Same-sex sexual behaviour (SSB) has been documented in a wide range of animals, but its evolutionary causes are not well understood. Here, we investigated SSB in the light of Reeve's acceptance threshold theory. When recognition is not error-proof, the acceptance threshold used by males to recognize potential mating partners should be flexibly adjusted to maximize the fitness pay-off between the costs of erroneously accepting males and the benefits of accepting females. By manipulating male burying beetles' search time for females and their reproductive potential, we influenced their perceived costs of making an acceptance or rejection error. As predicted, when the costs of rejecting females increased, males exhibited more permissive discrimination decisions and showed high levels of SSB; when the costs of accepting males increased, males were more restrictive and showed low levels of SSB. Our results support the idea that in animal species, in which the recognition cues of females and males overlap to a certain degree, SSB is a consequence of an adaptive discrimination strategy to avoid the costs of making rejection errors.
Same-sex sexual behaviour (SSB) is a widespread phenomenon in the animal kingdom, occurring in vertebrates as well as invertebrates [1,2]. In the last years, the number of studies that have tried to unravel the reasons for the evolutionary maintenance of SSB in populations has increased considerably, and various adaptive and non-adaptive hypotheses have been proposed [1,2]. Mistaken identity is one popular non-adaptive explanation, especially for the occurrence of homosexual behaviour in insects . In populations where the phenotypic cues (i.e. recognition labels) of males and females overlap and there are feminine males and masculine females, discriminating individuals inevitably will make mistakes. If the discriminating male has a high-acceptance threshold and is very restrictive, it will risk rejecting many females, while if it has a low-acceptance threshold and is very permissive, it will accept males as mating partners. Theory predicts that the optimal acceptance threshold depends on the benefits of accepting and rejecting the ‘right’ or ‘wrong’ individual, respectively, and the costs and likelihood of making an acceptance error (here: mating with a male) or rejection error (here: rejecting a female) . However, although the acceptance threshold theory has been tested in different contexts (e.g. [5,6]), there is currently no study that has investigated whether the occurrence of SSB depends on the costs of making mistakes and benefits of mating with a female. Indeed, in their recent review about SSB, Scharf & Martin  explicitly stated that the next research step should be predicting under which conditions males are expected to show SSB more or less often and to test these predictions experimentally.
In the current study, we created conditions in which male burying beetles, Nicrophorus vespilloides, perceived low or high costs of rejecting females and low or high costs of accepting males. If rejecting a female is very costly, males should exhibit more permissive discrimination decisions, and therefore higher frequency of SSB, than if the costs are low. If the costs of accepting a male are high, males are expected to engage in more restrictive discrimination decisions and therefore in a lower frequency of SSB than if the costs are low. We conducted two separate experiments in which we influenced the costs of rejection and acceptance error by manipulating (i) a male's searching costs (i.e. search time) for females and (ii) his reproductive potential (availability of a resource needed for reproduction). We used burying beetles as a model organism because it has been already established that they have a plastic recognition system and are able to flexibly adjust their acceptance threshold and, therefore, their discrimination decisions .
2. Material and methods
(a) Study organism and maintenance
Burying beetles (Nicrophorus spp.) provide elaborate biparental care to their offspring [8,9]. They compete for and breed on the carcasses of small vertebrates, which are an essential, but rare and ephemeral resource. Females and males mate both on and off a carcass. For our study, we used descendants of N. vespilloides beetles trapped in a deciduous forest in Freiburg, Germany. Experimental beetles were housed in plastic containers filled with moist peat and fed mealworms twice a week. All beetles were maintained at 20°C under a 16 L : 8 D regime.
(b) Treatment groups
In the first experiment, we created situations in which the costs of rejecting females differed for experimental males. Immediately after eclosion, males were kept at three different female densities to promote differences in perceived search time: (i) without a female, (ii) with one female and (iii) with three females. To control for general density effects, we included two additional female-deprived treatment groups, where we kept males (iv) with another male and (v) with three males. Males were either kept for 20–22 days under these conditions (20-day treatment) or for 55–60 days (60-day treatment; see the electronic supplementary material for additional information about the experimental design and sample sizes). After these time periods, we performed mating trials in which we presented the experimental males with either a male or a female (see the electronic supplementary material). In situations in which males were kept with one or several females, males perceived a lower searching time for females than the males in female-deprived situations, especially in the 60-day treatment. Hence, the males in the long-lasting female-deprived situations were expected to avoid making any rejection error and be more permissive, whereas the males that perceived a high female encounter rate and therefore low search costs throughout the 60 days of treatment could afford to be choosier and were expected to be more restrictive (cf. ‘search model’ ). In the shorter 20-day treatment, we did not expect many differences in discrimination decisions between the males of the different social environments, as perceived searching time should generally be low relative to the onset of sexual maturation (approx. on day 12–16; J. K. Müller 2013, personal communication). Hence, males of the 20-day treatment were expected to show a low level of SSB.
In the second experiment, we created situations in which we varied the costs of accepting males by manipulating the availability of a resource essential for reproduction. All focal males were kept alone for approximately 60 days. After this time period, the males were transferred to a small plastic container (7 × 3 × 3 cm) filled to one-quarter with plaster. For 2 h, about half of the males (n = 18) were provided with a mouse carcass; the other half (n = 16) did not obtain a resource. Immediately afterwards, we performed mating trials in which we confronted the experimental males with a conspecific male. Resource holders have higher costs of making an acceptance error, as it may result in injuries, death, complete loss of the carcass or, if the opponent makes the same mistake (he also might try to trick the resource holder to obtain access to the resource ), in wasting time and energy defending a resource without producing any offspring. Although the value of a female increases as well in the presence of a carcass, burying beetles have the chance to attract additional females via the emission of a long range sex pheromone . Hence, we expected a lower rate of acceptance error and therefore lower levels of SSB in resource holders than non-resource holders.
After 20 days, only a small proportion (mean ± s.e.: 12.22% ± 7.85) of males were showing SSB and female density had no influence (generalized linear model (GLM) with binomial errors, Wald-, p = 0.49; figure 1a). However, in the 60-day treatment, the occurrence of SSB was generally higher (mean ± s.e.: 34.82% ± 15.89; GLM, Wald-, p = 0.035), and there were large differences between social conditions (GLM, Wald- p < 0.001; figure 1b). Males in female-deprived situations engaged significantly more often in SSB than males kept with females (figure 1b). Males maintained with one or three females did not engage in SSB at all. Males kept in complete isolation showed the highest occurrence of SSB (80.0%; see the electronic supplementary material for a description of SSB).
Most males of the 20-day treatment (mean ± s.e.: 90.28% ± 2.45) copulated with females and there was no difference between social conditions (GLM, Wald-, p < 0.97; figure 1c). Also, a large proportion of males of the 60-day treatment (mean ± s.e.: 78.60% ± 9.65) copulated with females, but in this treatment group, the males' social environment affected the occurrence of matings (GLM, Wald-, p < 0.001; figure 1d). Males kept with three females copulated less often with a female than males kept in isolation or with a male (figure 1d).
The presence or absence of a carcass suitable for reproduction significantly affected the occurrence of SSB (χ2-test, , p < 0.001; figure 2). As in the previous experiment, males that were kept in isolation and never encountered a carcass showed high levels of SSB. However, if they were in the possession of a carcass prior to the mating trials, they showed low levels of SSB and in 77.80% of the cases the focal males attacked the encountered male.
Our results support the idea that the occurrence of homosexual copulatory behaviour is strongly context dependent and is influenced by the costs of making rejection and acceptance errors. As predicted by Reeve's acceptance threshold theory , if searching time for females, and thereby the costs of a rejection error, is experimentally increased, male burying beetles shifted their acceptance threshold and became more permissive. As a consequence, those males showed higher levels of SSB than males that experienced a very short searching time. In fact, males kept with several females started to be choosy, not only expressed a lack of SSB, but also did not mate with every female they encountered. The evolution of male choosiness is generally predicted if mating is costly [12,13]; however, we cannot rule out the possibility that the males kept at higher female density reduced their subsequent mating activity due to exhaustion or lower sperm count caused by multiple mating. We can, however, exclude that the lower mating rate is caused by a general higher beetle density, as males housed with one or several males copulated frequently and showed high levels of SSB.
The phenomenon that the incidence of SSB increases under the absence of mating opportunity is sometimes called the prison effect [2,14] and has been shown in a variety of animal species. Based on our results, we suggest that the prison effect is actually a consequence of a discrimination strategy and is caused by optimal adjustments of the acceptance threshold to the fitness consequences of rejecting females and accepting males. As our second experiment revealed, males in female-deprived situations are still flexible in their discrimination decisions. If we experimentally increased their perceived costs of accepting a male by providing them with a resource suitable for reproduction, males altered their discrimination decisions, were more restrictive again and hence expressed very low levels of SSB. This experiment also provides evidence that males under a female-deprived situation did not merely lose the ability to discriminate because they lacked the possibility to update their template of female cues. Furthermore, the second experiment demonstrates that sperm depletion or exhaustion cannot account for the difference in SSB levels, as males of both treatment groups were kept singly.
SSB has been shown in a wide range of insect species; however, its causes often remained unclear. Although studies revealed that the incidence of SSB increases or decreases in certain situations, SSB had never been investigated in the light of the acceptance threshold theory. The relative costs of making rejection and acceptance errors not only may explain the occurrence of SSB in burying beetles, but also may predict the level of SSB in many other species. We hope that our study will stimulate further research into the origin and maintenance of SSB.
Data from the Dryad Digital Repository: doi:10.5061/dryad.gh55k.
K.C.E. and S.S. designed the experiment; K.C.E. and L.M. collected the data; K.C.E. and S.S. analysed the data and co-wrote the manuscript. M.A. contributed to the writing of the manuscript.
The work was supported by a DFG grant (STE 1874/3-1) to S.S.
Conflict of interests
We have no competing interests.
We are very thankful to Eva Keppner for helping to rear and maintain the beetles, J. K. Müller for providing beetles to establish a colony and Per T. Smiseth and several anonymous reviewers for valuable comments on the manuscript.
- Received July 29, 2014.
- Accepted December 22, 2014.
- © 2015 The Author(s) Published by the Royal Society. All rights reserved.