

The present work offers a detailed account of how degus change their behavior with one-another as they become more familiar. This is one of very few systematic investigations examining the effects of repeated and extensive social familiarization on interactive behavior, placing a spotlight on a species whose members are known to live and cooperate with unrelated peers. The experiment employed multiple controls, including interleaved exposures to cagemate partners to control for familiarization with the experimental apparatus and protocols, as well as use of “new strangers” and “new cagemates”. We found that in males, stranger-cagemate differences decreased with familiarity—as anticipated. Female behavioral responses were relatively more complex. Some females didn’t interact more with strangers at all—even during an initial exposure. But those that did (typically consistently, across stranger peers) continued treating new relationships differently from long-established relationships, even after extensive familiarization. The evidence points toward females being more willing to interact with new individuals without conflict, rather than rapidly forming relationships that resemble those of long-term partners. To put this another way: if cooperative behaviors can arise either through relationship formation (experience-dependent mechanisms) or through a predisposition toward cooperation (experience-independent mechanisms); the present results suggests that degu females lean heavily toward the latter.

Higher levels of interaction between adult female, stranger degus has been reported previously (Insel et al., 2020; Lidhar et al., 2021; Villavicencio et al., 2009), and fits with the well-known tendency of animals to approach novelty (Berlyne, 1966; Hills, Todd, Lazer, Redish, & Couzin, 2015; Hughes, 1997). In the wild, degus also tend to be gregarious, with burrows found near one-another, and animals from different burrows often greeting one-another (Ebensperger et al., 2004; Fulk, 1976). In this experiment, increased levels of rear-sniffing and face-to-face behaviors between strangers of the SC-HIGH group also tended to remain higher after co-housing. In spite of gregariousness and tolerance for new individuals, the data are consistent with the possibility that new relationships formed in adulthood remain fundamentally different from peers (including siblings) reared together. While the generalizability of the observation is still unknown, and was only made visible by focusing on a subset of animals with higher, relative stranger-cagemate differences, there are some hints from prior studies that degus exhibit “in-group/out-group” differences that could depend on longer relationship histories or genetic relatedness. In the wild, female degus show preferences for members of their own burrow groups during daytime foraging and social greetings (Ebensperger et al., 2004), and in laboratory studies, genetic relatedness can influence social interactions—just not as strongly as familiarity (Villavicencio et al., 2009). It may be that either kinship, exposure during early developmental periods, or familiarity on much longer timeframes help establish long-term bonds in females that differ in nature from new relationships.

Unlike females, social behavior between male strangers did become more similar to those of cagemates across exposures. This convergence was most apparent in face-to-face and rear-sniffing behaviors (Figure 2B). Importantly, agonistic levels between strangers seemed to initially increase from the first exposure, relative to cagemates, and did not converge to become like cagemates within the 5 sessions. We conducted a more detailed analysis by examining interaction behaviors over seconds and minutes within the reunion sessions. Consistent with patterns observed in mice (W. Lee et al., 2019) there was a statistical trend for male but not female peers to increase agonistic asymmetry (i.e., more defined dominance roles) after a period of higher agonistic interaction. Although the tests were not statistically significant, the trend can be considered alongside other observed sex differences in familiarization effects, overall suggestive that coordinated behavior in males depends more on prior interactive experience between the individuals than it does in females.

The present experiment not only replicated higher interaction levels between degu strangers, but also an absence of differential huddling between cagemates and strangers (Beery & Shambaugh, 2021; Insel et al., 2020). Huddling was highly variable between sessions. In contrast with interaction levels, huddling between male degus—and even male strangers—was not clearly different from familiar male or female degus. The present study was not designed to tease apart the circumstances during which degus huddle, but the data do raise questions about what motivates huddling between degu peers. Anecdotally, huddling did not always take the same shape: sometimes animals faced the same direction, sometimes opposite directions, and in other cases one would rest itself fully on top of the other. Examining these variations in future studies may help reveal the social significance of huddling and further tease apart the effects of social novelty on degu behavior.

The key insight offered by the results is that a propensity for pro-social interaction with unrelated individuals does not mean a species quickly forms sibling-like relationships and bonds. While female degus are motivated to interact and huddle with initial strangers, behavior between new peers can remain distinct from long-established relationships for indefinite time periods. This adds important knowledge toward understanding prosocial, cooperative behavior in the wild, and raises new questions about the spectrum of psychological factors that support coherent, collective behavior within groups and societies.

Supplementary Material

Supinfo

Acknowledgements

Experiment was supported by NIH grant 1R15MH117611-01A1. We would like to thank Stephen Cooke, Kendra Kuehn, and Kendall Butler for help scoring behavior, Kinsey Webb, Janelle Shamp, and Dani Crandell, for help with the degu colony.

Footnotes

Competing interest statement: authors declare that they have no competing interests with this work.

Data availability statement

All raw videos are made freely and openly available on Databrary (https://nyu.databrary.org/volume/1297). All code is made freely and openly available on GitHub (https://github.com/ninsel/repeat_reunion). Processed data and spreadsheets are available through author request through UM Box.

REFERENCES

  1. Arnberg NN, Shizuka D, Chaine AS, & Lyon BE (2015). Social network structure in wintering golden‐crowned sparrows is not correlated with kinship. Molecular Ecology, 24(19), 5034–5044. doi: 10.1111/mec.13366 [DOI] [PubMed] [Google Scholar]
  2. Beery AK, & Shambaugh KL (2021). Comparative Assessment of Familiarity/Novelty Preferences in Rodents. Frontiers in Behavioral Neuroscience, 15, 648830. doi: 10.3389/fnbeh.2021.648830 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Benjamini Y, & Hochberg Y (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289–300. doi: 10.1111/j.2517-6161.1995.tb02031.x [DOI] [Google Scholar]
  4. Berlyne DE (1966). Curiosity and Exploration. Science, 153(3731), 25–33. doi: 10.1126/science.153.3731.25 [DOI] [PubMed] [Google Scholar]
  5. Davis GT, Vásquez RA, Poulin E, Oda E, Bazán-León EA, Ebensperger LA, & Hayes LD (2016). Octodon degus kin and social structure. Journal of Mammalogy, 97(2), 361–372. doi: 10.1093/jmammal/gyv182 [DOI] [Google Scholar]
  6. Ebensperger LA, & Bozinovic F (2000). Communal burrowing in the hystricognath rodent, Octodon degus: A benefit of sociality? Behavioral Ecology and Sociobiology, 47(5), 365–369. doi: 10.1007/s002650050678 [DOI] [Google Scholar]
  7. Ebensperger LA, Hurtado M, Soto-Gamboa M, Lacey EileenA., & Chang AnnT. (2004). Communal nesting and kinship in degus (Octodon degus). Naturwissenschaften, 91(8). doi: 10.1007/s00114-004-0545-5 [DOI] [PubMed] [Google Scholar]
  8. Ebensperger LA, Sobrero R, Quirici V, Castro RA, Tolhuysen LO, Vargas F, … Hayes LD (2012). Ecological drivers of group living in two populations of the communally rearing rodent, Octodon degus. Behavioral Ecology and Sociobiology, 66(2), 261–274. doi: 10.1007/s00265-011-1274-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ebensperger LA, Veloso C, & Wallem P (2002). Do female degus communally nest and nurse their pups? Journal of Ethology, 20(2), 143–146. doi: 10.1007/s10164-002-0063-x [DOI] [Google Scholar]
  10. Ebensperger LA, Villegas Á, Abades S, & Hayes LD (2014). Mean ecological conditions modulate the effects of group living and communal rearing on offspring production and survival. Behavioral Ecology, 25(4), 862–870. doi: 10.1093/beheco/aru061 [DOI] [Google Scholar]
  11. Faul F, Erdfelder E, Lang A-G, & Buchner A (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. doi: 10.3758/BF03193146 [DOI] [PubMed] [Google Scholar]
  12. Friard O, & Gamba M (2016). BORIS: A free, versatile open‐source event‐logging software for video/audio coding and live observations. Methods in Ecology and Evolution, 7(11), 1325–1330. doi: 10.1111/2041-210X.12584 [DOI] [Google Scholar]
  13. Fulk GW (1976). Notes on the Activity, Reproduction, and Social Behavior of Octodon degus. Journal of Mammalogy, 57(3), 495–505. doi: 10.2307/1379298 [DOI] [Google Scholar]
  14. Garroway CJ, Bowman J, & Wilson PJ (2013). Complex social structure of southern flying squirrels is related to spatial proximity but not kinship. Behavioral Ecology and Sociobiology, 67(1), 113–122. doi: 10.1007/s00265-012-1431-3 [DOI] [Google Scholar]
  15. Hamilton WD (1963). The Evolution of Altruistic Behavior. The American Naturalist, 97(896), 354–356. doi: 10.1086/497114 [DOI] [Google Scholar]
  16. Hamilton WD (1964). The genetical evolution of social behaviour. II. Journal of Theoretical Biology, 7(1), 17–52. doi: 10.1016/0022-5193(64)90039-6 [DOI] [PubMed] [Google Scholar]
  17. Hayes LD, Chesh AS, Castro RA, Tolhuysen LO, Burger JR, Bhattacharjee J, & Ebensperger LA (2009). Fitness consequences of group living in the degu Octodon degus, a plural breeder rodent with communal care. Animal Behaviour, 78(1), 131–139. doi: 10.1016/j.anbehav.2009.03.022 [DOI] [Google Scholar]
  18. Hills TT, Todd PM, Lazer D, Redish AD, & Couzin ID (2015). Exploration versus exploitation in space, mind, and society. Trends in Cognitive Sciences, 19(1), 46–54. doi: 10.1016/j.tics.2014.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hughes RN (1997). Intrinsic exploration in animals: Motives and measurement. Behavioural Processes, 41(3), 213–226. doi: 10.1016/S0376-6357(97)00055-7 [DOI] [PubMed] [Google Scholar]
  20. Insel N, Shambaugh KL, & Beery AK (2020). Female degus show high sociality but no preference for familiar peers. Behavioural Processes, 174, 104102. doi: 10.1016/j.beproc.2020.104102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lee NS, & Beery AK (2021). The role of dopamine signaling in prairie vole peer relationships. Hormones and Behavior, 127, 104876. doi: 10.1016/j.yhbeh.2020.104876 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lee NS, Goodwin NL, Freitas KE, & Beery AK (2019). Affiliation, Aggression, and Selectivity of Peer Relationships in Meadow and Prairie Voles. Frontiers in Behavioral Neuroscience, 13, 52. doi: 10.3389/fnbeh.2019.00052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee W, Fu J, Bouwman N, Farago P, & Curley JP (2019). Temporal microstructure of dyadic social behavior during relationship formation in mice. PLOS ONE, 14(12), e0220596. doi: 10.1371/journal.pone.0220596 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lidhar NK, Thakur A, David A-J, Takehara-Nishiuchi K, & Insel N (2021). Multiple dimensions of social motivation in adult female degus. PLOS ONE, 16(4), e0250219. doi: 10.1371/journal.pone.0250219 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu Y, Keller I, & Heckel G (2013). Temporal genetic structure and relatedness in the Tufted Duck Aythya fuligula suggests limited kin association in winter. Ibis, 155(3), 499–507. doi: 10.1111/ibi.12059 [DOI] [Google Scholar]
  26. Quirici V, Faugeron S, Hayes LD, & Ebensperger LA (2011). Absence of kin structure in a population of the group-living rodent Octodon degus. Behavioral Ecology, 22(2), 248–254. doi: 10.1093/beheco/arq196 [DOI] [Google Scholar]
  27. Raudenbush SW, & Bryk AS (1987). Examining Correlates of Diversity. Journal of Educational Statistics, 12(3), 241. doi: 10.2307/1164686 [DOI] [Google Scholar]
  28. Thatcher A, & Insel N (2023). In female degus, reunions are less variable when relationships are new. Behaviour, 1–28. doi: 10.1163/1568539X-bja10248 [DOI] [Google Scholar]
  29. Villavicencio CP, Márquez IN, Quispe R, & Vásquez RA (2009). Familiarity and phenotypic similarity influence kin discrimination in the social rodent Octodon degus. Animal Behaviour, 78(2), 377–384. doi: 10.1016/j.anbehav.2009.04.026 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supinfo
NIHMS1997695-supplement-Supinfo.docx (153.7KB, docx)

Data Availability Statement

All raw videos are made freely and openly available on Databrary (https://nyu.databrary.org/volume/1297). All code is made freely and openly available on GitHub (https://github.com/ninsel/repeat_reunion). Processed data and spreadsheets are available through author request through UM Box.

RESOURCES