Save

Sociosexual behaviour in wild chimpanzees occurs in variable contexts and is frequent between same-sex partners

In: Behaviour
View More View Less
  • 1 Department of Anthropology, University of Texas at Austin, 2201 Speedway, Austin, TX 78712, USA
  • | 2 Department of Anthropology, University of Michigan, Ann Arbor, MI 48109, USA
Full Access

Abstract

Many animals engage in sociosexual behaviour, including that between same-sex pairs. Bonobos (Pan paniscus) are famous for their sociosexual behaviour, but chimpanzees (Pan troglodytes) apparently do not engage in sociosexual behaviour frequently. However, sociosexual behaviour in chimpanzees may have been overlooked. We observed 584 instances of sociosexual behaviour in chimpanzees at Ngogo, Kibale National Park, Uganda during three years of study. All ages and sexes engaged in sociosexual behaviour, which included mounting, touching of genitals, and pressing genitals together. Most sociosexual behaviour was between adult males. Sociosexual behaviour was often during tense contexts, such as subgroup reunions and during territorial behaviour. Among males, grooming and dominance rank relationships do not explain patterns of sociosexual behaviour. Although sociosexual behaviour may be less frequent in chimpanzees than in bonobos, and bonobos remain distinct in their genito-genital rubbing, our findings suggest that sociosexual behaviour is a regular part of chimpanzee behaviour.

Abstract

Many animals engage in sociosexual behaviour, including that between same-sex pairs. Bonobos (Pan paniscus) are famous for their sociosexual behaviour, but chimpanzees (Pan troglodytes) apparently do not engage in sociosexual behaviour frequently. However, sociosexual behaviour in chimpanzees may have been overlooked. We observed 584 instances of sociosexual behaviour in chimpanzees at Ngogo, Kibale National Park, Uganda during three years of study. All ages and sexes engaged in sociosexual behaviour, which included mounting, touching of genitals, and pressing genitals together. Most sociosexual behaviour was between adult males. Sociosexual behaviour was often during tense contexts, such as subgroup reunions and during territorial behaviour. Among males, grooming and dominance rank relationships do not explain patterns of sociosexual behaviour. Although sociosexual behaviour may be less frequent in chimpanzees than in bonobos, and bonobos remain distinct in their genito-genital rubbing, our findings suggest that sociosexual behaviour is a regular part of chimpanzee behaviour.

1. Introduction

Sometimes it appears that animals are mating but they are not. Members of the same sex may engage in behaviours resembling a copulation, such as mounting (e.g., horned flour beetle, Gnatocerus cornutus (Lane et al., 2016); swamphen birds, Porphyrio porphyrio (Jamieson & Craig, 1987)). Since same-sex mounting cannot lead to reproduction (save for rare cases, such as in parthenogenetic whiptail lizards, Cnemidophorus uniparens (Crews & Fitzgerald, 1980)), it likely serves some other social function. Such mounting, along with other forms of non-conceptive genital contact within and between sexes, are considered ‘sociosexual’ behaviour.

Sociosexual behaviour occurs in a wide range of animal species, including between individuals of the same sex (Vasey, 1995; Sommer & Vasey, 2006; Bailey & Zuk, 2009). Same-sex behaviour occurs in bird species that typically form heterosexual pair bonds: same-sex pair bonds emerge in zebra finches (Taeniopygia guttata) when there are few members of the opposite sex (Elie et al., 2011). Similarly, experimentally skewing sex ratios leads to same-sex sexual behaviour in pigeons (Columba livia f. urbana) (Jankowiak et al., 2018). Sociosexual behaviour is also common in group-living mammals. For example, dolphins (Tursiops spp.) engage in frequent sociosexual behaviour, especially between adult males (Mann, 2006; Harvey et al., 2017). But sociosexual behaviour is not limited to same-sex pairs. Male baboons (Papio anubis) of various ages sometimes mount females without intromission (Owens, 1976), as do Japanese macaques (Macaca fuscata) (Takahata, 1982). However, the published literature seems to focus on that within sexes. For example, male flying foxes (Pteropus pselaphon) engage in genital licking with other males (Sugita, 2016). Anal pentration between males has also been occasionally observed, including in spider monkeys (Ateles geoffroyi (Busia et al., 2018)) and snub-nosed monkeys (Rhinopithecus roxellana (Fang et al., 2018)).

The function of sociosexual behaviour may depend on the taxa, but likely relates to social interactions given its prevalence in group-living species such as primates (MacFarlane & Vasey, 2016). Sociosexual behaviour has been well-documented in baboons and macaques (e.g., Hanby, 1974; Owens, 1976), and a range of other primate species exhibit non-conceptive sexual contact including gibbons (Hylobates lar (Edwards & Todd, 1991)), capuchins (Cebus capucinus (Manson et al., 1997)), and gorillas (Gorilla beringei (Grueter & Stoinski, 2016)). Sociosexual behaviour may be practice for adult mating behaviour (e.g., Japanese macaques, Macaca fuscata (Gunst et al., 2013)) or used to negotiate dominance relations (e.g., captive snub-nosed monkeys, Rhinopithecus roxellana (Huang et al., 2017)). Sociosexual behaviour may also function to reinforce social relationships or reduce tension. Baboons (Papio anubis and P. papio) mount and touch or tug on the penis during ‘greetings’ (Smuts & Watanabe, 1990; Dal Pesco & Fischer, 2018), spider monkeys (Ateles spp.) mount and embrace one another during subgroup reunions (Eisenberg & Kuehn, 1966; Schaffner et al., 2012), and bonobos (Pan paniscus) engage in sexual contact around food, during subgroup reunions, and following aggression (de Waal, 1990; Hohmann & Fruth, 2000; Hohmann et al., 2009; Clay & de Waal, 2015).

Bonobos stand out with regard to their sociosexual behaviour, so much so that they have been dubbed the ‘make love, not war’ ape (de Waal & Lanting, 1998). Bonobos engage in frequent sociosexual behaviour between all age and sex classes, and it is especially common between females (Savage-Rumbaugh & Wilkerson, 1978; de Waal, 1990). One salient and distinct manifestation of this is genito-genital rubbing (or ‘G-G rubbing’) in which females face one another and rub their genitals together, moving them horizontally, back and forth (Kano, 1980; Idani, 1991; Waller, 2016). Sociosexual behaviour in bonobos (1) has been suggested to be a mechanism to reduce tension, (2) may facilitate female alliances which enables female dominance over males, and/or (3) could be a by-product of selection against reactive aggression (de Waal, 1990; Hohmann & Fruth, 2000; Hohmann et al., 2009; Hare et al., 2012; Clay & de Waal, 2015; Hare, 2017). These hypotheses focus on other distinctly bonobo traits such as female dominance and social tolerance, as their closest relative, chimpanzees (Pan troglodytes) are considered to engage in sociosexual behaviour rarely (de Waal, 1988; Gruber & Clay, 2016).

Bonobos appear to engage in sociosexual behaviour more than chimpanzees (Woods & Hare, 2011; Gruber & Clay, 2016), but the focus on it in bonobos seems to have diverted attention from the existence of sociosexual behaviour in chimpanzee life (e.g., Wrangham, 1993; Waller, 2016). As a result, studies of sociosexual behaviour, including those focusing on same-sex behaviour in humans and other animals, highlight the behaviour of bonobos and omit reports on chimpanzees (Vasey, 1995; Bailey & Zuk, 2009; Gray & McIntyre, 2017). This is despite early descriptions of such behaviours from captive and wild chimpanzees including mounting and genital touching (Crawford, 1942; Nishida, 1968; van Lawick-Goodall, 1968; Sugiyama, 1969; Bygott, 1974; Malick & Savage, 1977). And such behaviours continue to be reported, although usually in the context of other social behaviours such as reassurance and reconciliation (de Waal & van Roosmalen, 1979; Arnold & Whiten, 2001; Kutsukake & Castles, 2004) or gestures (Hobaiter & Byrne, 2011; Luef & Pika, 2017), and rarely reported as sociosexual behaviour in its own right (in captivity: Anestis, 2004; anecdotally from the wild: Zamma & Fujita, 2004; Nishie, 2016; Wrangham, 2019). Understanding the extent to which chimpanzees engage in sociosexual behaviour and who they do it with it is important for understanding its function in bonobos, humans, and other primates.

In this study we describe the sociosexual behaviour of wild chimpanzees at Ngogo, in Kibale National Park, Uganda. We include data on males and females, but focus our analysis on the behaviour of adolescent and adult males because male chimpanzees engage in frequent cooperative and competitive behaviour with each other (Mitani, 2009a), and are thus likely to exhibit sociosexual behaviour. In addition, the data presented here were collected as part of a broader study of male social behaviour during adolescence and adulthood. However, we record cases of sociosexual behaviour by females and males of all ages. We describe the types of sociosexual behaviour and the contexts in which these behaviours occurred. The primary focus of our study is descriptive, but we also conduct several analyses to explore how rank and other aspects of social relationships impact the manner and frequency in which individuals engage in sociosexual behavior.

2. Material and methods

2.1. Study site and subjects

The Ngogo chimpanzees live in the centre of Kibale National Park in western Uganda and have been studied continuously since 1995 (Watts, 2012). At the time of our study, there were between 181 and 219 total individuals in the community. Since long-term study began in 1995, the precise ages of adults over 20 years are unknown, and ages are estimated to the nearest year based on physical appearance and known genetic relationships, whereas the ages of chimpanzees under 20 years are known to the nearest year, month, or day (Wood et al., 2017).

Within the Ngogo community, consistent subgroups or “neighbourhoods” exist based on association and space use (Mitani & Amsler, 2003). During our study, individuals were assigned to an eastern, central, and western neighbourhood. Between 2016 and 2017 the western neighborhood decreased its association with the eastern and central neighbourhoods, and by 2018 they rarely interacted.

2.2. Behavioural observations

A.A.S. collected data during two field seasons (24 August 2014 to 30 August 2015 and 16 October 2017 to 5 June 2018). R.B.R. collected data during four field seasons (22 May to 19 August 2015; 24 June 2016 to 12 August 2017; 18 May to 8 June 2018; and 17 July to 28 July 2019). We focused data collection on adolescent and young adult male chimpanzees (between 8 and 21 years old), as part of ongoing studies of male chimpanzee social behaviour during adolescence and the transition to adulthood. We conducted focal following sessions on these males, recording all occurrences of sociosexual behaviour and pant grunts, a formal signal of submission (Bygott, 1979). A.A.S. conducted hour-long following sessions, and R.B.R. conducted following sessions that typically lasted three hours. Together we made 2743 hours of focal observations on 35 adolescent and young adult male chimpanzees (8–21 years). In addition to focal animal sampling of adolescent and young adult male chimpanzees, we recorded sociosexual behaviour, grooming, and pant grunts of all chimpanzees of any age or sex class visible to us ad libitum. To account for this ad libitum observation effort, we indicated which chimpanzees were clearly in our view within 30-min intervals (including our focal subjects). Based on this measure, we spent on average (mean ± SD) 437 ± 205 contact hours with each of the 39 adult males present during the study, 351 ± 125 hours with each of the 22 adolescent males, 131 ± 76 hours with each of the 61 adult females, 134 ± 98 hours with each of the 19 adolescent females, 89 ± 69 hours with each of the 51 juvenile and infant females, and 88 ± 74 with each of the 43 juvenile and infant males. Males were determined to be adolescent (age 8–15 years) if their testes were visible (usually starting around age 9 years), and males were classified as adults beginning at age 16 years (Goodall, 1986; Pusey, 1990). Females were considered adolescents when they began to exhibit sexual swellings (e.g. 11 years) and adults after they gave birth (Goodall, 1986).

We defined sociosexual behaviours as physical interaction involving contact with the ano-genital region except for mating/copulations; we also excluded instances where a male chimpanzee inspected an oestrous female’s sexual swelling. We included behaviour that resembled mating, including mounting and thrusting between males and females if it was clear that there was no penile intromission and/or the female did not have an oestrus swelling. For each sociosexual interaction we noted the date and time of the event, the identities of the individuals involved, and the specific behaviours each individual who took part in the event demonstrated (e.g. Individual A reaches out hand, Individual B moves forward and presses penis to Individual A’s hand). For the descriptions of the behaviour itself, we also relied on pre-defined ethograms of certain behaviours previously reported in the literature, such as ‘mounting’ behaviour (van Lawick-Goodall, 1968; Nishida et al., 1999).

Where possible, we noted the behavioural context in which the event occurred. For events where a context was not described in notes pertinent to the specific sociosexual behaviour event, we consulted our daily notes regarding the timing of subgroup fissions, neighbouring chimpanzee calls and their direction, and other relevant behaviours. Sometimes, several of the preceding events occurred at the same time and precluded us from assigning a clear context to an event. In other cases where multiple events occurred, we could discern a primary context from the attention the chimpanzees paid to a certain individual, event, or stimulus (e.g. if an oestrus female and two males were begging another male for meat, the primary context would be ‘meat’. However, if the same two males were approaching the oestrus female with erections and one held remnants of a monkey, the primary context would be ‘oestrus female’). Given that we conducted continuous focal observations, and a considerable portion of sociosexual behaviour was recorded between individuals outside of the sampling session, there were many instances where a context was not recorded due to limitations in time or visibility.

In a dataset that can be accessed at 10.6084/m9.figshare.13546526, we include descriptions of sociosexual behaviour events and the context(s) in which they occurred. Other types of touch, such as embracing, kissing, and hand-holding, often occur in the same contexts as behaviours that involve genital contact (van Lawick-Goodall, 1968; Nishida et al., 1999). We restricted our analyses to behaviour that clearly involved ano-genital contact given the focus of our study on sociosexual behaviour.

2.3. Variables

We tallied instances of sociosexual behaviour and classified the age and sex classes of individuals involved on the date the instance occurred. We provide descriptions of the specific types of sociosexual behaviours that we observed, the dyads engaged in these behaviours, and the contexts in which sociosexual behaviour occurred. We discuss the rate of sociosexual behaviour only as it applies to adolescent and young adult males, given that these were the only individuals for whom we collected continuous observations.

To determine dominance ranks for males we calculated Elo ratings of all adolescent and adult males using pant grunts (Neumann et al., 2011) using the EloRating package in R (Neumann & Kulik, 2020). We assigned a rank to all males, and then calculated the rank difference between the pair. To assess social relationships between males, we tallied total instances of ad libitum grooming between pairs of adolescent and adult males. We assigned an age to males by calculating the mean age of the chimpanzee during every instance of sociosexual behaviour and grooming bout during the study period. We assigned chimpanzees to an eastern, central, and western neighbourhood using a qualitative assessment. There was perfect agreement by the authors on which individual belonged to which neighbourhood, and qualitative assessments corresponded to social network metrics (based on the leading non-negative eigenvector of the modularity matrix, Csardi & Nepusz, 2006; Newman, 2006). Males were classified as belonging to the same (1) or a different (0) neighbourhood.

2.4. Analysis

First, to assess predictors of sociosexual behaviour among individuals, we limited analysis to adolescent and adult males. Here, we asked whether a male’s rank, age, and the frequency with which he groomed with others impacted how often he engaged in sociosexual behaviour in general. We ran a linear model, with the total counts of sociosexual behaviours per individual as the outcome. The fixed effects were (a) the total counts of grooming, (b) the dominance rank of the male, (c) the age of the male, and (d) observation effort per male. To assess assumptions of the linear model, we plotted a histogram of the residuals and a quantile-quantile plot, and the residuals appeared normally distributed. We also plotted the fitted values against the residuals and there was no distinct pattern to the residuals.

Second, we assessed predictors of sociosexual behaviour at the dyadic level. We summed total cases of sociosexual behaviour between pairs of adolescent and adult males. Since most pairs (621/801) never engaged in sociosexual behaviour, and those that did often only did so once (160/201), we classified whether or not pairs ever engaged in sociosexual behaviour during our study period. In this analysis we investigated how rank differences between two males along with two other aspects of their social relationships influenced the probability that they engaged in sociosexual interactions with one another. Specifically, we assessed both how frequently they groomed, as an indicator of affiliative bond strength, and whether or not they belonged to the same neighbourhood, as males in different neighbourhoods meet less frequently and their fusions might produce more tension than fusions between males of the same neighbourhood. However, because males who belong to different neighbourhoods may nevertheless share strong affiliative bonds, we also included an interaction between grooming and neighbourhood membership in our analysis. We conducted a generalized linear mixed model with a binomial error distribution. The outcome was whether or not the pair engaged in sociosexual behaviour (1 or 0). Fixed effects were (a) total counts of grooming within the pair, (b) whether the pair belonged to the same neighbourhood within the community, (c) an interaction between grooming and neighbourhood, and (d) the rank difference between the pair. We included the identity of the two chimpanzees as two random intercepts. To produce comparable estimates for each fixed effect, we z-transformed variables by subtracting the mean from each value and dividing it by two standard deviations of the value (Schielzeth, 2010).

Third, we explored the impact of rank on male–male mounting behaviour in particular to test whether dominant males typically mount subordinate males. We conducted a model to determine, in cases of mounting, whether the mounter was more often than not higher ranking than the mountee. For each case of mounting, we assigned a rank to both males, and then calculated the rank difference by subtracting the mountee’s rank from the mounter’s rank. Since the same individual may appear multiple times as the mounter or mountee, each instance of mounting is not independent. To control for the non-independence of data due to repeated individuals in the sample, we ran an intercept-only generalized linear mixed model with a binomial distribution. Whether the mounter was higher ranking (1) or lower ranking (0) than the mountee was the outcome. We included the identity of the mounter and the identity of the mountee as two random intercepts. This generalized linear mixed model is similar to a binomial test. As a comparison, we also conducted a binomial test which asseses whether the mounter was higher ranking that the mountee more often than is expected by chance, although this analysis ignores the non-independence of data due to repeated individuals in the sample.

To assess the importance of variables in our models, we report coefficients and standard errors for each fixed effect. We used the lme4 package (Bates et al., 2015) in R (R Core Team, 2020). Additional R packages used for data organization included lubridate (Grolemund & Wickham, 2011) and tidyr (Wickham & Henry, 2020).

3. Results

We recorded 584 instances of non-mating genital touching, or sociosexual behaviour, in chimpanzees during three years of observation. Sociosexual behaviour took four main forms: mounting (335 instances = 57%), hand-to-anogenital-region touches (177 instances = 30%), rump-to-body (33 instances = 6%) and rump-to-rump contact (29 instances = 5%). Sociosexual behaviour occurred between every combination based on age and sex (Table 1), and it was most frequent between adult males (33% of cases) and between adult males and adult females (19%). We never observed intromission or ejaculation during any sociosexual behaviour. Although sociosexual behaviour was not common, per se, during this same period we observed 1681 copulations between males and females, which was only three times more frequent than sociosexual behaviour. When limited to data collected on adolescent and adult males during focal follows, for whom we could calculate a rate, we observed 153 cases of sociosexual behaviour across the study period, indicating that such behaviours occurred once every 166 hours per individual, which is approximately twice per month.

Figure 1.
Figure 1.

An adult male mounts another adult male, while a third adult male is vigilant to the left (photo by Nathan Chesterman).

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

Mounting was the most common form of sociosexual behaviour we observed, in which one chimpanzee stood bipedally behind a quadrupedally standing chimpanzee, often while holding their waist or back (Figure 1). Mounting was prevalent among male chimpanzees, accounting for 89% (250/393) of all sociosexual behaviours observed between sexually mature males. Mounting was also observed in all other age and sex classes, including between sexually mature males and sexually mature females (79/174 = 45%), between sexually mature females (8/38 = 21%), between sexually mature chimpanzees and immatures (46/88 = 52%), and between immature juveniles or infants (7/11 = 64 %). One-third of mounting cases included thrusting (110/335), and males and females of all ages were observed to thrust, and to assume both mounter and mountee positions, even within the same pairs of individuals. Sometimes individuals reached back to touch the genitals of a chimpanzee who mounted them. For these instances, we classified them as mounts rather than separate sociosexual behaviours.

Table 1.
Table 1.

Sociosexual behaviour between different age and sex combinations in chimpanzees at Ngogo, Kibale National Park, Uganda.

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

The second most common form of sociosexual behaviour was hand-to-genital contact, and these varied depending on the age and sex of the individuals involved. Hand touches directed at females of any age typically incorporated patting and rubbing of the clitoris, labia, and anal region, and often the individual who touched a particular female did so when the female’s back faced them. Sometimes a female presenting her rump initiated the direction of this touch, but other times another individual reached up to pat or rub her rump as she walked by, or ran to touch her when she was screaming. Hand touches to the penis and scrotum involved both reaches from the back/rump area and ventral reaches made while facing the individual (Figure 2). Penis- and scrotum-touching typically involved the clutching and cupping of the scrotum and sometimes patting the penis. Hand-to-genital touches were the most frequent form of sociosexual behaviour in which sexually mature female chimpanzees engaged, including between sexually mature females (22/38 = 58%) and between sexually mature males and sexually mature females (80/174 = 46%). Hand touches also occurred between other age and sex classes, including between sexually mature males (89/393 = 23%), between sexually mature chimpanzees and immatures (20/88 = 23%), and between immatures (2/11 = 18%).

Figure 2.
Figure 2.

An adult male touches his hand to the ano-genital region of an early adolescent male after the adolescent had screamed and approached the adult male (photo by Aaron Sandel).

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

We observed other forms of genital contact that occurred less frequently. Rump-to-body touches involved one chimpanzee pressing their rump into the side, chest, back, or other body part of another chimpanzee (6% of all sociosexual behaviour). They accounted for 7% (26/393) of all sociosexual behaviours observed between mature males, 3% (5/174) between mature males and mature females, 13% (5/38) between mature females, 11% (10/88) involving mature chimpanzees and immatures, and 9% (1/11) between immatures. We observed 29 instances (5% of all sociosexual behaviour) of rump-to-rump touches. Rump-to-rump touches involved two chimpanzees pressing their rumps together by backing into each other while standing quadrupedally, and the touching occasionally involved rubbing behaviour, especially when it occurred between two females. Rump-to-rump touches accounted for 5% (18/393) of all sociosexual behaviours observed between mature males, 6% (10/174) between mature males and mature females, 9 % (3/38) between mature females, 10% (9/88) involving mature chimpanzees and immatures, and 9% (1/11) between immatures.

In addition to the preceding incidents, we observed nine events that involved aspects of both rump-pressing and mounting behaviour but differed from the typical forms of either behaviour. In eight cases, a juvenile or young adolescent chimpanzee pressed their rump into the groin area of an adult male chimpanzee, touching his penis and/or scrotum. In a different case, after an aggressive event, a young adolescent male (Cannonball) ran screaming toward another young adolescent male (Fleck), who was resting on the ground, on his stomach. Cannonball stood behind Fleck in a mounting posture still screaming, and Fleck did not rise from the ground. Cannonball then thrusted his penis back and forth on Fleck’s back.

We also observed two less frequent, but notable behaviours to occur between sexually mature males. Males engaged in ventral-ventral genital contact on three occasions, and each occasion occurred between different pairs of males. One of these instances involved mouth-to-genital contact between a pair of adult males (Evans and Mulligan, see Video 1 at 10.6084/m9.figshare.13546526). In this case, upon joining together in the same party, the two adult males rushed toward each other, met in a tree, and one put his mouth to the other’s groin for approximately 30 seconds. Then, still in the tree, they repositioned and engaged in ventral-ventral contact as they rubbed their genitals together for an additional 30 seconds (Video 1 at 10.6084/m9.figshare.13546526).

3.1. Contexts of sociosexual behaviour

Sociosexual behaviour occurred in a variety of contexts. Due to limitations in visibility and sampling procedure, we could assign a clear context to 53% (311/584) of the events that we observed (Table 2). Sociosexual behaviour occurred (1) during or prior to fusions of chimpanzee parties or the arrival of a new individual to a party; (2) in situations where neighbouring chimpanzees were encountered, could be encountered, or where chimpanzees were actively patrolling for neighbours; (3) during within-group aggression, either by being the direct target of aggression or preparing to aggress another individual, while being surrounded by aggression, or, especially for infant and juvenile chimpanzees, when a conflict arose during play; (4) when chimpanzees were trying to access a monopolizable ‘resource’ such as a feeding spot in a tree, meat that one or several individuals held, or an oestrus female; (5) other arousing situations such as when encountering monkeys they might hunt, when a female arrived with a newborn infant for the first time, and if chimpanzees encountered other animals that caused them to alarm, such as wild pigs; (6) when chimpanzees interfered in copulations and engaged in sociosexual behaviour with one or both of the individuals mating with each other; and (7) between pairs who groomed.

Table 2.
Table 2.

Contexts of sociosexual behaviour across age- and sex-classes.

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

We observed adult and adolescent male chimpanzees (for whom we had the most observations), to engage in sociosexual behaviour in all of the preceding contexts, and for other age and sex classes to do so in many of them (Table 2). Notably, only sexually mature male chimpanzees who compete for mating access toward females showed sociosexual behaviour when in the presence of an oestrus female. In addition, infant and juvenile chimpanzees did not engage in sociosexual behaviours upon hearing calls from neighbouring chimpanzees, though they did do so while traveling quietly at the edge of the territory when on territorial patrols.

Below, we provide more detailed examples of sociosexual behaviour including notes on the style, context, and individuals involved.

3.1.1. During territorial behaviour

Upon hearing chimpanzees calls of neighbouring communities, male chimpanzees almost always mount one another (Figure 3). On one occasion, a subgroup of chimpanzees was near the border of their territory. Then they became alert, likely having heard neighbouring chimpanzees. As several chimpanzees ran toward the boundary of the territory toward the neighbouring group, other chimpanzees touched and mounted one another. Notably, two males mounted each other ventrally. One male fell on his back, and the other sat on top of him and thrusted, both ‘fear grimacing’ before they proceeded with the others toward the neighbouring group (Video 2 at 10.6084/m9.figshare.13546526).

Figure 3.
Figure 3.

Several adult males mount one another immediately after hearing a neighbouring group of chimpanzees, and just prior to an inter-group encounter (photo by Aaron Sandel).

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

3.1.2. Joining other subgroups within the community

The fission-fusion dynamics of chimpanzees necessitates that group members part and reunite. Reunions, whether between two individuals or two subgroups within the same community are, at times, marked by sociosexual interactions (Table 2). For example, on one occasion when two subgroups reunited, one male mounted another, and the mountee rubbed his foot to the mounter’s penis. On another occasion, an adult female and adult male met each other. Rather than pant grunt to the male, as is typically done when a female meets an adult male, both individuals turned around and pressed their rumps against each other.

3.1.3. Begging for and sharing meat

After chimpanzees hunt, one chimpanzee is usually in possession of meat and others gather around, reaching their hands out for meat, touching the meat and pulling on pieces, and grimacing or screaming. Many individuals are in close proximity, almost touching, and these instances lead to frequent genital contact. For example, six records of sociosexual behaviour occurred on one day (7 December 2018) during one bout of meat sharing.

3.1.4. During play bouts

Infants and juveniles tended to exhibit sociosexual behaviour during play, especially when play behaviour appeared to escalate to aggression. One individual would scream, play would end temporarily, the pair would engage in sociosexual behaviour, then resume playing. Such instances could be considered reconciliation.

3.1.5. Adolescent immigrant females and adult females

Although our observations of female behaviour are limited because of our male-biased research focus, adolescent females appeared to engage in sociosexual behaviour most frequently with adult females. Adolescent females engaged in sociosexual behaviour with adult females 16 times, more so than they did with adolescent males (10 times) or adult males (11 times). Adolescent females often sought specific older females out for reassurance generally or received reassurance from them, sometimes in the form of sociosexual behaviour reported here. For example, when subgroup fusion occurred, a recent immigrant screamed and ran to mount an older female with whom she had been frequently traveling. One month later, while aggression occurred around them, the same adolescent pressed and rubbed her rump into the chest of the same older resident female. On another occasion an adolescent immigrant screamed repeatedly and an older resident female with whom she traveled ran to her and mounted and thrusted. In another instance, an older female ran to a screaming young immigrant who often followed and groomed her, reached out and rubbed her swelling until she stopped screaming.

3.1.6. Social predictors of sociosexual behaviour

To assess the impact of grooming and dominance relationships on sociosexual behaviour, we limited our analysis to adolescent and adult males. First we looked at how often individuals engaged in sociosexual behaviour overall, regardless of who they did it with. Males that groomed frequently in general also engaged in more sociosexual behaviour (estimate (grooming) = 0.06 ± 0.02, p=0.003); neither dominance rank nor age had a strong effect on how often a male engaged in sociosexual behaviour (estimate (rank) = 0.002 ± 0.005, p=0.625; estimate (age) = −0.05 ± 0.088, p=0.596; estimate (observation effort) = 0.010 ± 0.006, p=0.095).

Second, we looked at the effect of grooming and dominance relationships on sociosexual behaviour at the dyadic level. Only 25% of pairs of adolescent and adult males (201/822) ever engaged in sociosexual behaviour with each other, and only 5% of pairs (41/822) did so more than once. Thus we assigned whether pairs did or did not engage in sociosexual behaviour at any point during the study period. We compared this to how often the same pair groomed each other, whether the pair belonged to the same neighbourhood within the community, and the rank difference between the pair. There were no strong predictors of whether a pair engaged in sociosexual behaviour (estimate(grooming) = 0.26 ± 0.26, p=0.315; estimate(same neighbourhood) = −0.09 ± 0.21, p=0.649; estimate(rank difference) = 0.24 ± 0.25, p=0.355; estimate(grooming by neighbourhood interaction) = 1.03 ± 0.62, p=0.095). Although how often a pair groomed had a minimal effect on sociosexual behaviour, there were some pairs who groomed frequently and also engaged in sociosexual behaviour frequently (Figure 4).

Figure 4.
Figure 4.

Grooming and sociosexual behaviour. How often a pair groomed was not positively related to how much the pair engaged in sociosexual behaviour.

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

To assess whether mounting related to dominance, we investigated the difference in rank between males engaged in mounting behaviour. When mounting occurred between males, the mounter was not more likely to be higher ranking than the mountee (Figure 5, generalized linear mixed model: intercept = −2.02, p=0.388; random effects: mounter variance = 7.50, mountee variance = 27.02). If anything, the mounter tended to be lower ranking than the mountee (Exact Binomial test: probability of mounter being higher-ranking = 40.2 %, 95% confidence interval = 31.4% to 49.4%, p=0.037). But there were also many instances in which the mounter was higher ranking than the mountee (Figure 5).

Figure 5.
Figure 5.

Difference in dominance rank between mounter and mountee. Dotted lines represent ±50 Elo rating difference from zero, indicating comparable ranks.

Citation: Behaviour 158, 3-4 (2021) ; 10.1163/1568539X-bja10062

4. Discussion

Sociosexual behaviour, including same-sex sexual behaviour, occurred in every sex- and age-class combination among chimpanzees at Ngogo. The most common type of sociosexual behaviour was mounting. Chimpanzees would also reach their hand to touch the rump or genitals of others and on rare occasions touch rumps or engage in genito-genital touching. Sociosexual behaviour occurred in multiple contexts. In nearly all cases for which we recorded the context, the context represented cases of tension, often due to aggression or the threat of aggression, including during food sharing, upon hearing neighbouring groups of chimpanzees, or prior to or after joining another subgroup within the community (Table 2).

Sociosexual behaviour was variable among the Ngogo chimpanzees. It appeared to be most common between adult males. This may be due to the cooperative nature of male chimpanzees, and the tension involved during certain activities, such as reunions that involve the reestablishment of dominance relationships. Adolescent and young adult male chimpanzees were the focus of our research, so they may be overrepresented in the sample. Indeed, we have considerably more observation time on them than any other age-sex class. However we frequently observed sociosexual behaviour between adult male pairs that were not the focus of our research. Despite our focus on adolescent and adult males, we recorded individuals in all possible age-sex class pairings engage in sociosexual behaviour including between sexually mature adult females (Table 1). Nevertheless, given our sampling procedure, females and infants are likely underrepresented. In fact, we likely underestimated the frequency of sociosexual behaviour in general, as such behaviours often occurred quickly and during somewhat chaotic events, including during aggression or just prior to the chimpanzees running toward a neighbouring group of chimpanzees, making it difficult to record.

Some individuals appeared to engage in sociosexual behaviour more than others. Several adult males accounted for a large proportion of sociosexual behaviours. One young adult male, somewhat low-ranking but rising in the male hierarchy (Evans), mounted others 30 times, and only twice was mounted by others. Another adult male, middle-aged and relatively low-ranking (Mulligan) was mounted nine times, and never mounted others. In fact, these two males engaged in one of the instances of genito-genital rubbing (Video 1 at 10.6084/m9.figshare.13546526). One of these males, Mulligan, was also involved in an unusual case of mounting in which he presented his rump to another male, and while the other male (Dexter) mounted him, Mulligan crouched down on the ground in a posture resembling that of females during a copulation. Other males that were over-represented in our sample included high-ranking adult males. It is possible that status and personality play a role in the frequency of sociosexual behaviour. High-ranking males may encounter tense situations frequently — competing for mating opportunities, sharing meat, participating in border patrols — instigating sociosexual behaviour as well as other forms of relationship regulation, such as grooming. In addition, some males may be more ‘nervous’ in general, and thus seek out reassurance from others.

With regard to the prevalence of sociosexual behaviour in chimpanzees, it is possible that it is more common at Ngogo than other chimpanzee sites. There are more males at Ngogo than any other community that has been studied. At Ngogo, chimpanzees, especially adult males, engage in frequent cooperative behaviour, group hunts, and border patrols (Mitani et al., 2000; Mitani & Watts, 2001; Watts & Mitani, 2001; Mitani, 2009b; Langergraber et al., 2017). Being surrounded on all sides by other chimpanzee communities, they regularly have intergroup encounters. As a result, there may be heightened need for regulating tension. However, we consider it unlikely that Ngogo stands out with regard to the nature and prevalence of sociosexual behaviour as similar behaviours have been described at multiple other sites (van Lawick-Goodall, 1968; Bygott, 1974; Nishida et al., 1999). For example, a study at Gombe found that 50% of all touching between male chimpanzees (N=194 total touches) involved touching of genitals, and 70% of all touching between female chimpanzees (N=56 total touches) involved touching genitals (Bygott, 1974). There also appeared to be variation among individuals in the frequency and patterning of sociosexual behaviour, including with some individuals being distinct in their sociosexual behaviour. Bygott (1974, p. 63) noted that “the adult male Faben often presented to other males, and when they held out a hand to him he would bounce his scrotum up and down against their hand. No other male was seen to do this.”

Further research is required to test hypotheses for the function of sociosexual behaviour in wild chimpanzees. Our findings suggest that sociosexual behaviour is unlikely related to dominance given that combinations of all age and sex classes engaged in these behaviours. Importantly, in cases of mounting between males, the mounter was not higher ranking; if anything, the mounter tended to be lower ranking than the mountee, although mounters were also frequently higher ranking than the mountee (Figure 5). Sociosexual behaviour may relate to other behaviours, such as affiliative bonds, but grooming was not a strong predictor of sociosexual behaviour (Figure 4). Some pairs that groomed frequently also engaged in sociosexual behaviour more than did other pairs, but sociosexual behaviour also occurred between males who associated less frequently because they occupied different social neighbourhoods within the community. Thus, at least for males, sociosexual behaviour may be a way to reduce tension among pairs who meet infrequently. The function of sociosexual behaviour likely varies by age and sex. For example, we observed infant females rubbing their genitals on others, and this may have been for a pleasurable sensation alone (Vasey & Duckworth, 2006). Further study of sociosexual behaviour is required in chimpanzees, particularly focused on females, and how the function of sociosexual behaviour changes with development, which will in turn inform its possible functions. Although we focused on genital contact, much of the behaviour that we observed resembled what other researchers have considered ‘reassurance’ behaviour (Goodall, 1986; Nishida et al., 1999). Prior studies seem to have lumped sociosexual behaviour within reassurance behaviour, and indeed, they may serve the same function. Future studies should analyse sociosexual behaviour along with other forms of touch, including mouth-to-mouth, hand-to-mouth, hand-to-hand, hand-to-body, and embraces without genital contact.

Given that much of sociosexual behaviour across mammals has been linked to cooperation and tension reduction in fission-fusion species, it is no surprise that chimpanzees do it. Indeed, this finding is not new. Some of the earliest studies of chimpanzees reported behaviours involving mounting and genital contact, both in captivity (Crawford, 1942) and the wild (Nishida, 1968; van Lawick-Goodall, 1968; Sugiyama, 1969). However, the ensuing decades have focused on such behaviours in bonobos, framing them as key for tolerance and female power (Hohmann & Fruth, 2000; Hohmann et al., 2009), and this seems to have downplayed its significance in chimpanzee life. For example, Hohmann et al. (2009) considered genital contact to be ‘habitual’ in bonobos, but not in other apes, and Grueter & Stoinski (2016) considered same-sex sexual behaviour “rare or absent among chimpanzees” in their study of such behaviours in wild gorillas. Similarly, sociosexual behaviour is considered “weak or infrequent” in East African chimpanzees (Pan troglodytes schweinfurthii) compared to hunting, which is “strong or highly frequent,” and infanticide, which is “moderate or frequent” (Gruber & Clay, 2016). In our study, sociosexual behaviour occurred considerably more frequently than did infanticides and was closer to the frequency of hunts, thus we would not consider it as “weak or infrequent”. Although we were not able to calculate a rate of sociosexual behaviour for most of our subjects, sociosexual behaviour was only one third as frequent as mating behaviour. And adolescent and young adult males tended to exhibit sociosexual behaviour approximately once every two months, which is likely an underestimate. It may have been that the study by Gruber and Clay (2016) considered a more limited array of behaviours as sociosexual, which could also account for its apparent absence between young chimpanzees compared to bonobos in an experimental feeding context (Woods & Hare, 2011). When analysing chimpanzee behaviour, scientists seem to classify mounts or genital contact within other functional behaviours, such as gestures or reassurance.

We do not dispute that genital contact may serve conciliatory or communicative purposes in chimpanzees, but we also suggest that it deserves attention in studies investigating sociosexual behaviour explicitly. Doing so will allow proper comparisons to bonobos, and may elucidate the evolution and function of sociosexual behaviour in primates more generally. Although we found chimpanzees to engage in a range of sociosexual behaviours, bonobos remain distinct. Bonobos exhibit frequent genito-genital rubbing, especially between adult females (Hohmann & Fruth, 2000), whereas, we recorded only three instances of face-to-face, genito-genital rubbing in chimpanzees, and it was only observed between adult males. However, different species of primates manifest different forms of sociosexual behaviour, such as ‘ritualized’ touching of the penis in some baboons species (Smuts & Watanabe, 1990; Dal Pesco & Fischer, 2018). For chimpanzees, mounting and touching genitals may be their species-typical manifestation of sociosexual behaviour.

That behaviours involving gential contact have not been conceptualized as ‘sociosexual’ in chimpanzees may also be due to cultural biases against homosexuality, as has been seen in the literature on other primates (Vasey, 1995). “Never, however, have we seen anything that could be regarded as homosexuality in chimpanzees,” Jane Goodall writes in her 1971 book, In the Shadow of Man. “Admittedly a male may mount another in moments of stress or excitement, clasping the other around the waist, and he may even make thrusting movements of the pelvis, but there is no intromission. It is true, also, that a male may try to calm himself or another male by reaching out to touch or pat the other’s genitals; while we still have much to learn about this type of behaviour, it certainly does not imply homosexuality. He only does this in moments of stress, and he will touch or pat a female on her genitals in exactly the same contexts” (Goodall, 2010, pp. 183–184). We agree that this does not indicate ‘homosexuality’ in the sense of a sexual orientation, but it does potentially represent sexual behaviour broadly defined, including that between members of the same sex.

Our findings reveal that sociosexual behaviour is a standard component of chimpanzee behaviour. Although sociosexual behaviour is, no doubt, more salient and frequent in bonobos, especially with their characteristic side-to-side genital rubbing, the range and type of sociosexual behaviour is similar in chimpanzees. Thus, there should not be such a distinction made between the two species. In addition, given the evidence of sociosexual behaviour in gorillas and a range of monkey species (e.g., Yamagiwa, 1987; Grueter & Stoinski, 2016), sociosexual behaviour is likely a common trait to haplorrhine primates. In some primate species, sociosexual behaviour has become key to negotiating relationships, as is the case in bonobos, some baboon species, and some populations of spider monkeys. Given its prevalence across taxa, sociosexual behaviour, including between members of the same sex, may be an important component of relationships that arose early on in primate evolution.

*

Corresponding author’s e-mail address: aaron.sandel@austin.utexas.edu

Acknowledgements

Research was approved by the Uganda Wildlife Authority and the Uganda National Council for Science and Technology and supported by the Makerere University Biological Field Station. Jacob Negrey provided important feedback on the development of this project. For additional comments, we also thank Rachel Voyt, Izzi Clark, Kiku Adatto, Elisabetta Palagi and two anonymous reviewers. Nathan Chesterman provided assistance in the field and video and photo material for this project. For additional support in the field, we thank John Mitani, David Watts, Kevin Langergraber, Sam Angedakin, Alfred Tumusiime, Ambrose Twineomujuni, Godfrey Mbabazi, Lawrence Ndangizi, Brian Kamugyisha, Chris Aliganyira, Charles Birungi, Jeremy Clift, Rebecca Davenport, Bethany Hansen, Sarah Dunphy-Leilii, Caleb Vogt, Braise Mugisha, Carolyn Rowney, Tom Struhsaker, William Sunday, James Tibisimwa, Charles Businge and the late Jerry Lwanga. A.A. Sandel was supported by L.S.B. Leakey Foundation, the University of Michigan, Arizona State University, the Nacey-Maggioncalda Foundation, and the National Science Foundation (1540259 and F031543). R. Reddy was supported by the National Geographic Society, the National Science Foundation (1613392 and 1256260), the Nacey-Maggioncalda Foundation, the L.S.B. Leakey Foundation, and the African Studies Center, International Institute, Rackham Graduate School, and the Department of Anthropology at the University of Michigan. Broader impacts of this project included conservation education and capacity building at Kyakagunga Primary School, Kamwenge District, Uganda and professional development workshops in Bigodi, Kamwenge District, Uganda.

References

  • Anestis, S.F. (2004). Female genito-genital rubbing in a group of captive chimpanzees. — Int. J. Primatol. 25: 477-488.

  • Arnold, K. & Whiten, A. (2001). Post-conflict behaviour of wild chimpanzees (Pan troglodytes schweinfurthii) in the Budongo forest, Uganda. — Behaviour 138: 649-690.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bailey, N.W. & Zuk, M. (2009). Same-sex sexual behavior and evolution. — Trends Ecol. Evol. 24: 439-446.

  • Bates, D., Maechler, M., Bolker, B. & Walker, S. (2015). Fitting linear mixed-effects models using lme4. — J. Stat. Softw. 67: 1-48.

  • Busia, L., Denice, A.R., Aureli, F. & Schaffner, C.M. (2018). Homosexual behavior between male spider monkeys (Ateles geoffroyi). — Arch. Sex. Behav. 47: 857-861.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bygott, J.D. (1974). Agonistic behaviour and dominance in wild chimpanzees. — Dissertation, University of Cambridge, Cambridge.

  • Bygott, J.D. (1979). Agonistic behavior, dominance, and social structure in wild chimpanzees of the Gombe National Park. — In: The great apes (Hamburg, D. & McCown, E., eds). Benjamin/Cummings, Menlo Park, CA.

    • Search Google Scholar
    • Export Citation
  • Clay, Z. & De Waal, F.B. (2015). Sex and strife: post-conflict sexual contacts in bonobos. — Behaviour 152: 313-334.

  • Crawford, M.P. (1942). Dominance and the behavior of pairs of female chimpanzees when they meet after varying intervals of separation. — J. Comp. Psychol. 33: 259-265.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Crews, D. & Fitzgerald, K.T. (1980). “Sexual” behavior in parthenogenetic lizards (Cnemidophorus). — Proc. Natl. Acad. Sci. USA 77: 499-502.

  • Csardi, G. & Nepusz, T. (2006). The igraph software package for complex network research. — InerJournal Complex Systems: 1695.

  • Dal Pesco, F. & Fischer, J. (2018). Greetings in male Guinea baboons and the function of rituals in complex social groups. — J. Hum. Evol. 125: 87-98.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Waal, F.B.M. (1988). The communicative repertoire of captive bonobos (Pan paniscus), compared to that of chimpanzees. — Behaviour 106: 183-251.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Waal, F.B.M. (1990). Sociosexual behavior used for tension regulation in all age and sex combinations among bonobos. — In: Pedophilia. Springer, Berlin, p. 378-393.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Waal, F.B.M. & Lanting, F. (1998). Bonobo: the forgotten ape. — University of California Press, Berkeley, CA.

  • De Waal, F.B.M. & Van Roosmalen, A. (1979). Reconciliation and consolation among chimpanzees. — Behav. Ecol. Sociobiol. 5: 55-66.

  • Edwards, A.-M.A. & Todd, J.D. (1991). Homosexual behaviour in wild white-handed gibbons (Hylobates lar). — Primates 32: 231-236.

  • Eisenberg, J.F. & Kuehn, R.E. (1966). The behavior of Ateles geoffroyi and related species. — Smithsonian Miscellaneous Collections 151: 4683.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Elie, J.E., Mathevon, N. & Vignal, C. (2011). Same-sex pair-bonds are equivalent to male–female bonds in a life-long socially monogamous songbird. — Behav. Ecol. Sociobiol. 65: 2197-2208.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fang, G., Dixson, A.F., Qi, X.-G. & Li, B.-G. (2018). Male-male mounting behaviour in free-ranging golden snub-nosed monkeys (Rhinopithecus roxellana). — Folia Primatol. 89: 150-156.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Goodall, J. (1986). The chimpanzees of Gombe: patterns of behavior. — Harvard University Press, Cambridge, MA.

  • Goodall, J. (2010). In the shadow of man. — Mariner Books/Houghton Mifflin Harcourt, Boston, MA/New York, NY.

  • Gray, P.B. & Mcintyre, M.H. (2017). Development of human sociosexual behavior. — In: The arc of life. Springer, Berlin, p. 41-53.

  • Grolemund, G. & Wickham, H. (2011). Dates and times made easy with lubridate. — J. Stat. Softw. 40: 1-25.

  • Gruber, T. & Clay, Z. (2016). A comparison between bonobos and chimpanzees: a review and update. — Evol. Anthropol. 25: 239-252.

  • Grueter, C.C. & Stoinski, T.S. (2016). Homosexual behavior in female mountain gorillas: reflection of dominance, affiliation, reconciliation or arousal?. — PloS one 11: e0154185.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gunst, N., Leca, J.-B. & Vasey, P.L. (2013). Development of sexual and socio-sexual behaviours in free-ranging juvenile male Japanese macaques, Macaca fuscata. — Behaviour 150: 1225-1254.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hanby, J. (1974). Male-male mounting in Japanese monkeys (Macaca fuscata). — Anim. Behav. 22: 836-849.

  • Hare, B. (2017). Survival of the friendliest: Homo sapiens evolved via selection for prosociality. — Annu. Rev. Psychol. 68: 155-186.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hare, B., Wobber, V. & Wrangham, R.W. (2012). The self-domestication hypothesis: evolution of bonobo psychology is due to selection against aggression. — Anim. Behav. 83: 573-585.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Harvey, B.S., Dudzinski, K.M. & Kuczaj, S.A. (2017). Associations and the role of affiliative, agonistic, and socio-sexual behaviors among common bottlenose dolphins (Tursiops truncatus). — Behav. Process. 135: 145-156.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hobaiter, C. & Byrne, R.W. (2011). The gestural repertoire of the wild chimpanzee. — Anim. Cogn. 14: 745-767.

    • Crossref